Active Traders - Active Trader Magazine 2003 April Issue

APRI APRIL L 2003 2003 ISS ISSUE UE 2 3 4

Editor’s Note Contributors Chat Room Inside the Market: By Jeff Ponczak

6

Finding the best definition of best execution The SEC has many things to consider when deciding its best execution policy — particularly since it’s so difficult to define best execution.

9

What’s a commission really worth? Is the price a broker charges for commission reflective of the service they provide?

10 Quick Scalps

33 Eurex comes to the U.S. European exchange Eurex announced its plans to begin trading U.S.-based products.

33 Fast Fills 34 Futures and Options Trading Stra Strategies The RSI trendline method Looking for a simple way to enter trades? Using trendline analysis on the relative strength index (RSI) can provide solid trend-following signals. By Thom Hartle

37 Futures Trading System Lab Trend with pattern entry

12 Web Watch: Best of the Web By Active Trader Staff

14 New New Products

39 Futures Snapshot 40 Active Trader Intervie Interview Relatively speaking: John Bollinger

15 Trader’s Bookshelf Bookshe lf A review of Gerald M. Loeb’s The Battle for Investment Survival By David Bukey

A conversation with analyst, commentator and money manager John Bollinger about the current market and the opportunities and challenges facing traders in the years to come. By Mark Etzkorn

16 Tech Techno nolo logy gy for for Tr Traders Software Screening: Ensign Windows Reviewed by David Bukey

Trading Strat Strategy 19 How to trade broadening tops It’s not one of the best-known reversal patterns, but the broadeningtop has the ability to identify major turning points. By Thomas A. Bierovic

46 The Face of Trading: From pennies to patterns By Kira McCaffrey Brecht

47 Risk Control and Money Management Road to ruin Probabilities can tell you a lot about your trading strategy, but they have a limit. Case in point: The probabilities of losing your trading stake using risk-of-ruin calculations. By Active Trader Staff

23 Opening shots Two types of narrow-range bars set up short-term trades in stocks and futures. By Thom Hartle

26 Outside bars: From concept to strategy The outside bar is one of the simplest price patterns around. Here’s a look at the process of attempting to use it as the basis for a trading system. By Timothy Andrews

Trading Basics 49 Taking a different route New to stock trading? Learn how Electronic Communication Networks (ECNs) function and the benefits they provide traders. By Jeff Ponczak

52 Indicator Insight: TSI/Ergodic oscillator By Active Trader Staff

55 Technical Tool Insight: Gaps By Kira McCafffrey Brecht

29 Trading System Lab: Trend with pattern entry 58 Trading and Investing: Give me land, lots of land… Futures and Options Watch :By Jeff Ponzcak 31 The Merc speaks out After a mandatory quiet period, the Chicago Mercantile Exchange discusses becoming the first U.S. exchange to go public.

32 Euro(N)ext for CBOT The Chicago Board of Trade chooses EuronextLIFFE to be its new electronic-trading platform provider.

32 “Down” year for options still very busy Although yearly options volume was down from its 2001 totals, it was still quite a busy year for the options exchanges.

ACTIVE TRADER • April 2003 • www.activetradermag.com

What’s really important when assessing housing stocks? Hint: It’s not what you might think. By Peter Navarro

61 The Business of Trading: The proof is in the return A comparison of tax forms shows the benefits traders have over investors. By Robert A. Green, CPA

65 After Hours 66 Trade Diary 1

EDITOR’S

Note

THEORY of relativity

E

verything’s relative in the markets. Today might bring a two-week high, but that high might also be the latest in a series of five lower swing highs over the past few months. Tomorrow’s horrendous horrendous selloff may be its biggest losing day in six months, but does that overshadow the fact a market just concluded its best three-month upswing in the past two years? It’s all a matter of perspective. Of course, perspective can be difficult to come by when a generation of traders raised on bull are suddenly force-fed a steady diet of bear. Disappointment with recent market performance is just one part of the limiting “Up good, down bad” mentality that dominates the public investment psyche. This is not to argue that bear markets in stocks are simply interchangeable with –— or just as desirable as — bull markets, when it comes to trading. It’s just that there might be more opportunity to make lemonade with the current market lemon than we’ve led ourselves to believe. At least John Bollinger thinks so. The subject of this month’s Active Trader Interview, Bollinger has been around long enough to see the market hit more than a few highs and lows. And while the market has blindsided millions of investors over the past three years, Bollinger thinks things are pretty much unfolding according to historical script, and he sees plenty of juicy scenes for traders in the future. In “Relatively speaking: John Bollinger” (p. 60), the analyst, commentator and developer of the trading bands that bear his name explains why he thinks the market may be stuck in a range (a very, very big one, that is) for quite some time — and why this environment might offer great opportunities for traders with the ability to accept and trade the market for what it is. After all, today’s conditions are similar to those Bollinger first experienced when he started his career in the late 1970s. At the time, traders who recognized the market’s oscillating

nature and capitalized on its intermediate-term swings were able to continue taking profits out of a market investors had given up for dead years before. Relativity also rears its head in the Trading Strategies section. Relative low-volatility and high-volatility patterns provide the background for two articles this month. “Opening shots” (p. 36) analyzes different techniques for trading narrowrange bars (including inside bars), while “Outside bars: From concept to strategy” (p. 40) uses outside bars to illustrate the process of developing a trading strategy, starting with an observation and building up to strategy rules and test procedures. The Futures Strategy section shows how the relative strength index can be used as a trend tool (rather than a momentum oscillator) and the basis for a T-bond futures system. No solution to any trading problem is perfect. Each involves trade-offs. Making a choice about how to use a particular strategy is a matter of determining if the rewards are worthwhile relative to the risks. Finally, if you haven’t been to our Web site in a while, you should take a look. In addition to our expanding menu of market commentary, we’ve opened message boards where traders can exchange ideas about everything from strategies and software to the economy and trader tax issues, a trading Web site directory, a trading dictionary and resource center, and — most importantly — an online store where you can purchase articles from past issues of Active Trader and download them immediately to your computer. See what’s new at www.activetradermag.com. www.activetradermag.com.

There might be more opportunity to make lemonade with the current market lemon than we’ve led ourselves to believe.

2

Mark Etzkorn, Editor-in-chief

www.activetradermag.com • April 2003 • ACTIVE TRADER

THIS MONTH’S

Contributors

Thom Hartle is a private trader and president of Market Market An A nalytics Inc. (www.thomhartle.com). (www.thomhartle.com). In a career spanning more than 20 years, Hartle has been a commodity account executive for Merrill Lynch, vice president of financial futures for Drexel Burnham Lambert, trader for the Federal Home Loan Bank of Seattle and editor for nine years of Technical Analysis of Stocks & Commodities magazine. q

Peter Navarro (www.p w.peternavarro.com) holds a Ph.D. in economics f rom rom Harvard University. He is an associate professor of economics at the Graduate School of Management, University of California-Irvine, and the author of If it’s Raining in Brazil, Buy Starbucks: The Investor’s Guide to News and Other Market-Moving Events (McGraw-Hill, 2001). q

Thomas A. Bierovic (t bi [email protected]), president president of the Synergy Synergy Trading Group, has been successfully trading his own account since 1971. He has presented technical analysis and trading seminars at conf erences in the United States and more than 40 other countries. He is the author of Playing for Keeps in Stocks & Futures: Three Top Trading Strategies That Consistently Beat the Markets. His daily stock recommendations recommendations (and a free copy of his new e-book, Synergetic Trading) are available at www. synergy rgytrader.com. q

q Kira McCaffrey Brecht , ,

a Chicago-based financial financial writer and technical technical analyst, has been writing about the markets for 12 years. Posts during her career include Chicago bureau chief at Futures World News , technical analyst at Bridge News and market analyst at MMS International. Tim Andrews ([email protected]) is a Chicago-based trader, analyst and system designer active in both the stock and futures markets. q

Robert A. Green ([email protected]) .com) is a CPA and his firm, firm, GreenTraderTax.com, consults traders on tax solutions, reviews or prepares their tax returns, and sets up business entities and retirement plans. GreenTraderTax.com GreenTraderTax.com also specializes specializes in hedge-fund creation and management, and offers trader tax guides and trade-accounting software. For more information, visit www.greentradertax.com or call (212) 658-9502. q

Dion Kurczek ([email protected]) is a private trader, software engineer and trading system researcher. In 2000 he founded Wealth-Lab Inc. and launched an interactive trading system development developmen t laboratory on the World Wo rld Wide Web (www.wealth-lab.com). His firm produces trading system development and backtesting software for traders. q

Volker Knapp has been a trading system researcher for more than 15 years. He is president of the VTAD (the German branch of the International Federation of Technical Analysts) and co-founder of Wealth-Lab Inc. He was also a professional hockey player and coach. q


3

CHAT

Room

Breaking news

I

’ve been doing my trading research for quite a while now and I’m getting close to going online with real money. (With little or no education or skills, I had tried day trading in 2000 during a three-week vacation from work.) My question begins with a recent observation: On Dec. 24, 2002, AGN opened at 52.64. At 10:01 a.m., BusinessWire announced AGN received an FDAapproval (stock rallies to 53.90). At 10:16, www.briefing.com posted the story (stock rallies to 56.00). At 11:18, Reuters posted the story (stock rallies to 57.15). What newswire service is going to offer me the quickest news directly to my monitor (not e-mail or fax, and not from a list of stocks that I pre-select)? From the above example, BusinessWire would be the one, but I don’t believe they offer individuals the service that I’m looking for. If I had www.briefing.com or Reuters I would not have had the opportunity to make money on this news. I would appreciate any help that you could give in this matter. —Patrick Hey, how did that three-week day-trading stint in 2000 work out? Never mind. Don’t ask, don’t tell — that’s our policy. And now, on to the news: Having spoken to several traders (and from per sonal experience), the bottom line is no single news service will always be first in line. One service will be first one day, a different one will be quickest the next day. Some of the traders we spoke to feel some services are quick in cer tain areas — e.g., economic releases, earnings numbers, and so on. One thing you can do is ask a particular source or site if they are buying their news from someone else; if you really need space, there’s no point in using a service that’s redistributing news. Some trading or analysis platforms will include several news services, or the option to subscribe to more than one. Head-to-head comparison is really the only way to find one that gives you the most “scoop” for your buck. One piece of stand-alone news-tracking software, NewsWare (www.news ware.com), combines multiple news and analysis services and allows you to set up various filters and alerts.

A commitment to trading

I

actively trade both the S&P E-Mini and Nasdaq E-Mini. I was reading your interview with Larry Williams in your December 2002 issue and he mentions two indicators, the Commitment of Traders (COT) and the Advisory Sentiment Index. How I can get these indicators? Any help you can provide would be greatly appreciated. — Jan A. Verleur Besides the CFTC site (www.cftc.gov), Larry Williams says Genesis data (www.gfds.com) has this information. CSI Data (www.csidata.com) also provides COT data. A quick Web search pro duced a few other possibilities for COT analy sis: Trading Systems Analysis Group (www.tsa group.com), which offers COT data and analysis tools, and MarketPit (www.marketp it.com ) , which has COT charts. The first thing you may want to do, though, is ask your broker (and software vendor, if they aren’t the same) if they have that information.

Six of one, 2000i of the other

I

am an Active Trader subscriber and I use TradeStation 6 (TS6). It seems that most of the code on your Web site is for TS 2000i. Why? I am a novice when it comes to EasyLanguage and one of the most attractive parts of your magazine is the fact that the ideas are coded for

4


New York, New York easy use — that is, unless the code won't “verify” because it’s written for an outdated (at least in my humble opinion) system. Please add TS6 code! — Scott Miller We began publishing code for indicators and strategies before TradeStation updated to TS6 —hence the 2000i code. However, more recent additions to the code page have been written on the TS6 platform. Yes, some older code will take some tweaking to make it usable in new program iterations, but that’s unavoidable. TradeStation has a new version of its software in the wings, and it may render some TS6 code “unverifiable,” too!

Connecting the connections

I

read in a previous Active Trader article that there is a router on the market that can dynamically switch Internet connections if one (such as cable or DSL) should fail. I cannot remember what month the article was in, but I believe you would just plug in a Cable modem, DSL line, and even a 56k backup line. The router would use the primary line unless it was not functional, and in that event it would shift to the next line. Can you help? — Bob Byrne

The product you’re referring to is the Nexland Pro800 Turbo router. Our article on the product, written by David Bukey, appeared in the January 2003 issue. As you mentioned, it allows you to plug your broadband and dial-up connec tions into one central port, and it switches con nections if one fails.


I

have been a proprietary trader trading New York Stock Exchange stocks for the last three months. I trade with a firm in Canada. Because I was the first group of trainees with this company, they didn’t really have much of a strategy for us to follow. I was just wondering if there were any tips (i.e., dealing with a specialist, technical analysis, differences between the NYSE and the Nasdaq) for day trading the NYSE. Most of the material out there seems to be Nasdaq-oriented. — Dan Roch Strategically, there’s no real difference. Technical analysis is technical analysis, whether it’s applied to NYSE or Nasdaq stocks on an intraday basis, or the crude oil market on a long-term basis. Tactically, the primary difference between trading NYSE and Nasdaq stocks are their respective specialist and market-maker systems and their slightly different uptick rules. Because order flow always goes through a designated specialist on the NYSE, and because specialists are required to fill customer orders before their own (if those orders are at the same price), some traders look for situations in which the specialist may be alone on the bid or offer, and then trade “with” the specialist. Active Trader has previously published articles dealing specifically with trading NYSE stocks. Check out “Playing the special ist’s game” in the May 2000 issue; “Bullish on the Big Board” in the Jan. 2002 issue; and “NYSE short selling: Bear raid warning” in the July 2002 issue. All three articles are written by Christopher A. Farrell. These articles, as well as all other past articles, can now be purchased from the Active Trader Web site (www.active tradermag.com). Click on the “down load articles” link at the top of the page.

Questions about an article or trading issue? Send them to [email protected]. Active Tr ad er reserves the right to edit letters for clarity and length.

5

INSIDE THE

Market

BY JEFF PONCZAK

Good, better, best?

Finding the best definition of best execution

I

t wasn’t all that long ago that best execution in the stock market was a concept more than an actual rule. Sure, there was legislation on the books that required it, but nobody (from the brokers to the regulators) paid much attention to it. The trading public was essentially at the mercy of brokers, who may or may not have filled their customers’ orders in a timely manner, and may or may not have done it at the best price available. However, advanced technologies have changed all that. ECNs allow traders to execute trades in fractions of a second, and there are myriad places traders can find the best bid/offer for a particular stock. As a result, there is more emphasis on best execution than ever before. The Securities and Exchange Commission (SEC) has mandated all execution venues provide data on how fast they execute an order and at what price. This has allowed for a direct comparison between execution firms. Still, best execution means different things to different traders — scalpers want an order filled at the price showing on their screen; institutional investors with a one-million share order are more concerned about their order getting filled quickly than they are about price. In the third of a three-part series examining challenges facing the SEC, we’ll look at the issues surrounding best execution and the difficulty of keeping all market participants happy.

What does it mean? With the exception of trades executed through ECNs, all orders are first sent to a broker, who then sends the order to be executed. Although the brok er/dealer does not necessarily oversee the execu6

tion of the trade, SEC rules say it is the broker/dealer who is responsible for ensuring the customer receives best execution. In theory, this puts the onus on the broker/dealer to find an execution venue that consistently provides the best execution. The problem, though, is finding a comprehensive, objective definition of best execution. “Best execution is a combination of things including speed, effective spread, price improvement, liquidity and cost of execution,” says David Herron, CEO of the Chicago Stock Exchange. “As each broker/dealer builds a business model and chooses which markets to go to, they are able to compete for customer business. The customer will vote with their commission dollars. “The exchanges are also required to provide for best execution in making sure the specialist and floor brokers are

looking out for the best interests of their customers,” he says. “If they don’t, the system breaks down.” When execution data is reported by market participants, all price data is compared to the national best bid/offer (NBBO). The NBBO has long been accepted as the “best” price available in the marketplace. However, even that is not as cut and dried as it seems. “The simplicity of the NBBO is not as simple as it seems,” says Matt DeSalvo, managing director of Morgan Stanley’s equity division. “When we have an unequal playing field with regard to access standards and access speeds, you cannot determine price as the standard for best execution. If we don’t address equal access standards, we cannot get good determination of price as best execution.” From an access speed standpoint, the inequality is apparent when both an


ECN and a floor-based exchange are showing the best price. An order sent to an ECN can be executed immediately; an order sent to the floor may be held by the specialist, and price may change before the order is filled. On the other hand, the price shown on ECNs does not take into account access fees. In the case of floor-based exchanges, no fees apply. “At a minimum, a price in one market should be equal to another market,” DeSalvo says. “When I access a $10 bid, it should be 10 to me, not 10 plus an access fee. In addition, there should be a minimum level of access standard s between markets. At that point, you at least have a skeleton to work with.”

The missing link? One plan that has been suggested more than once is connecting all market participants via high-speed linkages and giving all execution venues auto-execution capabilities. Not surprisingly, the floor- based exchanges are reluctant to support any plan that would diametrically change a business model that has been successful for decades. “There are a lot of ways to access the NYSE,” says Chris Quick of Fleet Trading, an NYSE specialist. “Linkage is not a problem.” One of those ways is the Intermarket Trading System (ITS), an electronic platform that gives other exchanges access to NYSE quotes. For years, ECNs had no access to the ITS and therefore no ability to post NYSE quotes. Some ECNs have since joined the ITS through Nasdaq, albeit reluctantly. The ITS allows a three-second delay from the time an order is received to the time it must be acted on. For an ECN, three seconds is a lifetime, and even Quick admits the ITS has its flaws. “In 1975, it probably accomplished the obligation of best execution,” he says. “I’m not sure today that it’s the vehicle to continue on to the future.” Additionally, ITS rules prohibit a market from trading through the quotes of any other. So, if a specialist is sitting on the best bid or offer, it may be a while bef ore anything can be done in that stock.

“The simplicity of the NBBO is not as simple as it seems. When we have an unequal playing field with regards to access standards and access speeds, you cannot determine price as the standard for best execution. If we don’t address equal access standards, we cannot get good determination of price as best execution.” —Matt DeSalvo, Morgan Stanley “[No trade-throughs] would make sense if all quotes were electronically, instantaneously reachable, but they are not,” says Thomas Peterff y, y, chairman and CEO of direct-access firm Interactive Brokers. “These rules provide traditional exchanges somewhere between 30 and 90 seconds to decide if they want to trade with an order or not. “The SEC should examine whether exchanges and ECNs with electronically accessible limit order books could be cut f ree ree from ITS participant quotes,” he says. “If the answer is yes, these electronic venues could form their own ITS rules. There are a sufficient number of them, so you could certainly come up with a fair price.” The members of the ITS are considering changing the rules to allow tradethroughs. DeSalvo thinks any discussion of link-


age is moot unless equal access and fee standards are first addressed. Obviously, those from the electronic-trading arena disagree. “How do you have equal access when you have floor-based models and electronic models?” says Janet Angstadt, associate general counsel of Archipelago. Archipelago. “When you’re an electronic market and you’re trying to access a manual market, what does that mean for you? Linkage is a good next step, and then we can decide what access is for each individual market.” Another idea floating around is that of letting the customer decide what best execution is. If the customer desires speed, then his broker will be judged on how fast he fills an order. If he desires price, that will be the benchmark. However, Peterffy thinks drawing distinctions between the two is silly. “The only reason people want speed is because they worry price will move away from them,” he says. “So it’s only price we are into. Best execution certainly means the best aggregate price for the order.”

Fee, fi, fo, fum Morgan Stanley’s DeSalvo is by no means alone in his displeasure of ECN fees. For firms such as Morgan Stanley and others that operate as both traders and market makers, the system has its inequities. If Archipelago routes an order to Morgan Stanley for execution, there is no access fee involved (in fact, it is forbidden by SEC law). However, if Morgan Stanley goes to the Archipelago book to execute a trade, that trade comes with an access fee. “We have the same business model,” DeSalvo says. “But we can’t charge an access fee.” In defending the ECNs, Angstadt says every participant should be able to charge an access fee. And, Nasdaq’s SuperMontage trading platform gives users a choice of viewing prices with or without access fees being considered. Moreover, she says, other market participants charge transaction fees. “Why are we making a distinction between transaction fees and access fees?” Angstadt asks. “If, as a service, I 7

INSIDE THE

Market

am providing liquidity and I am able to charge a transaction fee, why should I have to give that product away?” Complicating matters even further is the fact that ECNs and the Nasdaq provide rebates to traders who add liquidity by placing limit orders. Some see this practice as no different from payment for order flow, a controversial controver sial activity involving execution venues paying broker/dealers ker/dealers for the right to execute their order flow. Critics of payment for order flow claim broker/dealers broker/dealers send their orders to the participant participant who pays them the most, not necessarily the one who will provide the best execution. Likewise, critics of market rebates believe it will give certain participants a disproportionate amount of orders, even if they are not the best venue for executing them. However, Angstadt says there is a clear distinction between rebates and payment for order flow. “When people pay for order flow, it’s because they believe they can make a lot of money on that order flow,” she says. “On a rebate, we’re glad to give some back because we have saved money by being able to cross order internally. We want to pass those savings along.” The system makes perfect sense to Peterffy. “Liquidity takers must pay liquidity providers,” he says. “That’s either done via paying the spread or by rebates for limit orders. If you look at what is happening at the ECNs, spreads are extremely tight, so that’s a good deal for the customer. “When we route a limit order to an ECN, we get two-tenths of a penny. We also route to ECNs because they provide

the quickest way to cancel an orde ord er, which is part of best execution.” Besides, not everybody is convinced that payment for order flow is necessarily a bad thing. “There is no reason why payment for order flow itself has to come at the expense of best execution,” says Bernie Madoff, principal of Madoff Investment Securities, a Nasdaq market maker. “In the equities market, I could prove that the venues paying for order flow are getting better execution than those that aren’t. The problem is that most competing markets use payment for order flow as a way of disparaging other markets.” Madoff says the mandatory publicizing of execution data has taken the negative perception away from payment for order flow. “Payment for order flow is not a problem because it’s very easy for a firm to look at the execution stats and clearly make a determination as to where to route their order,” he says. “It’s important we don’t characterize payment for order flow as an evil child, because that’s not the case.” Still, some market participants are displeased with rebates of any kind. They pay millions of dollars in access fees annually, and they believe if rebates are eliminated either by mutual agreement or legislation, the need for access fees will be eliminated and the entire system will run much smoother. Angstadt, though, says access fees are not designed to offset rebates. “Routing out and having people access us costs money,” she says. “Plus, consider the ITS. It’s free, but it’s slow. You should be able to charge for the value you provide.”

TRADING ROOM ANGEL I'm installing your new operating system.

Darn, I hate when that happens.

Allow me.

But the old one's working just fine. 8

It's not safe around here. Let's go over to “On the Job.”

Additionally, Instinet CEO Ed Nicoll believes there is a need to create an incentive for placing limit orders. “When somebody places a limit order, they are putting themselves at a disadvantage to someone who doesn’t place a limit order,” Nicoll says. Joel Hasbrouck, professor of finance at New York University, believes the rebates provided by ECNs have helped narrow Nasdaq spreads. “We need to make sure that we not only have incentives for limit orders, we don’t present disincentives –— i.e., charges for changing and canceling limit orders,” he says. “In listed markets, specialists having the ability to delay posting of limit orders and delay executing of market orders are disincentives. The reasons spreads have not narrowed are a direct result of that.”

What’s next? Although a central limit order book has been proposed more than once over the past few years, it enjoys only moderate support today. The SEC has long favored favore d innovation and technology, as is apparen appare nt in allowing ECNs to charge access fees when other market participants participants don’t. Many think the next logical step is a complete linkage, similar to the one recently enacted among options exchanges. However, this would certainly meet with great opposition, and is years off even with complete participation. The SEC is often hesitant to make radical changes to any aspect of the market unless situations get completely out of hand. While this patience often makes it the brunt of criticism, it also suggests the SEC will continue to have a wait-and-see attitude toward best execution.Ý Must be that new System Restore feature I heard about.

Does she do servers?

Race you to page 98!

©2003, Active Trader Magazine


Is the price right?

What’s a commission really worth?

W

hile improvements to a trading platform are sometimes announced with nothing more than a press release, it’s not unusual for firms to roll out reduced commissions with a full-blown marketing blitz, including print and media advertising, and new signup specials. This no doubt due in part to the fact that while more experienced traders also consider a brok broker’s trading platform, execution quality, customer support, etc., commissions are often the only point of reference for traders just entering the business. “The newer entries into the industry are price shoppers,” says MB Trading CEO Ross Ditlove. “As they mature into the industry, they begin to educate themselves as to the benefits beyond price, and often times the lowest price provider is not going to give you the highest level of service.” Indeed, commissions are just one part of the overall package a firm brings to its customers. And finding the right balance between price and service isn’t always the easiest thing to do. “It’s important to have low costs,” says Chris Doubek, president of Terra Nova Trading “But there is not enough focus being placed with the customer on being profitable because firms think profitability profitability is something subjective that they can’t address. And that is absolutely not true. Being concerned about your risk is much more important than being concerned about whether you have a $7, $8, $10 or $20 commission.” It should be noted that non-directaccess firms, where less active traders and investors typically trade, routinely have cheaper commissions because they often have lower-end platforms and slower execution speeds. Most trades at Terra Nova cost $15, although there are incentives available to reduce the costs. Doubek isn’t convinced a firm that con-

sistently offers cheap commissions can adequately provide the services necessary to ensure an active trader’s success. “You’re going to give up something when your commissions are as low as you would like them to be,” he says. However, However, Andy Gill, vice president of marketing for CyberTrader, believes it is possible to offer low commissions and still provide a complete range of services. CyberTrader has commissions as low as $9.95, with $12.95 as a top tier. “We look at commissions as one of the levers we use to provide the best value to our clients,” he says. “And we look at value as the combination of low commissions with high-quality executions, knowledgeable and available client service, and services such as education to help the trader. “I think we compare favorably [to firms that charge more money],” Gill says. “When we launched our new streaming Web platform, we made sure we were committed to providing a value to that client, recognizing that their needs were going to be different than the needs of somebody who has traded for 20 years at a very high volume. Consequently, our clients services are trained to handle those sorts of questions, and we want to help those people be successful in their trading style, whether it’s 10 or 100 trades a month or 100 trades a day.” However, Gill knows that traders who choose CyberTrader won’t stay just because of low commissions. “We have to provide the best value,” he says. “You have to take things that aren’t a direct cost of the trade — client services, technical analysis, trader education — and understand what your clients think the value of those are in determining that price.” According to Steve Sanders, executive vice president of Interactive Brok Bro kers, “The key to whatever you’re going to charge is your infrastructure.” infrastructure.” “If you’re a high-cost broker with old


mainframe systems and you use other people to do things, obviously you’re not going to be able to charge a very low price,” he says. “At Interactive, we have our own clearing system and our own execution system. We have never gone outside for a consultant. We link into the exchanges directly and we view that as a competitive advantage. We can offer the lowest commissions, and anybody else who is trying to offer the same low commission without a system like ours won’t be able to sustain that in the long run.” Interactive has one of the lowest commissions plans available, with certain trades costing as little as a penny a share. Sanders says Interactive often turns 9

INSIDE THE

Market

away less-experienced traders, suggesting they trade with a broker that provides more services. “We aim at more of the professional type,” he says. “If someone wants help and advice, we’re not going to do that. We offer the lowest commissions and professional customer service. We’ll help you with problems or issues, but we won’t provide the hand-holding that other brokers will provide.”

Added importance Cyber (Charles Schwab) and Interactive Brokers (Timber Hill) are both owned by larger organizations, and Terra Nova is

the largest “independent” direct-access broker. As a result, their financial situations and infrastructures are more stable than those of many other firms. For smaller firms such as MB Trading, commissions take on even greater significance. “We are constantly balancing three things with the company — price, service and quality,” says Ditlove. “We attempt to pick two at all times, because you can’t pick three. Those priorities sometimes change. It’s always service first, and when quality is below price (i.e., we don’t off er all features as our competitors), we bump up the quality. Once we catch up or

exceed the industry, we can emphasize keeping price in line. But our goal is not necessarily to be the cheapest.” MB offers traders the option of paying per trade or per share. As a result, some trades are less than $10, and Ditlove says it’s difficult to make a profit on an individual trade when commissions are that low. “As you go through various economic growth cycles, you have supernatural growth, normal growth and then you plateau out and profits go toward zero,” he says. “Once you start getting to the point where profits are zero, you have to continuously watch the bottom line as far as expenses are concerned. And this

• • • • • • • • QUICK SCALPS YOU SCRATCH MY BACK… Brokerage firm Robertson Stephens was censured and fined $28 million in early January after the NASD found it received inflated commissions from clients in exchange for IPO shares in 1999 and 2000. Half the fine goes to the NASD, the other $14 million to the SEC. In some instances, customers were paying more than $1 a share for orders — well above the typical charge of six cents. Robertson Stephens was the lead manager of more than 75 IPOs in 1999 and 2000. q

IT’S EVEN MORE SUPER In early February, the SEC approved a change that allows non-registered Nasdaq participants to post limit orders in SuperMontage. The SEC approval is for a 90-day pilot program. Under the rule change, broker/dealers and other order-entry firms that do not have direct access to SuperMontage will be able to enter limit orders into the system. Plus, market makers will be allowed to display orders in SuperMontage for stocks in which they do not make markets. The Nasdaq made the change in an effort to increase SuperMontage liquidity. q

TAKE YOUR TIME With the index still mired in a slump, and prices of many of its stocks trading near or below a dollar, the Nasdaq has decided to continue a pilot program that gives issuers additional time to avoid delisting for failing to meet minimum bid requirements. The rule has been extended until Jan. 31, 2004. According to the rules of the program, all Nasdaq issues that fail to meet certain listing requirements would have 180 calendar days to demonstrate compliance with minimum bid requirements and another 180 days if they comply with other listing criteria. For SmallCap stocks failing to meet bid requirements, the grace period is also 180 days. However, it increases to 540 days (about 18 months) for stocks that meet other listing criteria. q

LEGAL MATTERS U.S. lawyers enjoyed a victory of sorts in late January when the SEC decided to extend its comment period on the “noisy withdrawal” provisions of the Sarbanes-Oxley Act. The rule would have, in certain circumstances, required lawyers representing publicly traded companies to blow the whistle on their clients if they found evidence of wrongdoing. Attorneys claimed this would be a violation of attorney-client privilege. The SEC did pass provisions that require lawyers to report violations to upper management or, if that fails, to the board of directors. And, companies will be allowed to form “qualified legal compliance committees” to recommend appropriate response to a material violation. The rules will not apply to lawyers of foreign-based, U.S.-listed companies, as long as those lawyers are not advising their clients on U.S. law. q

DON’T BLAME ME…BLAME MY BOSS The NASD’s ongoing investigation of Wall Street analysts could be expanded to include the analysts supervisors, published reports said. The NASD is considering a case against former Merrill Lynch Internet analyst Henry Blodgett, who became famous for his outlandishly bullish reports on stocks such as Qualcomm and Amazon.com. However, the NASD is also investigating the supervisory structure of Merrill Lynch to see if it did enough to prevent analysts from making unwarranted claims. q

10


is why you have seen such tremendous fall on the customers. consolidation within the industry. You “When the question is asked of the have to grow as far as trades or shares firm, “What are your commissions,” traded per day, because if you don’t, somewhere in that conversation should you’re going to get eaten alive. That’s be a question about how profitable the what you see happening.” person is on those trades,” he says. Ditlove admits MB Trading watches “Ultimately, if you’re not profitable, it the competition closely to ensure it is doesn’t matter what the commissions competitively priced, although he says are. You should be looking first at risk previous price changes have been a management. Then you look at making direct result of customer feedback. some money on the trade. Once you’re profitable and you don’t have to think Share the blame about each trade being profitable, then While Doubek thinks some firms need to you can start working on your expenses, be less concerned with commissions, he such as commissions.” also thinks some responsibility has to “If a group of customers continue to

put downward pressure on broker/dealers for lower prices, broker/dealers really have nowhere else to go but to continue to eliminate services and eliminate infrastructure,” Doubek says. “It’s a downward spiral. “I think it’s important for us as professionals in the industry to dictate some of the pricing and the focus,” he says. “I don’t think there’s a successful trader who will tell you that a $20 commission vs. a $10 commission on each trade made the difference in their success. If you have so little profit that you become unprofitable after a difference of $10, you’ve got some problems.” Ý

SPEND IT WISELY q It appears the SEC will have more money to play with in the next fiscal year. The federal budget, submitted in early February, calls for an SEC budget of around $650 million. That represents about a $100 million increase from the last fiscal year, which ends Sept. 30. The total SEC budget is less than Democrats wanted, but it’s an increase on the initial figure proposed by Republicans last summer.

NEW, POSSIBLY IMPROVED q The New York Stock Exchange recently introduced a revised composite index that will serve as the basis for an exchange-traded fund. The index will represent nearly every stock that trades on the NYSE. While the old NYSE Composite Index took into account every issue on the NYSE, the new index would eliminate closed-end funds, preferred stocks, derivatives and limited partnerships. All told, 2,077 stocks will comprise the new index.

HEY, DIDN’T I SEE YOU ON TV? Todd Eberhard, a stockbroker who was also a frequent guest on financial TV, was arrested in early February on charges he cheated his clients for his own personal profit. The 39-year-old Eberhard, a guest on CNN’s Moneyline and other shows, is charged with making excessive trades in his customer’s accounts (known as churning) for the purposes of creating excessive commissions. The actions caused one customer to lose more than $10 million. The charges allege Eberhard began the scheme in 1999. If convicted, he could face up to 25 years in jail. Eberhard was previously named in numerous complaints that caused him to pay more than $4 million in settlements. q

THAT IS SO(ES) ILLEGAL In concluding an investigation that began in 1993, the SEC in mid-January charged four individuals and day-trading firm Heartland Securities with fraud related to Nasdaq’s Small Order Execution System (SOES). Total penalties in the case total more than $70 million, with two individuals — Sheldon Maschler ($29.2 million) and Jeffrey Citron ($22.5 million) — paying two of the largest individual fines the SEC has ever handed down. The group is charged with using SOES to execute trades for their own accounts — a violation of NASD rules. The group created fictitious records and filed false reports with the SEC. q

THE FEELING IS MUTUAL q After months of discussion, the SEC in early February proposed new rules for mutual funds. The rules will mandate that internal controls need to be tightened and will consider creating a self-policing organization for the mutual fund industry. These rules follow on the heels of an SEC vote weeks earlier that forces fund managers to tell their investors how the managers voted on shareholder questions asked by the companies that comprise the fund. The proposed rules are subject to a 60-day public comment period, after which they are expected to be approved.


11

WEB

Watch Best of the Web BY ACTIVE TRADER STAFF

O

ver the past three years, Active Trader has written about dozens of trading Web sites. Some have been mainstays of the online trading community; others have been consigned to the dustbin of digital history since we wrote about them. Now that the field has been pared down a bit, we revisited some of the sites and looked at new ones to develop a list of top trading and investment sites that have stood the test of time, however modest that time period may be. We focused on sites offering useful free content to visitors, as well as pay sites providing a service or product not readily available anywhere else.

uct you sold at a loss less than 30 days ago), which can cause tax losses if not properly reported. It also has an education center that answers many tax questions. Gainskeeper’s subscriptions can cost as much as $149/year — even more if you have in excess of 1,000 transactions. However, a 30-day free trial is available, which should be adequate time to determine if the product is right for you.

ECN sites

Island.com and Archipe lago.com (www.island.com, www.archipelago.com): These two popular ECNs have plenty of helpful data on their Web sites, including their respective order books, which can help traders get an idea of the liquidity of a particular stock at Island or Archipelago. Additionally, Island.com has other data tools such as priceTrader tax and business site volume charts and time and sales for individual stocks, and topGainskeeper.com (www.gainskeeper.com): This site was 5, 10 and 20 lists that show Island’s most-traded stocks. While launched in the late 1990s as a way to help traders automate the the information provided is only for trades entered and executprocess of reporting trades for tax purposes. Gainskeeper has con- ed on Island.com (and not for all of the Nasdaq), it is fre e. tinuously made changes and additions to the site, and it remains Considering how popular Island is (on certain days, the ECN helpful for traders trying to make sense of their tax situation. does 20 percent of total Nasdaq volume), the price showing on Gainskeeper’s best feature is its ability to download individ- Island.com is often the best bid/offer on the Nasdaq. That ual trade data from a brokerage (more than 30 brokers have increases the quality of the information Island.com is providing. partnered with Gainskeeper). The data can be transferred Archipelago also provides lists and updates, but the best directly into an individual’s account; in some cases, it must be part of its Web page is its program allowing traders to make downloaded into an Excel file then transferred to the Web site. sense of the execution data provided by market centers. SEC Gainskeeper then compiles all the trades and prints them out rules mandate all execution venues (i.e., ECNs, market makers, on a Schedule D (the tax form required by the IRS). For an exchanges) must disclose where they are routing their orders active trader, this will save hours, if not days, of paperwork. to, at what price orders are getting executed, etc. Gainskeeper handles wash sales (the repurchase of a prodThis data is released in spreadsheet form, with dozens of columns and thousands of rows. It’s virtually impossible to make any sense of the FIGURE 1: GAINSKEEPER.COM raw data. Archipelago, though, has taken Active traders with hundreds of transactions to report may find Gainskeeper. com the raw data and presented it in a way to be a great timesaver. that everybody can understand. For example, you could compare execution time between Island and SuperSOES on all Oracle orders between 2,000-4,999 shares. Twenty-three different market participants can be compared on speed, price and price-to-speed. The three most recent months of data can be viewed.

Online system testing Wealth-Lab.com (www.wealth-lab.com): There are several software programs that let you design and test trading systems. Wealth-Lab.com provides free access to sophisticated testing tools through its Web site. Wealth-Lab.com offers access to its library of trading systems and allows users to create their own using the program’s “WealthScript” language (a more sophisticated version of Wealth-Lab is available as stand-alone software). Users 12


FIGURE 2: ARCHIPELAGO.COM The Archipelago Web site can help make sense of the execution data provided by market centers.

can also create detailed charts and examine volumes of trade and systems statistics and performance characteristics. Wealth-Lab.com can back-test trading systems on individual equities or stock portfolios. There are dozens of systems to experiment with, and a forum through which to share thoughts with other traders and systems designers.

Market research MarketHistory.com (www.markethistory.com): If you’re looking for probabilities and statistics about different price moves and market relationships, MarketHistory.com is one of your best bets on the Web. With research and number crunching powered by Market Information Machine (previously software sold by the firm Logical Information Machines), MarketHistory foregoes vague ref erences to what usually happens when indicator x crosses threshold y , and instead answers questions such as, “How has the Dow performed in the past during the month of December when it has been down more than 10 percent going into the month?” and “The Dow on Tuesday declined 172.98 points, or -2 percent. What’s happened in the past when the Dow has made a one-day decline of more than 2 percent two days before Thanksgiving?” The answers blend written summary with tables and charts that break down the results, usually over several years, so you can get a handle on the probabilities. Covering the stock, futures and Forex markets, MarketHistory puts the numbers at your fingertips, allowing you to factor historical market tendencies into your trading plan. Subscriptions range from $24.95 to $200 per month.

implied volatility, delta and theoretical option values, historical volatility, current 30-day implied volatility index for calls and puts, and price and volatility charts. Also free is the basic options calculator, which allows you to calculate theoretical option prices and Greeks by entering the parameters (option style, price of the underlying instrument, strike price, expiration date, implied volatility, interest rate and dividends) of your choice. Various subscription packages allow you to add option and volatility scanners, rankers, and advanced worksheets and calculators.

Economics

Online research/browsing tool

The Dismal Scientist (www.economy.com/dismal): Even if it was a shoddy Web site, The Dismal Scientist would be worthy of praise for its name and URL. Fortunately, Dismal, an affiliate of Economy.com, is a great online resource for all things economic. If you weren’t paying attention in Econ 101 (and re ally, who could blame you?), Dismal has educational ref erences and numbers galore. Commentary, forecasts and analysis of various economic data (domestic and overseas) abound, and the site has neat tools, including a CPI calculator that gives you an idea of how money changes value over time as inflation ebbs and flows. There’s also a section devoted to the Federal Reserve for those who like to read the minutes of FOMC meetings. Subscriptions start at $19.91 per month ($189.95 annually) and the site offers a free two-week trial.

SpeedResearch.com (www.speedresearch.com): If you’re an information junkie, this is the site for you. SpeedResearch is a downloadable browser with links to dozens of market-related Web sites. Switching from, say, BigCharts to Yahoo Finance to TheStreet.com can be done with a single click of the mouse. The best part is that if you’re looking for information on a specific stock, you need only enter the stock symbol at the top of the browser and all subsequent sites you visit will have the data for that stock. For example, if you’re analyzing a chart of Dell on Prophet.net and you want to see how Dell is trading in the Island book, a click of the Island link will send you directly to the Island order book for Dell. SpeedResearch also links to various lists — i.e., 10 most active, biggest percentage gainers/losers, etc. You can enter and track entire portfolios, and any stock or stocks you find on a Web site can be instantly transferred into a portfolio (e.g., if you find a list of the 10 stocks that did the most volume in the first hour of trading, with one click you can move them all into a portfolio and track them the remainder of the day). There is a one-time payment of $37.99 for SpeedResearch. All upgrades are free, and the site comes with a 30-day free trial. Ý

Options/volatility analysis Ivolatility.com(www.ivolatility.com): When we first looked at Ivolatility.com in September 2000, the site was just starting out. More than two years later it has mushroomed into a multifaceted research center for traders who want the inside scoop on option value and volatility. Better yet, it still has a lot of free information and tools, including general stock and option price, volume and open interest data, option chains with ACTIVE TRADER • April 2003 • www.activetradermag.com

Next month: more Best of the Web. 13

NEW

Products

ESignal recently released eSignal 7.2, the latest version of its real-time quote service and charting software application. Among its features, eSignal 7.2 includes a Learning/ Replay Mode, with which users can replay data after hours into a chart to see how indicators and strategies worked out, and Market Depth, which provides detailed exchange information including Nasdaq SuperMontage and Island’s ECN Book data. Version 7.2 also includes Formula Wizard, through which users can back test, build studies, apply alerts and initiate trades automatically; single-stock f utures data from OneChicago; and the ability to mix real-time and delayed services from the markets. Anew chat function is also available through the recently unveiled eSignal Central Web site (www.esignalcentral.com), which supports file sharing and instant messaging, and features a learning center and product training. Monthly subscriptions of eSignal 7.2 start at $49 for active traders and $175 for professionals. For information, visit www.esignal.com/f eatures. q

FutureSource has released ProNET 4.2, an upgrade to its real-time charts, quotes and news application. New features include chartable expressions, which allow users to perform any mathematical operation on one or more contracts; redesigned price pages; and the ability to e-mail charts and weather maps. Also, six additional indicators (Moving Average Convergence Divergence, Spread Absolute, Spread Relative, Stochastic Fast, Stochastic Slow, and Volume and Open Interest) are available through ProNet Express , and users will find support for Delta, Gamma, Theta, Vega and Implied Volatility. There is also a new user authentication service, among other enhancements. For more information on the updated ProNET 4.2 or for a 14-day free trial visit http://pronet.futuresource.com. q

The Options Industry Council has added Options University (OU) to its Web site, www.888options.com. OU offers a variety of interactive options classes based on experience level — beginner, intermediate and advanced — including Buying Calls, Buying Puts, Introduction to Spreading, Options in a Bearish Market and Options in a Bullish Market. Once registered, students have free access to: phone and e-mail support; options strategy overviews and articles; options calculators and glossary; and Online Options Investigator software . Quizzes are also part of the offering. For more information or to register, visit the Web site. q

The Trader Bulletin, a weekly magazine focused on professional equity trading, has launched at www.trader bulletin.com. Membership at the Trader Bulletin is free and q

14

includes a weekly e-mail bulletin with the latest trading industry headlines. New editions are released every Thursday morning at the Web site after the market opens. Recent topics have included trading innovations on the NYSE, the evolution of ECNs, new buy and sell-side technology, day-trading scandals and market data controversies.

GreenTrader Tax.com, a tax site for traders, has a brand-new look. The new Web site is easier to navigate and has an improved Education Center. There is additional content on hedge funds and trader retirement plans, plus updated information on key issues such as trader tax status and mark-tomarket accounting. Available on the site are the 2003 versions of the Trader Tax Guides, written by GTT CEO Robert A. Green. The new guides are completely updated to reflect 2002 tax law changes, and can be downloaded directly from the site. For more information, go to www.greentradertax.com. q

Global Forex Trading DealBookFX customers now have access to MMS International’s Market Pulse, Central Banks and FX Fundamentals sections. As a result of a partnership between the two companies, users will have access to market news and analysis, intraday commentary, interday highs and lows in the USD pairs and non-dollar crosses, central bank policy outlook and intervention tracker, as well as an FX forecast matrix, daily FX commentary and special FX reports. In addition, coverage of MMS Market Movers, Foreign Exchange, Viewpoint, Major Markets, Emerging Markets and financial calendars are also included. For more information visit www.gftforex.com. q


TRADER’S

Bookshelf

The Battle for Investment Survival By Gerald M. Loeb (Foreword by John Rothchild) John Wiley & Sons, 1996 (originally published in 1935) Hardcover, 320 pages $29.95

REVIEWED BY DAVID BUKEY

G

erald M. Loeb’s The Battle for Investment Survival is a great book for active traders because it focuses on practical, short-term trading ideas rather than long-term investing concepts. Loeb began his career as a stockbroker and financial journalist in 1921 and published this book in 1935 — about the same time Benjamin Graham advocated long-term investing in his influential book, Security Analysis . John Rothchild notes in his f oreword to Loeb’s book that many investors interpreted Graham’s ideas to mean buy and hold indefinitely — regardless of what the market does. Loeb, however, sees long-term investing based on hope rather than judgment, which leads to complacency and, ultimately, serious losses. By contrast, the author argues, getting in and out of the market in short bursts is a safer method. He believes short-term trading provides flexibility long-term investing doesn’t: Having your assets fully invested can leave you vulnerable to unexpected risk, while leaving your portfolio in cash limits risk and allows you to take advantage of future trading opportunities. Loeb does not focus on specific trading rules, which will disappoint some readers. However, the real value of the book lies in its general principles for executing and managing trades. Not surprisingly, Loeb’s most relevant piece of advice is to limit losses: Sell positions that have declined 10 percent and sell the bottom 10 percent of your portfolio each year. Cutting losses allows you to preserve capital and view the market with fresh eyes. Instead of waiting for a weak stock to rebound, you can reassess your opportunities after taking the loss. This approach helps take emotion out of trading and forces you to stay focused on trades with the most profit potential at any given moment. Loeb describes his pyramid trading technique while explaining the “Ever-Liquid Account.” First, identify and enter a promising trade. Cut your loss if the trade goes against you and buy more shares as it becomes profitable. Loeb sees this strategy as more successful than dollar-cost averaging or diversification. While Loeb admits diversification is the best


approach for inexperienced traders, he believes in concentrating one’s holdings: “The greatest safety lies in putting all your eggs in one basket and watching the basket.” Context is another key issue for Loeb. Traders must have a “[T]here is no cheaper insurance than accepting a loss quickly. That is the tactic of retreat rather than capitulation. I think it would be very difficult for an investor losing, say, five to 10 percent each time on a succession of ventures, to continue to lose time and time again without checking his errors or stopping altogether. On the other hand, a buyer who holds regardless of unfavorable news or action can become involuntarily locked in his ‘investment’ for years… It is important to regard the situation with an open mind, unbiased by a bad stale position, and it is important to be able to act each time convictions are very strong.”

— From page 104

sense of general market conditions before committing to any positions. For example, Loeb warns readers that increasing price action and volume may mean different things depending on market conditions. Individual stock movement must always be put into the context of the underlying market. The Battle for Investment Survival contains Loeb’s original, short text, as well as additional chapters that reiterate the book’s main themes. Loeb’s advice may seem simplistic and unoriginal to experienced traders, but his ideas are powerful because they are much easier to dismiss than to act on with discipline. However, readers shouldn’t expect his advice to automatically lead to profits. In fact, Loeb’s goal is to remind his readers just how risky the markets can be. Ý

15

Technology for TRADERS

Software SCREENING:

Ensign Windows

SOFTWARE SUMMARY

BY DAVID BUKEY

E

nsign Windows is a real-time chart analysis program with features similar to higher-end trading software, but without the high-end price tag. Ensign’s design interface may seem less polished than its peers, but the software is quite comprehensive and flexible. Its programming language — Ensign Software Programming Language (ESPL) — allows advanced users to create chart studies and indicators, trading systems and market scans. However, the language’s complexities may frustrate beginning and intermediate users.

Data sources Ensign Windows is compatible with real-time data feeds from eSignal, BMI and DTN Satellite (at pre ss time DTN.IQFeed compatibility was scheduled for Fe bruary). Ensign Internet is an end-of-day program that costs the same as Ensign Windows and downloads delayed intraday, daily, weekly or monthly chart data directly from Ensign, bypassing the need to subscribe to a separate end-of-day data source. Both versions can handle intraday futures data and offer the same features.

Features

Program: Ensign Windows: version date 12/20/02 What it is: Real-time chart analysis software that includes customizable charts, quotes, news, paper-trading, scanning, system design and back-testing capabilities. Who it’s for: Stock, option and futures traders Skill level: Intermediate to advanced users Company: Ensign Software, 2641 Shannon Court, Idaho Falls, ID 83404 Tel: (801) 328-1382 Fax: (208) 525-8781 Web site: www.ensignsoftware.com; www.ensignsupport.com Price: $39.95 per month for both real-time and end-of-day versions. Additional data feed subscription needed for real-time Ensign Windows. Seven-day free trial can be downloaded from Web site. Upside: Flexible, powerful ESPL programming language. Comprehensive quote screens. New Play Back and Sim-Broker features make simulated trading easier. Various training and customer support resources. Downside: The programming language is difficult to understand for non-programmers and the manual is not as user-friendly as it could be. Minimum IBM compatible computer; Windows 98, Windows ME, system Windows 2000, Windows NT or Windows XP; requirements: 300 MHz Pentium PC; 100 MB of available hard disk space; 256 MB of RAM memory; 800 x 600 Video resolution (higher resolutions preferred); Mouse (or equivalent pointing device); Internet connection for downloading data, upgrading software and receiving Internet data feeds.

Charting. Ensign’s charts are user-friendly — they have many options and are easy to modify. Charts can be displayed over a variety of time frames (tick and intraday to daily, weekly and monthly), and you can choose from four chart types (bar, candlestick, line and close ticks) and display them in either regular or logarithmic scale. You can choose how much historical data to view, but you cannot customize the range of the data. For example, there is no way to view a chart just for the first week of December 2002. You can overlay more than 30 predefined chart studies and indicators on a price series, or plot them in one of nine subwindows. The Color Bars function scans a chart for various predefined criteria and identifies 25 types of bullish and bearish characteristics on both price and volume bars. For instance, selecting the “Close Outer 10%” color bar scan changes bars with closing prices in the top 10 percent of the bar to green and changes bars with closing prices in the bottom 10 percent of the bar to red. Similarly, the “Close vs. Open” scan distinguishes bars that closed higher than they opened from those that closed lower than they opened. In addition to standard trendlines, Ensign includes more

16

than 20 line studies, including Fibonacci levels, Gann Fans and Andrew’s Pitchfork. It also includes more unique tools such as PyraPoint, a Gann-based line study that shows support and resistance levels and projects price turning points. Like other chart elements, these tools are all easy to apply and modify. Right-clicking anywhere on a chart allows you to add, delete and change its objects and properties. You can change a chart’s background, gridlines and bars to any imaginable color. Toolbars allow you to edit the time frame and chart type, apply studies, indicators and color bars, and draw trendlines with a couple of mouse clicks. After tweaking a chart to your liking, you can save it as a template that will add your analysis settings to any chart you open. Once you open multiple charts and quote screens, you can save the layout as a workspace. Quotes. Ensign’s real-time quote screens are comprehensive and clearly organized. Pre-formatted quote pages are divided into categories such as NYSE, Nasdaq, indices, futures, corporate and municipal bonds, and foreign stocks. You can create


FIGURE 1 ENSIGN CHARTS Setting up multiple chart and quote windows as a workspace enables Ensign Windows users to track market changes. Here, an S&P E-Mini daily futures chart and Microsoft daily chart are shown with a standard quote screen and Level II window.

Source: Ensign Windows

custom quote pages to keep track of hundreds of stocks, bonds and futures. Quote page entries can be modified and sorted based on as many as 51 price, volume and fundamental categories (depending on your data feed). Saving quote screens to Excel, MetaStock or AIQ formats, or establishing a Dynamic Data Exchange (DDE) link to an Excel or Quattro Pro spreadsheet, is a snap. Loading a chart is as simple as double-clicking on the appropriate symbol. Ensign includes Level II and time and sales windows that are similar to other mid- to upper-echelon platforms. In addition, you can overlay market maker bid and ask prices onto an underlying stock chart to compare changes in both. Paper trading. Ensign has developed three unique ways for users to develop their trading skills via “paper trading”: using DEMO, a simulated stock; replaying the price action of previous days; or placing orders through Sim-Broker. DEMO’s simulated prices are updated every second. The program can generate up to four days of intraday data. DEMO behaves just like any other security, except it trades whenever you want it to — just pick a start price and market phase (random, trending or correction) and the program generates data, starting at the beginning of the trading day, and then scrolls


intraday price bars on a chart. Also, DEMO is us eful for training or practicing with other people because it can be synchronized among different computers. Play Back works like a tape recorder and downloads intraday price data from a previous trading session for any security. The program then plays it back, tick-by-tick on a chart, in real time or up to 10 times the normal speed. The ability to specify start times is convenient if you are interested in a particular period of a past day. Sim-Broker places simulated trades and keeps track of prof it and loss on real data and DEMO, and works well with Play Back. You can enter market, limit, stop or stop-and-reverse orders while choosing quantity and commission fees. SimBroker includes a transaction log that displays pending and filled trades. It also places arrows on charts to mark entries and exits and summarizes each trade’s profit and loss. Programming. The most powerful — but least accessible for non-programmers — element of Ensign is its ESPL programming language. ESPL is similar to the Delphi or Pascal programming languages. Unlike some system-design program languages, ESPL is geared toward serious programmers; it cannot be learned in an afternoon.

17

FIGURE 2 ENSIGN TRADING SYSTEM PROGRAMMING SCRIPT ESPL programming is at the heart of Ensign Windows’ capabilities. The code here backtests trading rules on IBM’s historical price data and shows the profit and loss results.

Source: Ensign Windows

However, those with programming skills will be able to take by keyword. News alerts will also notify you of articles conadvantage of ESPL’s ability to design original chart studies, taining keywords. Double-clicking a headline will display the indicators, trading systems and market scans. There is virtual- full article at the vendor’s Web site. Options analysis. Ensign has basic option analysis tools. ly no limit to what you can create. The language is flexible — ESPL functions can automate the operation of many of The options spreadsheet uses the Black-Scholes model to estiEnsign’s features and can access, modify and close any chart, mate an option’s value. In addition to listing values at different quote screen, trading system or scan within a program. strikes, the program also calculates values such as implied With ESPL, experienced users can design and test trading sys- volatility and delta. Trad ing . Ensign Windows now offers online trading tems on any individual security and view the performance results. You can also use ESPLto create market and chart scans to through Preferred Trade. Double-clicking on the bid or ask find stocks, bonds and futures with user-specified criteria. For price in a quote screen will initiate a trade. Ensign allows you example, ESPL will scan all stocks on the Nasdaq and display to track up to 500 trading accounts (more than enough for your only those with the volume and price characteristics that you indi- average trader these days, we’re guessing). cate. However, while these two tools are useful, again, their reliance on ESPLmakes them less accessible to non-programmers. Customer support Learning ESPL can be particularly frustrating because its Ensign offers a free customer support line and its Web site prodocumentation is also somewhat complex, and seems to be vides chart and line study examples, downloadable templates, written for a programming-literate audience. Although the newsletter archives, ESPL programming lessons and sample ESPL manual does contain sample code to experiment with, code. In addition, Ensign programmers host tutorials and beginning and intermediate users could probably use more answer questions each week in the Ensign chat room tutorials, examples, and pre-coded functions and systems to (www.paltalk.com). help them learn the basics of the language. A programming drawback of Ensign is that it does not have Bottom line a library of pre-defined trading or analysis functions or vari- Ensign Windows’ charts, quotes, back-testing and scanning abilables you can link together when designing technical studies ity are of the level normally found in more expensive software . or systems. This means you must declare all variables and pro- The program’s reliance on its ESPL programming language is cedures — there are no short cuts. both its greatest asset and biggest drawback: Programmers can For example, you can plot a simple, custom indicator in use ESPL to customize Ensign’s features and create sophisticatMetaStock with one line of its formula language (i.e., a 10-day ed analysis tools and trading systems, but the language’s learnAverage True Range divided by its 60-day Average True Range: ing curve and inadequate manual may prevent new users from ATR(10)/ATR(60)). Plotting the same indicator with ESPL taking advantage of all Ensign has to offer (at least initially). requires at least 30 lines of code. Ensign Windows is a compelling program with many powNews. Ensign’s news options are extensive. Headlines are erful and clever features. Traders who are comfortable writing organized by vendor and subject, and can be saved or searched code will appreciate its value. Ý 18


TRADING Strategies

How to trade BROADENING T OPS In previous articles, Tom Bierovic has discussed several classic chart patterns, including flags, head-and-shoulders, inside days, rectangles, triangles and wedges. He completes his series with a lesson on broadening tops — perhaps the least appreciated and most misunderstood of major patterns. FIGURE 1 SYMMETRICAL BROADENING TOP

BY THOMAS A. BIEROVIC

S

ome technical analysis books that feature chart patterns ignore broadening tops entirely. Others claim broadening tops are so hard to identify and trade, it’s lucky they occur so infrequently. In reality, broadening tops are no more difficult to find and trade than most other patterns. They occur often enough to be worth looking for and, most importantly, they are as accurate and profitable as other major chart formations.

Why no broadening bottoms? Edwards and Magee addressed this question in 1948 in their book, Technical Analysis of Stock Trends: “It has been assumed in the past that Broadening Bottoms must exist, but this writer has never found a good one in his examination of the charts of thousands of individual stocks over many years, and only one or two pat terns which bore a resemblance to it in the charts of the averages. Apparently, the circumstances which create Broadening Formations do not exist after a prolonged decline in prices. This would seem to bear out our earlier characterization of this sort of pattern

In a symmetrical broadening top, the upper line connects rising swing highs and the lower line connects declining swing lows. A sell signal occurs when price penetrates the pattern’s bottom line.

E

C

A

B

D

F

as suggesting active, excited trading with much public (and, hence, not too well informed or managed) participa tion. Such conditions are naturally associated with the final phases of a Bull Market.”

Symmetrical broadening tops There are three kinds of broadening tops: symmetrical, ascending and descending. Figure 1 shows a symmetrical broadening top, the most common of the three types. The salient feature distinguishing broadening tops from many other chart patterns is that price swings within a broadening top increase in size, whereas the swings within other patterns stay the same size (as in rectangles) or decrease in size (as in triangles and wedges). Note the rising top line in Figure 1 that connects a series of higher swing highs and the declining bottom line that


connects a series of lower swing lows. (Aswing high has a lower high to its immediate left and right; a swing low has higher lows to its immediate left and right.) In a proper broadening top, the swing highs and lows alternate on the horizontal axis: Point A is a high, point B a low to the right of A, point C a high to the right of B, point D a low to the right of C, and so on.

Trading the pattern A broadening top is a reversal pattern — i.e., it usually signals a trend change from up to down. In Figure 2, an uptrend precedes the beginning of the pattern at point A, while a downtrend follows the breakout from the pattern at point F. The following examples illustrate an effective way to trade broadening tops: Sell short when price drops below the bottom line of the pattern, place a conservative protective stop and take profits at a pre-defined price objective. In Figure 2, sell short below the bottom line (58.51) of the pattern at point F (58.50) and set a protective stop at the high of the day before entry (59.53), risking 1.03 on the trade. Next, determine two price objectives: Goal 1, a signal to lower the stop to 19

FIGURE 2 SETTING STOPS AND PRICE OBJECTIVES Goal 1 was reached in two days and the protective stop was lowered to breakeven. Goal 2, the price objective, was reached 16 days later. 63

Loews (LTR), daily

E

62

C

A

61 60 Protective stop

B

59

D

Sell short

Breakeven stop

H

Goal 1

58

F 57

G

J

I

56 55

Goal 2

K L

25

April

8

15

22

29 May 6

13

20

June

10

54 17

Source: TradeStation Platform by TradeStation Group

FIGURE 3 A SLIGHT TWIST In this example, the price swing that began at D failed to reach the broadening top’s upper line, ending instead with a swing high at E. Such a partial rise is usually a bearish sign. Also, the stop was placed at the midpoint of the candle before entry because placing it above the high of the candle would represent too large a risk. William Wrigley Jr (WWY), daily

60

C

A

E

58

F 56

Protective stop

B

Sell short

D

54 Goal 1 52

G Goal 2

50

H

48 46 44

13

20

June

10

17

24


20

July

8

15

22

29 Aug.5

breakeven, and Goal 2, a signal to take profits. Goal 1 is a multiple of the trade’s initial risk subtracted from the entry price. Any multiple between 1.15 and 1.55 is reasonable — the objective is for Goal 1 to be a little farther below the entry price than the initial protective stop is above it. The following examples use a multiple of 1.25. Goal 2, the price target, is the height of the pattern subtracted from its bottom line on the day of the breakout. To calculate Goal 1 for the trade in Figure 2, multiply the trade’s risk by 1.25 (1.03*1.25 = 1.29) and subtract the result from the entry price (58.50 - 1.29 = 57.21). To determine Goal 2, first measure the height of the broadening top by subtracting the price level of the bottom line on the breakout day (bar F) from the price level of the top line directly above it (62.60 - 58.51 = 4.09). Then, subtract the pattern’s height from the bottom line at F (58.51 - 4.09 = 54.42). After a close below Goal 1 (57.21), lower the protective stop to breakeven (58.50, the entry price). Some broadening top trades have an additional trade-management step between moving the stop to breakeven (at Goal 1) and taking profits (at Goal 2). After a stock posts its lowest close since the trade entry, look to the left on the chart between the lowest close and the entry for a swing high below the breakeven stop. If you find one, lower the stop to that swing high. (If you find more than one swing high, lower the stop to the most recent one.) For example, in Figure 2, price posts a new low close at point I, so lower your stop to 58, the swing high at point H, which is the most recent swing high to the left of I. (Although technically a swing high, the high at H is better described as a “tweezer top,” candlestick terminology


for consecutive highs at the FIGURE 4B DESCENDING BROADENING TOP FIGURE 4A ASCENDING BROADENING TOP same, or nearly the same, In an ascending broadening top, the In a descending broadening top, the price.) At point K, price pattern’s upper line connects rising swing upper line is flat, and the lower line is makes another new low close, highs and its lower line connects declining. so lower your stop to the horizontal swing lows. swing high at J (56.10). Take profits of 4.10 when price declines to Goal 2 (54.42). Figure 3 shows a common A C E variation in broadening-top E C formations. Note that the A price swing that began at the pattern’s bottom line at point D failed to reach the top line, B ending instead at the swing D B D F F high at E. This partial rise is a bearish sign: Price wasn’t strong enough to rally all the way to the top of the pattern. Sell short at candle F at 54.70, on the open below the pattern’s bottom line. line and a falling bottom line (Figure 4b. at candle F at 20.92, below the pattern’s The high of candle F is too far from the Figure 5 shows a market example of bottom line, and set a protective stop at entry price to provide a reasonable point an ascending broadening top. Note the 21.61, a little above the midpoint of the for a protective stop, so set the stop just rising line connecting the swing highs at tall, black candle before F. above the midpoint of the candle’s real points A and C, and the flat line connectCalculate Goal 1 by multiplying the body instead. To determine the stop, ing the swing lows at B and D. Sell short trade’s risk by 1.25 and subtracting the subtract the bottom of the candle’s real body from its top, FIGURE 5 TRADING AN ASCENDING TOP divide the difference by two, and add that number to the bot Ascending and descending broadening tops are traded the same as the symmetrical tom of the real body (56.60 version of the pattern. Note in this example that seven days after entry, price 55.00 = 1.6, 1.6 divided by 2 = pulled back to the pattern’s bottom line. Pullbacks after breakouts are a common .80, and .80 + 55.00 = 55.80). Set occurrence. the stop at 55.85, a little above Ceridian Corp (CEN), daily the midpoint, for a risk of 1.15. 23.5 To calculate Goal 1, multiply E 23.0 the risk by 1.25 and subtract the C A difference from the entry price 22.5 (1.15 x 1.25 = 1.44, and 54.70 – 22.0 1.44 = 53.26). To determine Goal Protective stop 2, subtract the height of the pat21.5 tern at F (59.66 - 54.90 = 4.76) 21.0 from the bottom line at F (54.90 Sell short B D F 4.76 = 50.14). After price closes 20.5 below 53.26 at G, lower your Goal 1 20.0 stop to breakeven (54.70, the G entry price). Take profits (4.76) 19.5 at H when price falls to 50.14.

Ascending and descending broadening tops

The final examples consist of ascending and descending broadening tops. An ascending broadening top has a rising top line and a flat bottom line (Figure 4a, above); a descending broadening top has a flat top

19.0

Goal 2

H

18.5 18.0 17.5

April

May

June

July

17.0



21

FIGURE 6 TRADING A DESCENDING TOP

result from the entry price (.69 x Note that after Goal 1 was reached, the stop was lowered two times before the 1.25 = .86, and 20.92 - .86 = exit — first, to the swing high at H after the new low close at I and, second, to the 20.06). Determine Goal 2 by subswing high at J after the new low close at K. tracting the bottom of the pattern at F from its top at F and ACE Limited (ACE), daily subtracting the difference from E A C 45 the bottom (23.02 - 20.93 = 2.09, and 20.93 - 2.09 = 18.84). After 44 the close below Goal 1 (20.06) at 43 candle G, lower the stop to 42 breakeven. Take profits (2.09) at candle H at 18.84. 41 Protective stop In Figure 6 (above), price B 40 forms a descending broadening D 39 top. The line connecting the Sell short swing highs at Aand C is flat; the F 38 line connecting the swing lows at 37 H B and D is declining. Sell short at Goal 1 F (at 38.89), just below the bottom 36 J G line. The open of candle F (41.00), 35 I the top of the real body, provides 34 a reasonable place to set a protecK tive stop (the high at F is too far 33 Goal 2 L away from the entry price for a 32 prudent stop). The risk is 2.11 February March April May June (41.00 - 38.89 = 2.11). Source: TradeStation Platform by TradeStation Group Calculate Goal 1 by multiplying the risk by 1.25 and subtracting the result from the entry price (2.11 x works, it has not yet been shown these 1.25 = 2.64, and 38.89 – 2.64 = 36.25). sophisticated techniques produce better Determine Goal 2 by subtracting the bot- results (i.e., bigger profits) than the simAdditional reading tom of the pattern at F from its top at F ple chart patterns that have been tested and subtracting the difference from the and proven in the markets by thousands Playing for Keeps in Stocks & of traders over the last 70 years. bottom (44.76 – 38.90 = 5.86, and 38.90 – Futures: Three Top Trading A 1948 quote from Edwards and 5.86 = 33.04). Strategies That Consistently Beat Price closes below Goal 1 at candle G, Magee still rings true today: the Markets so lower your stop to breakeven. Price by Thomas A. Bierovic “…Prices move in trends and trends posts its lowest close since entry at point (John Wiley & Sons, New York, tend to continue until something hap I, so lower the stop to the swing high at 2001) pens to change the supply-demand bal H. After the next new low close (at K), ance. Such changes are usually lower the stop to the swing high at J. Technical Analysis of Stock Trends detectable in the action of the market Take profits (5.86) at candle Lat 33.04. by Robert D. Edwards itself. Certain patterns or formations, and John Magee levels or areas, appear on the charts, Wrapping things up (Originally published in 1948, which have a meaning and which can All the patterns in this series ( Active Seventh Edition, CRC Press, be interpreted in terms of probable Trader , August 2002 to April 2003) on Boca Raton, Florida, 1997) future trend development. They are not classic chart patterns were described by infallible, it must be noted, but the odds Richard Schabacker in 1932 in Technical Technical Analysis are definitely in their favor. Time after Analysis and Stock Market Profits and by and Stock Market Profits time, as experience has amply proved, Robert Edwards and John Magee in 1948 by Richard Schabacker they are far more prescient than th e best in Technical Analysis of Stock Trends. (Originally published in 1932, informed and most shrewd of statisti Although technical analysts today can reissued by Pitman Publishing, cians.” Ý employ new, advanced tools such as artiLondon, 1997) ficial intelligence, chaos theory, fuzzy logic, genetic algorithms and neural net- For information on the author see p. 3. 22


TRADING Strategies

Narrow-range bars and inside bars represent short-term

Opening shots

volatility lows out of which price can move sharply. This strategy uses a simple volatility measurement to determine where to enter trades to capitalize on this behavior. FIGURE 1 IDENTIFYING SHORT-TERM VOLATILITY CHANGES The bars at A, B and C are all relatively narrow-range bars that represent short-term consolidations. Applying specific entry rules allows you to take advantage of moves out of the consolidation when volatility increases.

IBM (IBM), daily

BY THOM HARTLE

82.50 80.00

C

I

n the late 1980s, Tobey Crabel published a series of articles detailing various short-term price patterns for futures trading. The patterns were designed to detect short-term volatility changes — that is, a market would move into a short-term consolidation, and then break out of it. Two of Crabel’s patterns will be covered here. The first is an inside day that is smaller than the previous four days (IDnr4) and the second is a day whose range is less than that of the previous six days (NR7). An NR7 day does not have to be an inside day. Figure 1 contains examples of these price patterns (in a stock rather than a f utures contract). Bar A is an IDnr4: It is an inside day and its range is smaller than the ranges of the previous four days. Bar B is an NR7 day. In this case, the day’s range is smaller than the previous six days’ ranges. Finally, bar C qualifies as both an IDnr4 and an NR7 day.

77.50 75.00

B

72.50

A

70.00 67.50

29

August 1 5

12

19

26

Source: CQGNet

Analyzing IBM daily bar data fro m Oct. 1, 2001, to Nov. 15, 2002, revealed that 15 IDnr4 and 44 NR7 days occurred over this 286-trading day period. This means these patterns are somewhat rare. However, many trading opportunities present themselves if you look for these patterns across a number of stocks.


Trading the patterns Crabel recommended trading these patterns the next day using an entry strategy based on the opening price and a preset value called the “stretch.” He had two versions of the strategy — one called the opening range breakout (ORB) and the second the opening range breakout preference (ORBP). 23

FIGURE 2 TAKING A STRETCH The “stretch” is an amount added to a bar’s opening price to determine the entry point and stop level, respectively. $1.20 $1.00 $0.80 $0.60 $0.40 $0.20 $0.00 Day

FIGURE 3 FACTORING IN THE TREND Whether a market is in a trend (defined here by a 30-day EMA) dictates the trade entry technique that is used.

The stretch value is the 10-day average of the difference between the open and the low or high, whichever is smaller. For example, if the open was 79.50, the high was 80.33 and the low was 77.98, you would use the diff erence between the open and the high (0.83) vs. the open and the low (1.52). Figure 2 shows the stretch range for IBM over the last year: from an extreme of just under $1 to just over $0.20, with the average just over $0.50. The entry strategy using ORB is to place a buy stop just above the opening price plus the stretch amount, and place a sell stop just below the opening price minus the stretch amount. The first stop triggered is the trade, and the other automatically becomes the protective stop. If other technical indicators provide evidence of a strong trend, then use the ORBP: Place the entry stop in anticipation of the trend continuing. For example, if the stock is in an uptrend, enter only a buy stop above the opening price plus the stretch amount. If filled, place the protective stop below the stretch subtracted from the open.

Time is of the essence IBM (IBM), daily

82.50 80.00

C

77.50 75.00

B

72.50

A

70.00 30-day EMA

67.50

29 Source: CQGNet

24

August 1 5

12

19

26

According to Crabel, the earlier in the trading session the entry stop is hit the more likely the trade will be profitable at the close. A trend triggered quickly in the session could rack up a substantial profit by the close and should be held for a possible two- to three-day run. Once filled (and if the market is exhibiting a strong trend early), then move the protective stop to breakeven. If you are not filled early in the session, reduce the size of the position as time passes through the day. Any positions filled near the close are suspect. In this case, you may have only a small unrealized profit, and probably should not hold the trade overnight.

Rules in action Figure 3 is Figure 1 with the addition of a 30-day exponential moving average (EMA). Bar A is the IDnr4 day. Considering the market moved above the previous resistance level at 72.50 in www.activetradermag.com • April 2003 • ACTIVE TRADER

the act of confirming a double bottom pattern, and the 30-day EMA was rising and provided a support level, it’s logical to conclude the stock is in an uptrend. Accordingly, we want to use the ORBP guidelines. At this time, the stretch was 0.49. The next day IBM opened at 71.55, so the entry stop price was 72.04 (71.55 + 0.49) and the stop loss was 71.06 (71.55 – 0.49). That day IBM climbed as high as 73.78, fell as low as 71.19, and closed at 71.90. In this situation, when there is no profit at the close, you should take the loss (in this case, 14 cents plus any slippage). Bar B is the NR7 bar. The market is

closed at 79.35 for a profit of 2.37 that day, not including slippage. This open profit warranted holding the position overnight. The next day, the stretch is used as a trailing stop, based on the opening price (79.35). The stretch had increased to 0.53, making the stop level 78.82 (79.35-.53), which was not hit. The next day, the stock opened at 81.56. Subtracting the stretch (0.53) sets a new stop at 81.03, the approximate level at which the trade would have been stopped out. Bar C is the combination IDnr4 and NR7 bar. Still trading from the long side, the market opened the next day at 81.90.

FIGURE 4 DETERMINING LOGICAL TARGETS To increase the odds of profitability, you can take partial profits when price moves in your direction by 66 percent of the 10-day average daily range. 4.0

10-day MA of the daily ranges

3.5 3.0 2.5 2.0 1.5

Stretch

1.0 0.5 0 Day

taking, you would have been stopped out at 81.48 (using the stretch the following day to determine the stop-loss).

Profit targets One way to take partial profits is to use a percentage of the 10-day moving average of the daily ranges as a target. Figure 4 compares the 10-day average of the daily ranges compared to the stretch amounts. The average daily range is 2.55, but it ranges from just above 1.50 to just over 3.50. A reasonable target is 66-percent of the current 10-day moving average, just to be sure to put away some partial profits when you have the chance. For example, at the bar after A in Figure 3, the current 10-day moving average of the range was 2.78. Assuming that once you are stopped into the long position you are confident the low at 71.19 is in place for the day, add 66 percent of 2.78 (1.85) to 71.19 for a target of 73.04. This would have given you partial profits of 0.98 on this day before slippage, which offsets the loss taken had you held the position at the close. In addition, had you traded against the trend after the bar C trigger, you would have been short at 81.37. The 10day moving average of the daily ranges had increased to 3.06, so the targ et would have been the high of 82.33 – 2.04 = 80.31 (3.06 * 0.666 = 2.04), which would have been reached. Finally, once in a trade, consider moving your stop to breakeven if the market has reached a point that represents 50 percent of the 10-day average of the daily range.

Be a specialist clearly in an uptrend, and the prudent thing to do is trade with the trend, again following the ORBPguidelines. After the NR7 day, the stock opened at 76.50; the stretch was 0.48. In this case, the stock made a low of 75.98 (0.52 below the open) in the first few minutes, which would have triggered a short sale if we did not use the ORBP rule to trade with the trend. The long-entry stop was 76.98, which was hit later in the morning. The stock

Adding the current stretch of 0.53 created an entry stop of 82.43. But because the high was 82.00 on this day, no long trade was executed. As an example of what would happen taking the trade against the trend, subtracting 0.53 from the opening of 81.90 would have put you short at 81.37. The stock closed at 81.00, which would have resulted in a small profit if you exited at the close. If you stayed short, however, expecting some follow-through prof it


Avery important aspect of applying this kind of pattern-based technique is your ability to execute trades with as little slippage as possible. Identifying these types of price patterns, and combining them with rules for entering, managing risk and taking profits in various markets amounts to specializing. And specialization is as popular and prof ita ble an endeavor in trading as it is in other disciplines.Ý For information on the author see p. 3. 25

TRADING Strategies

OUTSIDE BARS: From concept to strategy A trading strategy usually begins with a simple observation about market behavior. In the first of two articles, we explore the “meaning” of outside bars and illustrate the process of designing a trading strategy around our conclusions.

BY TIMOTHY ANDREWS

A

nyone who has attempted to design a technical low and a higher high than the previous bar reflects upward trading system knows how difficult it can be to momentum, especially if it closes near the top of the bar’s develop rules that consistently and mechanically range and higher than the previous bar’s close. Several such capitalize on price patterns. bars in a row would constitute a clear short-term uptrend. Because we tend to naturally focus on the potential profit Although they do not provide quite as decisive a picture of (instead of the risk) in a trade scenario, our eyes sometimes momentum as this, outside bars have certain characteristics deceive us by seeing chart formations that, though obvious that, when placed in the context of the surrounding market and easy to trade in hindsight, are much more ambiguous and conditions, can signal which direction a market is likely to difficult to define when unfolding in real-time trading. move in the near future. In short, outside bars can provide Also, we tend to focus on patterns and signals that seem to potential entry signals, if you know what to look for — and put us in the market at advantageous times, without regard to what to look out for. how we will get out of trades or how much risk we will be exposed to along FIGURE 1 OUTSIDE BARS the way. An outside bar has a higher high and a lower low than the previous bar. Walking through the process of develDepending on market conditions, outside bars can signal reversals or contin oping a rule-based trading strategy uations of the existing price trend. based on a simple, “visual” price pattern highlights these challenges, but also Natural Gas (NG), daily hones the skills necessary for developing better trading systems in the future as 3.600 well as reinforcing trading habits that benefit both discretionary and systematic traders. 3.400

The outsiders The pattern that will form the basis of the strategy is the outside bar, which is a price bar with a higher high and lower low than the preceding bar (see Figure 1). Because they probe both higher and lower than the bars that precede them, outside bars are sometimes viewed as ambiguous price events — they send out mixed signals about market direction by moving above and below the preceding bar, or bars. As a result, traders often ignore them and look for more obvious price patterns. For example, a price bar with a higher 26

3.200

3.000

2.800

28

3 10 17 24 1 8 June July Source: QCharts Quote.com

15

22

29

5 12 August

19

26

3 9 September


FIGURE 2 SOME GOOD, SOME BAD, SOME SO-SO

The challenge is developing ru les based on these characteristics. A quick review of the general price characteristics of outside bars will lay the groundwork for the strategy.

The vertical lines identify outside bars. Some of these bars — including those at the more significant tops and bottoms — seem to provide potential ly good entry signals, but defining them objectively is more difficult than finding them on a chart. (Note: The October low bar is an outside bar; its dashed line is not visible.) S&P Depository Receipts (SPY), daily

What outside bars represent

96

94 The most obvious characteristic of an outside bar is that it represents a volatili92 ty increase — an expansion of the price range — from the previous bar. 90 Volatility, however, is non-directional 88 — it doesn’t tell you anything about the direction a market is moving, or is likely 86 to move. A high-volatility market could be moving up dramatically, down dra84 matically, or thrashing back and forth in 82 a wide trading range. Or, if the high volatility conditions have existed for a 80 while, it could mean the market is ready to “quiet down,” after exhausting itself. 78 Viewed in isolation, outside bars are September October November December January 2003 typically considered non-dire ctional. However, looking at any market event in Source: TradeStation Platform by TradeStation Group isolation rarely gives an accurate impresFrom one perspective, this kind of continuation behavior sion of its meaning. Similar to the trending bars (higher highs, higher lows and higher closes) described earlier, we can look implies the “indecision” often attributed to outside bars is for clues in outside bars themselves — how they close and how actually a sign of trend strength, up or down. For example, if a stock forms an outside bar on a day when a piece of bearish their highs and lows compare to previous bars — as well as in news hit the market — e.g., a poor earnings announcement or the surrounding price action to find tradable price moves. an increase in unemployment — it could be evidence the marOutside bars occur on all time frames, but they are less significant (although not completely meaningless) on intraday ket has the strength to absorb the sell-off that followed the time frames because the position of the close within the bar — news and still rally to a new high. However, the most conspicuous outside days in Figure 2 and how the high and low relate to the previous high and low were those that accompanied the October swing low and the — are more arbitrary on five-minute or 10-minute (or 60December and January swing highs. These bars closed in the minute, etc.) bars than on daily or weekly bars. As a result, the opposite direction of the short-term trend — at the upper end strategy is based on daily bars. of the bar during a downtrend and the lower end of the bar during an uptrend. However, the October low outside bar First impressions Figure 2 shows a daily chart of the S&P Depository Receipts closed below the EMA and the December and January high outside bars closed above the EMA. (SPY) with all outside days marked with dashed blue lines. The first impression this chart gives is that outside bars There are 13 outside days among the roughly 112 days shown might be useful tools for identifying intermediate-term reveron this chart. sals, and to a lesser extent, momentum “thrusts” in the direcThe price action that follows them is mixed: The second September outside day, second October day, first December tion of the short-term trend. day and mid-January outside day were followed by significant price moves; others, including the first September outside day, Distinguishing characteristics first October day and second December day, were followed by The next task is to translate these observations about outside bars into logical statements that can be incorporated in a stratsmaller moves. However, few of these bars were followed by much sideways egy and tested objectively. Let’s first address the characteristics of those outside bars movement; for the most part, somethin g usually happened that appear to offer the most favorable trade opportunities — shortly after most of the outside bars. Closer examination the reversal bars. In terms of a bullish outside reversal bar that reveals several of the outside bars that are followed quickly by occurs at the bottom of a downtrend, the bar should close notable up moves or down moves share a few characteristics. First, many of the outside days tended to close in the dire c- strongly (in the opposite direction of the short-term trend, as tion of the short-term trend, which is re presented in Figure 2 by noted earlier). That could be defined as a close that is in the a 10-day exponential moving average (EMA). The days with upper half of the bar’s range, above the open, and above the close of the close two days ago (instead of just above yesterhigh closes were more often than not above the EMA, while the opposite was true of the days with low closes. These bars were day’s close). These criteria are summarized in the following logical statement: often followed at least by a brief continuation of the trend.


27

FIGURE 3 OUTSIDE REVERSAL BARS The green and red bars are the outside bars from Figure 2 that met the cri teria we developed to identify reversal points. Only a handful of the outside bars met the requirements.

(high - close)/(high - low) < .50 and close > open and close > close[2]

S&P Depository Receipts (SPY), daily

96 94

where

[1], [2], etc., represent “one day ago,” “two days ago,” etc.

92 90

To make sure this bar is occurring in a downtrend, we can add the conditions that its 10-day EMA value be less than yesterday’s and that at least some of the bar is below the 10-day EMA:

88 86 84 82

10-day EMA< 10-day EMA[1] and low < 10-day EMA. Finally, adding the definition of the outside bar results in the following definition of the bars we are trying to identify:

80 78 September

October

November

December

January 2003


with the outside days again highlighted with dashed lines. Of the nine outside days shown here, only one (highlighted green) fulfilled the outside reversal bar criteria we’ve developed. One signal over five months constitutes a technical drought, but the time period was dominated by a major downtrend. This is a reminder that we need to revisit the other type of outside bar we identified — those that close in the direction of the trend and which can lead to short-term momentum moves — to see if we can make more use of outside bars. In the follow-up to this article, we will develop trading rules based on the information we’ve put together about outside bars, and test the strategy to see what kind of results we might get in real trading. Ý

high > high[1] and low < low[1] and (high - close)/(high - low) < .50 and close > open and close > close[2] and 10-day EMA< 10-day EMA[1] and low < 10-day EMA.

The definition would be inverted (low replacing high, high replacing low, “greater than” replacing “less than” and “less than” replacing “greater than”) for bearish outside reversal bars. Figure 3 shows the same time period as Figure 2, with bullish and bearish outside reversal bars colored green and re d , respectively. Only five of the 14 outside bars fulfilled the crite- For information on the author see p. 10. ria, but with the exception of the first and last outside bars, these criteria identified FIGURE 4 EXPANDING THE HORIZON several of the important intermediateterm turning points on this chart (and as A look at the period preceding the time window shown in Figures 2 and 3 of press time, the jury was still out on the reveals there were far fewer outside bars in general, and only one outside last bar on the chart). reversal bar (in July).

Hold your horses It’s too soon to have even the slightest idea whether these rules are the beginning of a good trading strategy. First, we’ve constructed these rules based on the observation of one chart — a limited data sample, to say the least. Second, we have not even determined a specific entry rule yet, let alone rules for controlling risk, taking profits or determining how much capital to risk per trade. All we know so far is that the ru les have identified some bars that occur in the vicinity of some of the important swing highs and lows on this chart. Figure 4 shows the five months leading up to the period in Figures 2 and 3, 28

S&P Depository Receipts (SPY), daily 115 110 105 100 95 90 85 80 April

May

June

July

August

Source: TradeStation Platform by TradeStation Group www.activetradermag.com • April 2003 • ACTIVE TRADER

FIGURE 1 EQUITY CURVE

Trend with pattern entry (TPE)

The light green area shows how much account capital is invested at any given time. The dark green area shows how much cash is available. 1,050,000 1,000,000

Markets: Stocks.

950,000

System concept: The Trend with Pattern Entry (TPE) system combines a trend-following technique with a simple pattern to pinpoint entries. If the system determines a stock is in an uptrend, it looks for a brief pullback and enters long on the assumption the trend will resume. The stop-loss level is based on current market volatility (expanding as volatility increases, and vice versa), which gives a trade sufficient breathing room to reach a profitable exit level. This test is for the long-only version of the system. The Futures Trading System Lab version (see p. 54) includes both short and long trades.

900,000 850,000 800,000 750,000 700,000 650,000 600,000 550,000 500,000 450,000 400,000 350,000 300,000 250,000 200,000 150,000 100,000 50,000 0 12/1/92

Rules: 1. Trend-detection rule: The closing price must be above the 100-bar moving average.

11/2/93

11/2/94

11/2/95

Equity

Cash

FIGURE 2 SAMPLE TRADES This chart of Cisco Systems (CSCO) shows entry and exit points during an uptrend, along with the 100-bar moving average that defines the long-term trend. Sell

Cisco Systems (CSCO), daily

Sell

28.00 27.00

Sell

1 1/1/96

26.00 25.00

11/4/98

Linear reg

11/5/99

11/6/00

12/3/01

12/2/02

Buy & hold

2. Pullback entry rule: When two consecutive bars close below the previous bar’s close, enter long on the next bar’s open. 3. Stop rule: The stop-loss level is equal to the closing price minus four times the 10 bar average true range (ATR). Keep this stop-loss order active for the duration of the trade. 4. Exit rule: Exit at the market open when the trade closes with a 5-percent profit.

24.00

Buy Sell

23.00 22.00

Buy Buy

21.00 20.00 19.00 18.00

Buy

17.00 16.00

Volume 20M

January 1999 Source for all figures: Wealth-Lab Inc. (www.wealth-lab.com)


February

Money management: Risk 2 percent of current account equity per position. This is done by calculating the difference between the position’s “basis price” (the closing price when the signal was triggered) and the stop-loss level. Buy the number of shares that results in a 2-percent loss of account equity when the stop-loss is hit. For example, given a current equity level of $120,000, a basis price of $20.55, and a stop-loss price of $17.40, the trade’s maximum point loss is $3.15. Two percent of the current equity is $2,400; dividing this amount by the maximum point loss gives the number of shares to buy ($2400 / 3.15 = 762). Starting equity: $100,000. $10 slippage/ commission deducted per trade. 29

FIGURE 3 DRAWDOWN CURVE A fairly long drawdown started in mid 2000, coinciding with the broad market’s decline.

Test data: The system was tested on the Active Trader Standard Stock Portfolio, which contains the following 18 stocks: Apple Computer (AAPL), Boeing (BA), Citibank (C), Caterpillar (CAT), Cisco (CSCO), Disney (DIS), General Motors (GM), Hewlett Packard (HPQ), International Business Machines (IBM), Intel (INTC), International Paper (IP), J.P. Morgan Chase (JPM), Coke (KO), Microsoft (MSFT), Sears (S), Starbucks (SBUX), AT&T (T) and Wal-Mart (WMT).

0% -2% -4% -6% -8% -10% -12% -14% -16% -18% -20% -22% -24% -26% -28% -30% 12/1/92

11/1/93

Test period: December 1992 through December 2002. Test results: The system produced a 295-percent total net profit, compared to 328 percent for buying and holding over the same period. Total market exposure was 72 percent, which means on a risk-adjusted basis the system was slightly less efficient than buy and hold. The system experienced a substantial drawdown of 31 percent near the end of the test period. However, buy and hold had a 67-percent maximum drawdown during the test period. The system protects the portfolio from a more substantial drawdown by avoiding the market when prices are below the 100-bar moving average. The equity curve (Figure 1) shows the system is mostly in cash after prolonged down moves. System parameters: The system parameters for the test were selected subjectively, not as a result of optimization. A 100-bar STRATEGY SUMMARY

11/1/94

11/1/95

11/1/96

11/2/98

11/1/99

1 1/1/00

11/1/01

11/1/02

moving average (MA) captures the long-term trend while being more responsive to price changes than the more commonly used 200-bar MA. A 5-percent profit target is a reasonable goal. The ATR stop level gives the trade plenty of space, and should prevent the position from being stopped out because of random market noise. You can tweak or optimize these parameters to see the effect on the system’s dynamics. For example, by decreasing the ATR level, we’d expect to see the average loss decrease, but we’d also expect to see a larger number of losing trades (because of whipsaws). Adjusting system parameters usually involves trade-offs such as this. Bottom line: These simple rules yielded a moderately profitable trading system. It did not beat buy and hold in terms of profit on our test portfolio, but its drawdown was significantly smaller. In addition to the filter that limits trades to uptrends only, the position-sizing rule ensures a maximum total loss of 2 percent per trade. This combination is an effective defense against capital deterioration. Ý — Compiled by Dion Kurczek of Wealth-Lab Inc.

Profitability Net profit ($): 212,926 Net profit (%): 212.93 Exposure (%): 72.09 Profit factor: 1.22 Payoff ratio: 0.52 Recovery factor: 1.56

Trade statistics No. trades: 656 Win/loss (%): 72.87 Avg. gain/loss (%): 1.36 Avg. holding time: 17.79 Avg. profit (winners): 6.55 Avg. hold time (winners): 15.13

Drawdown

Avg. loss (losers):

12.57

Weekly

0.24%

0.78

7.72% -8.63%

56.08%

8

7

Avg. hold time (losers): Max. consec. win/loss:

24.91 23/7

Monthly

1.05%

0.79

13.50% -13.67%

56.20%

8

6

Quarterly 3.13%

0.78

18.95% -15.35%

68.29%

10

3

Annually 13.11%

0.76

35.49% -16.50%

80.00%

8

2

Max. DD (%): Longest flat days:

30.52 600

LEGEND: Net profit — profit at end of test period, less commission • Exposure — the area of the eq uity curve exposed to long or short positions, as opposed to cash • Profit factor — gross profit divided by gross loss • Payoff ratio — average profit of winning trades divided by average loss of losing trades • Recovery factor — net profit divided by max. drawdown • Max DD% — largest percentage decline in equity • Longest flat days — longest period, in days, the system is between two equity highs • No. trades — num ber of trades generated by the system • Win/Loss% — the percentage of trades that were profitable • Avg. profit — the average profit for all trades • Avg. hold time — the average holding period for all trades • Avg. profit (winners) — the average profit for winning trades • Avg. hold time (winners) — the average holding time for winning trades •Avg. loss (losers) — the average loss for losing trades • Avg. hold time (losers) — the average holding time for losing trades • Max. consec. win/loss — the maximum number of consecutive winning and losing trades

PERIODIC RETURNS Avg. Sharpe Best Worst return ratio return return

% Max. Max. Profitable consec. consec. periods profitable unprofitable

LEGEND: Avg. return — the average percentage for the period • Sharpe ratio — average return divided by standard deviation of returns (annualized) • Best return — best return for the period • Worst return — worst return for the period • % Profitable periods — the percentage of periods that were prof itable • Max. consec. profitable — the largest number of consecutive prof itable periods • Max. consec. unprofitable — the largest number of consec utive unprofitable periods Trading System Lab strategies are tested on a portfolio basis (unless otherwise noted) using Wealth-Lab Inc.’s testing platform. If you have a system you’d like to see tested, please send the trading and money-management rules to [email protected] .

Disclaimer: The Trading System Lab is intended for educational purposes only to provide a perspective on different market concepts. It is not meant to recommend or promote any trading system or approach. Traders are advised to do their own research and testing to determine the validity of a trading idea. Past performance does not guarantee future results; historical testing may not reflect a system’s behavior in real-time trading.

30


&OPTIONSWatch

FUTURES

Lend me your ear

The Merc speaks out

T

he Chicago Mercantile Exchange certainly had plenty to say about becoming the first U.S. exchange to go public, but the SEC-mandated “quiet period” kept Merc officials mum in the weeks before and after the Dec. 6 IPO. Once the proverbial muzzle came off , the Merc wasted no time in telling the financial world about its long-term plans as a publicly traded company. “There are three key reasons we went public,” says Merc chairman Terry Duffy. “Being public will facilitate future access to capital markets, provide a stock currency for mergers and acquisitions, and create liquidity for our existing shareholders.” And, at least from a volume standpoint, the CME has never been more in demand. Trading volume in 2002 was a record 558 million contracts. That was 36 percent higher than the record set in 2001. Globex, the Merc’s electronic trading system, had record volume of 198 million contracts in 2002. Total open interest at the Merc is the most for any options or futures exchange in the world, and the Dec. 12 single-day record of 24.8 million open contracts represente d appro ximately $16 trillion in notional value. Volume in the E-Mini Nasdaq exceeded its old record by 67 percent; volume in the S&P E-Mini bested the former high by 194 percent. The Merc also opened a remote data facility in 2002 and reported record revenue and earnings. If stock value is any indication, the IPO has been a moderate success so far. The stock, which trades on the New York Stock Exchange under the ticker symbol CME, opened at 35 and had not traded below that level as of Feb. 12, when it was trading around 42.70. “As risk management continues to become an increasingly important component in the portfolios of institutions and individual investors around the world, we believe we have great potential to attract more customers,” Duffy says. The Merc’s recent success is tied to a number of factors. For starters, the E-

Mini contracts on the S&P have been an enormous success, as they have allowed small retail traders to partake in index futures trading. Overall, Globex has proven to be a nice complement to the Merc’s floor-based business. Not surprisingly, the Merc is interested in expanding its product line. The exchange won’t specifically say what new products might be on the way, but President James McNulty admits they are always looking in the interest rate area. The Merc also wants to expand Globex’s ability to trade its own products as well as increase its reach internationally. The Merc recently added a Globex hub in London, and Chairman Emeritus Leo Melamed sees China, South America and Eastern Europe as good places for future growth. However, it’s that expansion of Globex that creates a catch-22: To compete in today’s market environment, a significant electronic presence is required. However, the more reliant the Merc is on Globex, the greater the perception the exchange is trying to phase out the open-outcry part of its business. In 2002, about half of the Merc’s volume was electronic, up from only 12 percent in 1996. But McNulty denies the Merc wants to reduce or eliminate pit trading. In fact, he says the administration has no input on closing trading pits. Rather, the Merc gauges volume and open interest to determine whether a pit remains open. This is not to say the Merc plans to stand in the way of future progress. “We have put money into technology, and we will continue to do that and let customers make the decision as to which platform they want to choose,” says Duffy. Melamed points to currency trading as proof that the Merc’s strategy in balancing floor and electronic trading is working. “A few years back, everyone thought we had lost the FX market, and we cer-


tainly were losing market share,” Melamed says. “But it needed the electronics to move, and we provided that when we raised the level of efficiency in Globex. Almost the moment that happened, the FX market started to come back to us. “Our strategy is to provide both venues in the best fashion possible,” he says. “We spend all the money we need to build the best electronic system in the world, and we spend all the money we need to maintain the floor and keep it efficient and viable, and we let the market dictate which way it wants to go. It has worked extremely well.” The Eurodollar, the Merc’s most actively traded contract, does most of its volume in the open-outcry market. One of the Merc’s big initiatives for 2003 is the launch of EAGLE — an electronic way to trade Eurodollar spreads through Globex. “The Eurodollar pit is an incredibly efficient open-outcry mechanism for a very complex product,” McNulty says. “It’s that value added that has kept it as lively as it has. We also need to be able to provide that on a 23-hour-a-day basis (the pit only trades during market hours; Globex virtually around-the-clock), so it’s extremely important for us to have our Eagle release.” However, while many outsiders are curious to see how the increased reliance on electronic trading will eventually play out, Duffy says it’s not a concern to the people who matter most — the members of the Merc. “I talk to them all the time,” Duffy says. “And what’s not on their agenda are questions about how long the floor will be open. The members are more about innovation and products. I always have members come up to me with different ideas on products and innovation and what they think will be successful in trading.”Ý 31

And here comes linkage!

&OPTIONS Watch continued

FUTURES

“Down” year for options still very busy

F

Making the CONNECTion

Euro(N)ext for CBOT

T

he Chicago Board of Trade is saying good-bye to one European exchange and hello to that exchange’s main com-

petitor. The CBOT dumped Eurex as its electronic platform provider and agreed to a deal with EuronextLIFFE. Under the terms of the existing contract, Eurex’s a/c/e platform will be used by the Board of Trade until Dec. 31, 2003. When trading begins in 2004, LIFFE CONNECT

will be the CBOT’s new electronic trading platform. “Our selection will bring enormous benefits to our customers and will allow the Board of Trade to continue to off er markets of the highest integrity, relia bility, liquidity and flexibility for years to come,” says CBOT chairman Nick Neu bauer. “Our board reached a decision that reflects our long-term business objective of continued independence and growth.” The CBOT’s decision came a day after Eurex announced plans to launch a U.S. based exchange (see “Eurex comes to the U.S.,” p. 48). CBOT president Bernard Dan says Eurex’s decision had nothing to do with the selection of Euronext, and he expects the remaining months with Eurex to go smoothly. 32

The original agreement between the CBOT and Eurex called for Eurex to provide electronic trading capability until October 2003, with the Board of Trade retaining exclusive rights to trade certain futures contracts. The agreement was modified in July 2002 to shorten the length of the technology agreement and allow Eurex to offer contracts similar to those trading on the CBOT. Dan doesn’t expect any difficulty in switching over to the new platform. “We anticipate a seamless transition, particularly as the majority of our current electronic trading volume comes from customers and members who are already established users of the LIFFE CONNECT system and are familiar with the functionality and benefits offered by that platform,” he says. Neubauer says that he expects CBOT volume — which set a record with more than 343 million contracts traded in 2002 — to increase even more in 2003 because of Euronext’s distribution range. Dan adds, “We’re looking forward to significant enhancements to our interproduct and intraproduct spread trading and to considerably more flexibility in the matching algorithms. We’ll have enhanced product development and product maintenance capability, which will translate into uniquely attractive new product offerings.” Dan says there is nothing specific on the new product front, but the focus will be on fixed-income products. And, he maintains the selection of LIFFE CONNECT in no way endangers the CBOT’s floor-based business. “A significant portion of financial futures are already traded electronically,” he says. “LIFFE will increase capa bility and functionality to ensure we remain competitive, not increase our electronic presence. It will enhance our ability to align our floor and electronic trading.”Ý

or the first time in 11 years, total and equity options volume failed to set a new record. However, there’s no reason to shed tears for the options industry. Equity option volume for 2002 was just less than 710 million contracts — about 13 million less than the record set in 2001, and the second-highest total ever. Equity option volume also was the second highest ever; at 780.5 million it missed breaking another record by less than 1 million contracts. Ind ividually, the International Securities Exchange continued to be the story. Total volume at the ISE increased 133 percent from the year before to 152 million contracts. The Chicago Board Options Exchange continues to be the volume leader among the five U.S.-based options exchanges. Total volume at the CBOE was more than 267 million, the thirdhighest total in exchange history. However, that represented a 13-percent drop from 2001. Equity options volume at the CBOE was down 26 percent to 173 million — a number directly attributable to the ISE, as the ISE trades only equity options. In other options news, the long-awaited linkage between the five U.S. option exchanges (ISE, CBOE, American Stock Exchange, Pacific Coast Exchange, Philadelphia Stock Exchange) officially began Jan. 31. The linkage, which has been more than three years in the making, will begin with a batch of 14 actively traded options. At press time, all options were scheduled to be linked by March 4. The options linkage is designed to prevent “trade-throughs” — an option trade executed on one exchange even though a better price can be found on another exchange. Without linkage, this has been a common occurrence. With linkage, option market makers will have the ability to access quotes from the other exchanges and execute trades at the best price. Ý


&OPTIONS Watch continued

FUTURES

Won’t they need a name change?

Eurex comes to the U.S.

F

utures exchanges in the U.S. will rates, indices and equities be getting a new competitor in early 2004. Although trading in U.S. soon — and one that comes with bonds has long been the a long track record of success. Eurex, the German derivatives exclusive domain of the exchange that leads the world in futures Chicago Board of Trade, volume, announced its intentions to Eurex gained the right to offer those begin trading derivatives on U.S. interest products by forming an alliance with the

• • • • • • • • FAST FILLS HAVING MORE OPTIONS q The International Securities Exchange signed a licensing agreement with Standard & Poors in mid-January, allowing the ISE to list options on nine Select Sector SPDR exchangetraded funds. Select Sector SPDRs are comprised of the stocks that make up various S&P sectors. The options will be on the Consumer Discretionary Select Sector SPDR fund (XLY), Consumer Staples Select Sector SPDR fund (XLP), Energy Select Sector SPDR fund (XLE), Financial Select Sector SPDR fund (XLF), Health Care Select Sector SPDR fund (XLV), Industrial Select Sector SPDR fund (XLI), Materials Select Sector SPDR fund (XLB), Technology Select Sector SPDR fund (XLK) and Utilities Select Sector SPDR fund (XLU). Earlier in the month, the ISE began trading options on the Financial Select SPDR fund (XLF), Technology Select Sector SPDR Fund (XLK) and Utilities Select Sector SPDR Fund (XLU).

WELL FED q The Chicago Board of Trade began trading options on Fed

Funds in mid-March. The Fed Funds futures contract often reaches record volume on days of Federal Open Market Committee meetings, prompting the CBOT to introduce an option on the contract. Fed Funds options will be traded in an open-outcry pit.

TAKE A DIP IN THE INCENTIVE POOL Eurodollars are already the most traded contract at the Chicago Mercantile Exchange, but the Merc has launched a new incentive program to promote liquidity in the back months of the contract. The back months are contract months that trade three to 10 years into the future, and the Merc will create an incentive pool that provides rebates to those accounts that meet minimum volume thresholds in the contracts. Additionally, the Merc is increasing the size of the pit that trades back-month Eurodollar contracts. q


CBOT. Although the alliance will end at the end of 2003 (see “Euro(N)ext for CBOT,”p. 46), Eurex swapped its electronic trading platform (a/c/e) for the right to list products already traded by the CBOT. Leaders of the top U.S. future s exchanges were quick to downplay the significance of the Eurex announcement. “We are aware of the competition,” says Chicago Board of Trade president Bernard Dan. “Eurex has several steps they need to make to get established in the U.S. We don’t view them as a particular threat.” BrokerTec, an all-electronic exchange that traded fixed-income products, was created as an alternative to the CBOT. Although BrokerTec traded more than 2 million contracts its first full year in business, it failed to steal substantial business from the CBOT and was recently bought by London-based brokerage ICAP. “We have seen competition before,” says CBOT chairman Nick Neubauer. “We have seen how difficult it is to compete with an established exchange. It’s even more difficult when that exchange is not local and is based in another country. As long as we do our business properly and provide the best value we can to our customers, we’ll be successful.” Because of Eurex’s familiarity with the CBOT (the a/c/e platform handles more than 70 percent of the global bond trade), it’s likely CBOT products will be Eurex’s initial focus. However, the Chicago Mercantile Exchange, a big provider of interest-rate products, will also see some direct competition from Eurex. “We have many excellent competitors around the world,” says Merc president James McNulty. “We live with that on a daily basis and we’re happy to continue to.”Ý 33

&OPTIONSStrategies

FUTURES

The RSI TRENDLINE method Although it is most commonly used as an overbought-oversold indicator, the relative strength index (RSI) also functions as an accurate trend indicator. Applying trendlines to the RSI and trading in the direction of the trend results in many favorable entry points.

T

BY THOM HARTLE

raders who have used a certain technical indicator for a long time tend to become so comfortable with a particular interpretation of the tool they have trouble conceiving diff erent uses for it.

FIGURE 1 TRADING IN AN UPTREND When the RSI closes above the descending trendline drawn between two pivot highs, a long trade is entered. This trade was exited on the fourth day after entry, when price traded below the low of the previous day. 30-year T-bond futures (US), daily

101-00

4 3

100-00

Long

1

2

Exit

99-00

98-00

97-00

RSI

A 28 Source: CQGNet

34

3 June

B

C 10

17

70 65 60 55 50

For example, the relative stre ngth index (RSI) is almost always discussed in terms of signaling short-term over bought or oversold conditions. However, “The Trend-Following RSI” (Active Trader January 2003, p. 44) illustrated how the 14-day relative strength index (RSI) can be used as a de facto trend indicator: RSI readings greater than 50 were shown to be equivalent to the market being above a 28-day exponential moving average, while RSI readings below 50 were the same as the market being below a 28-day exponential moving average. In addition, uptrends were shown to be characterized by persistent overbought RSI readings, and downtrends by persistent oversold readings. Because of its trend-following aspect, a simple trendline technique can be applied to the RSI to generate entry signals in the direction of the trend. To do this, it is necessary to plot the indicator as a histogram (a series of vertical bars) with the baseline set to 50, as shown in Figure 1). For an explanation of the RSI, see “The relative strength index” (opposite page).

RSI trendline rules The following rules describe how to use a trendline technique on the 14-bar RSI (daily charts): Long trades Setup: 1. Find markets where the RSI is greater than 50 and forms two pivot highs, the first higher than the second.


FIGURE 2 SELLING SHORT With the market in a downtrend, a trendline is drawn between two ascend ing RSI pivot lows. The trade is entered when the RSI penetrates the trend line to the downside. Nasdaq 100 E-Mini futures (NQ), daily Short

1050.00

Exit

1000.00

7

(Apivot high is a high with a lower high preceding and following it; a pivot low is a low with a higher low preceding and following it.) 2. Draw a downward-sloping trendline connecting the two pivot highs. 3. The RSI must close above the trendline. Entry: 1. If price closes in the upper 50 percent of the range for the day (which confirms rising momentum), go long on the close. Stop-loss and exit: 1. Set an initial stop two ticks below the low of the entry bar. Then trail a stop below the low of the previous day for the remainder of the trade’s life. Reverse the rules for short trades

Trade examples Figure 1 is a classic long trade in T-bond futures (US). The market is in an uptrend, as indicated by the RSI readings above 50. Point A of the RSI histogram is the first pivot high — the RSI value on that day is higher than the previous and following day. Point B is a lower pivot high, so a downward trendline is drawn to connect points A and B. The RSI closes above the trendline at point C, and price closes at its high for the day, so a long trade is entered on the close at 99 12 ⁄ 32. For the first full day of the trade, the stop-loss is placed under the low of the entry bar. On the second day, the stop was moved to under the low of the first day, and so on. The trade was exited on day 4, when the trailing stop (set under the low of day 3) was hit at 100 20 ⁄ 32.

1 950.00

2

6 3 900.00

4

5

50 45

RSI

B

40 35

C

A 1 July

8

30 15

22

29

Source: CQGNet

The relative strength index (RSI) The relative strength index (RSI) is an oscillator that ranges from 0 to 100. The formula is:

RSI = 100 - (100/[1+RS]) where

RS = relative strength = the average of the up closes over the calculation period (e.g., 10 bars, 14 bars) divided by the average of the down closes over the calculation period. For example, when calculating a 10-day RSI, if six days closed higher than the previous closes, you would subtract the previous close from the current close for these days, sum the differences, and divide the result by 10 to get the up-close average. (The sum is divided by the total number of days in the lookback period and not the number of up-closing days.) For the four days that closed lower than the previous day’s close, you would subtract the current close from the previous close, sum these differences, and divide by 10 to get the down-close average. If the up-close average was .8 and the down-close average was .4, the relative strength over this period would be 2. The resulting RSI would be 100 - (100/[1+2]) = 100 - 33.3 = 66.67.


35

&OPTIONS Strategies continued

FUTURES

Uptrends are characterized by persistent overbought RSI readings, and downtrends are reflected by persistent oversold readings. Figure 2 is an example of a short sale in the Nasdaq 100 E-Mini (NQ) futures contract. The market is in a downtrend — the RSI readings are all below 50. The bars preceding and following point A have higher RSI readings, establishing it as a pivot low.

Drawing appropriate trendlines Figures 1 and 2 show fairly straightforward trade examples. However, in some instances it might be less clear which RSI pivot highs or lows should be connected to form the correct trendline for a trade.

FIGURE 3 CONNECTING THE PROPER POINTS The trendlines drawn between points A and B and B and C resulted in a loss and no signal, respectively. However, the trendline connecting points A and C best reflected the renewed momentum and provided the basis for a success ful trade. 30-year T-bond futures (US), daily

111-00

110-00

Long

109-00

1

2 3

RSI A

19

C

26

Source: CQGNet

Point B is a higher pivot low, s o we draw an upward trendline connecting the two points. At point C, the RSI closes below the trendline, and price closes right at the low of the day, so a short position is established at the close at a price of 1005. After setting the initial stop, the trailing stop is implemented for the remaining days of the trade. The position is stopped out at 929 on day 7, when the market traded above the high of day 6. 36

107-00

106-00

Exit

B

108-00

75 70 65 60 55

downside penetration of the RSI up trendline. That being the case, the goal is to draw trendlines on the RSI that signal momentum is returning sooner rather than later. Figure 3 illustrates a situation in which more than one trendline could be drawn. The trendline drawn along points A and B on the RSI histogram generated a buy signal on bar 1. This turned out to be a losing trade, stopped out two bars later. Another pivot high formed at point C, allowing a trendline to points B and C. The RSI is above this trendline at bar 2, but price closes below the 50 percent point of the day’s range on this bar, so no entry signal occurs. However, a down trendline connecting pivot highs A and C creates an entry signal on bar 3 (two bars after RSI bar C). A long trade entered here would turn out to be very profitable. Although bars B and C were pivot highs, the fact they were nearly horizontal indicates they were part of the sideways trading action. Because we are trying to enter trades as soon as there is an indication momentum is returning, the RSI pivot highs or lows that should be connected are the ones that result in the steepest possible trendline. A steep trendline will be penetrated earlier than one connecting two similar pivot points.

From momentum to trend

The RSI is a momentum indicator based on calculating the difference between 3 9 September closing prices over a given period. If a market is moving from a trending phase to a congestion phase, the RSI levels will move closer to 50, which is the indicaWhen a market is trending upward, tor’s midpoint. This is a sign that the RSI will peak, and then edge its way momentum is decreasing. back to 50 — forming declining pivot The RSI trendline method alerts you to highs — in a consolidation period. The expanding momentum — exactly what a object is to identify when a trend is just trader wants, because it defines an about to begin; we look for a move above emerging trend. Ideas to improve this the RSI down trendline to signal upward approach include back-testing to see if a momentum is reasserting itself. The con- better stop-loss point or preset profit tarcept is the same for a downtrending mar- get becomes apparent.Ý ket, except we look for signs downward momentum is returning by virtue of a For information on the author see p. 3. www.activetradermag.com • April 2003 • ACTIVE TRADER

FUTURES

Trading System Lab

Trend with pattern entry (TPE)

FIGURE 1 PORTFOLIO EQUITY CURVE The light green area shows how much capital is invested at any given time. The dark green area shows how much cash is available. The top border of the light green area is the combined equity curve of the short and long trades.

Markets: Futures. System concept: System designers must confront two trading realities: Complex or “innovative” strategy ideas often sound good but fail to stand the test of time; and, most strategies that are effective on stocks do not work on futures. This test shows the performance of the Trend with Pattern Entry (TPE) system, a straightforward strategy that combines a trend-following rule with a simple pattern to pinpoint trade entries. If a market is in an uptrend, the system waits for a brief pull back and enters long. The system gives trades plenty of breathing room and the chance to reach the profit target (see the stock Trading System Lab on p. 44 for more details). Unlike the stock version of the system, the futures system goes short as well as long.

700,000 650,000 600,000 550,000 500,000 450,000 400,000 350,000 300,000 250,000

Rules: For long trades 1. Trend detection rule: The closing price must be above the 100-bar moving average. 2. Pullback entry rule: When two consecutive bars close below the previous bar’s close, enter long on the next bar’s open. 3. Stop rule: The stop-loss level is equal to the closing price minus four times the 10-bar average true range (ATR). Keep this stop-loss order active for the duration of the trade. 4. Exit rule: Exit at the market open when the trade closes with a 5-percent profit.

Invert the rules to execute short trades. Money management: Risk 2 percent of current account equity per position. This is done by calculating the difference between the position’s “basis price” (the closing price when the signal was trig-

FIGURE 2 SAMPLE TRADES

200,000 150,000 100,000 50,000 0

8 /1 6/ 93

7 /1 /9 4

Equity

4,000 3,900

Short

Short

3,400 3,300 3,100

Cover

Cover Cover

2,700

Cover

Nasdaq 100 index futures (ND), daily

2,600 2,500 2,400 40,000

Volume

20,000

November 2000 Source for all figures: Wealth-Lab Inc. (www.wealth-lab.com)

December 2000


Short

1 2/ 5/ 00

12 /4 /0 1 1 2/ 3/ 02

Long

Starting equity: $100,000. Deduct $10 slippage/commission per contract (entry and exit).

Test period: August 1993 until November 2002.

3,500

3,200

Cover

Linear reg

1 /3 /0 0

2,900 Short 2,800

3,600

Cover

3 /2 /9 8

3,000

3,700

Short

Cash

4 /1 /9 7

Test data: The system was tested on the Active Trader Standard Futures Portfolio, which contains the following 20 futures contracts: DAX30 (AX), corn (C), crude oil (CL), German Bund (DT), Euro dollar (ED), Euro Forex (FX), gold (GC), copper (HG), Japanese yen (JY), coffee (KC), live cattle (LC), lean hogs (LH), Nasdaq 100 (ND), natural gas (NG), soybeans (S), sugar (SB), silver (SI), S&P 500 (SP) and T-notes 10 year (TA). For this article we used ratio-adjusted data from Pinnacle Data Corp.

3,800

Short

5 /1 /9 6

gered) and the stop-loss level. Buy the number of contracts that results in a maximum loss of 2 percent of account equity when the stop-loss is hit. For example, given a current equity level of $1.2 million, a basis price of $.008750 in the Japanese yen (JY) futures, and a stop-loss price of $.008450, the trade’s maximum point loss is $.000300. The dollar risk is the point risk times the contract size ($.000300*[12,500,000] = $3,750). Two percent of the current equity is $24,000; dividing this amount by the dollar risk gives the number of contracts to buy: $24,000/$3,750 = 6 (rounded).

The system improved its entry points in the Nasdaq downtrend. Also note that the stop levels (green dots) on the chart stay constant until a new position is entered. Short

6 /1 /9 5

Test results: The system produced a total net profit of 572 percent with a total market exposure of only 38 percent, which gives you the financial leeway to use the system on other markets, or to diversify with other systems. The system’s biggest drawdown was 17 percent (on June 6, 2000). Most futures systems have a low percentage of winning trades, which makes it difficult to stick to the system over time. The TPE futures 37

FIGURE 3 DRAWDOWN CURVE

system has a very high winning percentage, combined with a low drawdown. Figure 1 shows the equity curves for both the long trades (black line) and the short trades (red line) that are responsible for the smooth overall equity curve. Although the short trades experienced a hefty drawdown of over 47 perc ent in recent months, the long trades were able to offset this.

A fairly long drawdown (more than one year) occurred at the beginning of the test period. However, the depth of the system’s drawdowns were modest. 0.00% -2.00% -4.00% -6.00%

-8.00% Practical trading issues: Moving-average-based systems very often wipe out a large amount of trading capital in side-10.00% ways markets. If you wait until a market has crossed above -12.00% or below a moving average (either by a certain percentage -14.00% move or for a certain number of days) before acting on a trade signal, you end up buying the highs and selling the -16.00% lows. The TPE system avoids this problem and improves the 8/16/93 7/1/94 6/1/95 5/1/96 4/1/97 3/2/98 1/3/00 12/8/00 12/7/01 entry point by taking a position when the market retraces after a move above or below the moving average. TABLE 1 PARAMETER STABILITY Because the original stop level is not a trailing stop, the This table compares system results using different moving average lengths. system uses a profit target to lock i n some profits and waits The system results are stable across the range of parameters, even when for the pattern to trigger a new signal before establishing a comparing long and short trades (not shown). new trade. Also, with each new position you adjust the position size based on the recent market volatility and the resultSMA 100 SMA 75 SMA 50 SMA 25 ing ATR calculation. This happens only if the trend defined Starting capital $100,000 $100,000 $100,000 $100,000 by the moving average is still moving in the same direction Ending capital $672,634.81 $601,189.69 $559,724.00 $470,290.13 it was before. Net profit % 572.63 501.19 459.72 370.29 The tested parameters were not the result of optimization. Exposure % 38.34 38.41 39.61 37.98 To verify the results, you can test the system on different All trades 834 823 787 807 markets and see whether the results are consistently profWinning trades 540 (64.75%) 527 (64.03%) 500 (63.53%) 499 (61.83%) itable across different parameter values. Table 1 shows the Max. consecutive 12 13 13 12 results of a quick optimization of the moving average periLosing trades 294 (35.25%) 296 (35.97%) 287 (36.47%) 308 (38.17%) od from 25 to 100 in steps of 25, using the same profit target, Max. consecutive 5 6 7 7 stop-loss and money management rules. Max. drawdown -16.57% -15.79% -21.46% -28.69% The system performs much better on commodity futures than it does on stocks because commodities generally have longer trends and less volatility than equity markets. As a and ride more profitable trends in futures markets. Another factor result, trend-following systems such as this one can usually catch is the futures version uses long and short signals, while the stocks version uses long signals only.

STRATEGY SUMMARY Profitability Net profit ($): 572,634 Net profit (%): 572.63 Exposure (%): 38.34 Profit factor: 1.34 Payoff ratio: 0.73 Recovery factor: 6.85

Trade statistics No. trades: Win/loss (%): Avg. gain/loss (%): Avg. holding time: Avg. gain (winners) Avg. hold time (winners):

834 64.75 0.98 29.39 6.00% 27.20

Drawdown

Avg. loss (losers):

-8.25%

Avg. hold time (losers): Max. consec. win/loss:

33.39 12/5

Max. DD (%): Longest flat days:

16.57 341

Bottom line: All traders need to imprint their own personalities and styles to a trading method. For traders who like to diversify among markets, the TPE strategy provides a good starting point. Managing your risk at the portfolio level offers exciting opportunities for futures traders. — Compiled by Dion Kurczek and Volker Knapp of Wealth-Lab Inc.

PERIODIC RETURNS Avg. Sharpe Best Worst return ratio return return

Weekly Monthly Quarterly Annually

0.42% 1.29 1.80% 1.37 5.47% 1.40 22.19% 1.22

9.53% 19.43% 26.88% 57.66%

-9.42% -7.83% -10.53% -1.53%

% Max. Max. Profitable consec. consec. periods profitable unprofitable

60.70% 68.14% 73.68% 90.00%

10 10 6 7

5 3 2 1

LEGEND: Net profit — profit at end of test period, less commission • Exposure — the area of the equity curve exposed to long or short positions, as opposed to cash • Profit factor — gross profit divided by gross loss • Payoff LEGEND: Avg. return — the average percentage for the period • Sharpe ratio — average profit of winning trades divided by average loss of losing ratio — average return divided by standard deviation of returns (annualized) trades • Recovery factor — net profit divided by max. drawdown • Max • Best return — best return for the period • Worst return — worst return DD% — largest percentage decline in equity • Longest flat days — longest for the period • % Profitable periods — the percentage of periods that were period, in days, the system is between two equity highs • No. trades — num - profitable • Max. consec. profitable — the largest number of consecutive ber of trades generated by the system • Win/Loss% — the percentage of profitable periods • Max. consec. unprofitable — the largest number of trades that were profitable • Avg. gain — the average profit for all trades • consecutive unprofitable periods Avg. hold time — the average holding period for all trades • Avg. gain (winners) — the average profit for winning trades •Avg. hold time (winTrading System Lab strategies are tested on a portfolio basis (unless ners) — the average holding time for winning trades •Avg. loss (losers) — otherwise noted) using Wealth-Lab Inc.’s testing platform. the average loss for losing trades • Avg. hold time (losers) — the average If you have a system you’d like to see tested, please send the tradholding time for losing trades • Max. consec. win/loss — the maximum ing and money-management rules to [email protected]. number of consecutive winning and losing trades Disclaimer: The Trading System Lab is intended for educational purposes only to provide a perspective on different market concepts. It is not meant to recommend or promote any trading system or approach. Traders are advised to do their own research and testing to determine the validity of a trading idea. Past performance does not guarantee future results; historical testing may not reflect a system’s behavior in real-time trading.

38


FUTURES

Legend:

Snapshot

Sym: Ticker symbol. Exch: Exchange on which the contract is traded.

Date: Feb. 11, 2003 The following table summarizes the trading activity in the most actively traded futures contracts. The indicator readings are NOT trade signals. They are intended only to provide a brief synopsis of each market’s liquidity, direction, and levels of momentum and volatility. See the legend (right) for explanations of the different fields.

Vol: 30-day average daily volume, in thousands.

Market

STT: Short-term trend direction, determined by whether a short-term moving average (MA) is above/below its value one month ago.

Sym S&P 500 E-Mini ES Nasdaq 100 E-Mini NQ ED E ur od ollar CL Crude oil TY 10-y r.T-note SP S&P 500 index US 30-y r. T-bond NG Natural Gas GC Gold FV 5-y r. T-note Cor n C EC E ur ocur re ncy HO Heating oil S Soy beans YM Mini Dow HU Unleaded gasoline JY Japanese yen Dow Jones Ind. Avg. DJ ND Nasdaq 100 index SB Sugar W W heat SI Silv er CD Canadian dollar SM Soybean meal SF Swiss franc BO Soybean oil HG Co p per Russell 2000 E-Mini MR LC Live cattle BP British pound TU 2-y r. T-note KC Co ff ee ZG Mexican peso CT Cotton CC Cocoa LH Lean hogs AD Aussie dollar EM LI BOR RL Russell 2000 index FF Fed Funds NK Nikkei 225 index OJ Orange juice

Exch Vol CME 622.30 CME 231.30 CME 94.90 NYMEX 88.70 CBOT 68.10 CME 65.00 CBOT 40.10 NYMEX 40.00 NYMEX 36.10 CBOT 34.10 CBOT 33.70 CME 29.30 NYMEX 27.70 CBOT 27.30 CBOT 24.70 NYMEX 23.80 CME 19.50 CBOT 18.40 CME 15.80 NYBT 15.20 CBOT 12.80 NYMEX 12.60 CME 12.30 CBOT 11.70 CME 11.50 CBOT 10.30 NYMEX 8.70 CME 7.70 CME 7.50 CME 6.80 CBOT 6.30 NYBT 6.30 CME 5.70 NYBT 5.60 NYBT 5.50 CME 3.70 CME 3.50 CME 3.00 CME 2.20 CBOT 1.90 CME 1.40 NYBT 1.30

OI 0.89 205.31 729.47 149.85 753.47 575.12 403.00 57.19 108.15 668.00 206.20 101.00 42.28 70.49 15.53 36.89 87.56 27.84 72.73 124.99 53.60 64.68 88.09 37.66 57.62 54.27 50.73 17.93 46.73 36.50 108.15 32.82 30.69 42.23 32.59 15.44 45.03 12.90 24.65 53.70 15.60 16.02

STT ITT LTT q

p

q

q

p

q

q

p

p

p

p

p

p

p

p

q

p

q

p

p

p

p

p

p

p

p

p

p

p

p

q

q

p

p

p

p

p

p

p

q

p

p

q

p

q

p

p

p

p

p

p

q

p

q

q

p

q

p

p

p

q

q

p

p

p

p

p

p

p

p

q

p

p

p

p

q

p

p

p

p

p

q

p

q

p

p

p

p

p

p

p

p

p

p

p

q

q

q

q

p

p

p

p

p

p

q

p

q

p

p

p

p

p

p

q

p

q

q

p

p

q

q

q

q

q

p

Trend% 3.50/1.00 6.00/0.75 1.13/0.00 1.45/0.32 1.86/0.14 3.19/1.00 1.11/0.25 1.85/0.44 2.18/0.00 2.33/0.22 2.00/0.12 2.07/0.22 2.20/0.36 0.74/0.36 2.08/1.00 2.47/0.78 1.30/0.46 3.27/1.00 5.00/0.69 4.00/1.00 2.43/0.53 2.20/0.19 2.25/0.51 2.09/0.44 3.17/0.47 2.35/0.47 1.52/0.47 1.90/1.00 2.08/0.05 2.27/0.66 2.17/0.10 1.36/0.25 5.00/0.83 2.23/0.81 1.68/0.75 1.69/0.75 2.82/0.44 0.79/0.00 4.64/1.00 1.32/0.58 2.40/0.07 1.10/0.71

Vltly% 0.91/0.92 0.85/0.37 0.70/0.56 1.08/0.44 0.92/0.24 0.90/0.88 0.96/0.32 1.09/0.36 1.44/0.88 0.89/0.32 0.80/0.39 1.17/0.98 1.59/1.00 0.97/0.86 0.90/0.86 1.15/0.83 0.93/0.31 0.89/0.86 0.84/0.36 1.17/0.86 0.81/0.32 1.29/0.95 1.09/0.98 1.05/0.97 1.15/0.98 0.96/0.59 1.17/0.98 0.92/0.83 1.10/0.58 1.05/0.69 0.84/0.36 0.93/0.81 1.08/0.78 1.14/0.80 1.15/0.58 0.79/0.92 1.00/0.61 0.59/0.07 0.88/0.81 2.09/0.81 0.82/0.22 1.04/0.53

OB/OS N N N OB N N N N OS N N N N N N N N N N N N OS N OB N N N N N N N N N N OS N N OS N N N N

This information is for educational purposes only. Active Trader provides this data in good faith, but it cannot guarantee its accuracy or timeliness. Active Trader assumes no responsibility for the use of this information. Active Trader does not recommend buying or selling any market, nor does it solicit orders to buy or sell any market. There is a high level of risk in trading, especially for traders who use leverage. The reader assumes all responsibility for his or her actions in the market.


OI: Open interest, in thousands.

ITT: Intermediate-term trend direction, determined by whether an intermediate-term MA is above/below its value three months ago. LTT: Long-term trend direction, determined by whether a long-term MA is above or below its value nine months ago. Trend%: The first number in this field is a measure of relative trend strength (NOT direction) expressed as the current short-term trend strength divided by the intermediate-term trend strength. The higher the number, the higher the current short-term relative trend strength (up or down). The second number shows the percentile rank of the current reading compared to the readings of the past two months. (In other words, a reading of .09, or 9%, means only 9 percent of the readings over this period were lower than the current reading.) Vlty%: The first number in this field measures relative volatility by dividing the current short-term volatility by the intermediate-term volatility. The higher the number, the higher the current short-term relative volatility. The second number shows the percentile rank of the current reading compared to the readings of the past two months. (In other words, a reading of .75, or 75%, means 75 percent of the readings over this period were lower than the current reading.) OB/OS: Whether a 10-day momentum indicator registers the market as overbought (OB), oversold (OS) or neutral (N). Note: Overbought and oversold signals are NOT trade signals. They are warnings that upside momentum is high or low (compared to the market’s recent activity) AND MAY REMAIN SO FOR AN UNDETERMINED AMOUNT OF TIME.

39

Active TRADER Interview

BY MARK ETZKORN

John Bollinger on market cycles, overlooked opportunities and why a “consolidating” market may be just what the doctor ordered for a new breed of swing trader.

I

n the markets, what goes around comes around. You just have to wait a while sometimes. When Active Trader last spoke with analyst and money manager John Bollinger in late 2000, there was still lively debate in the financial community as to whether the sell-off the market had suffered that year was simply a nasty correction or an actual bear market. Given U.S. traders and investors had recently surfed the seemingly endless wave of an historic bull market, it’s not surprising many people were loathe to accept the idea the tsunami had finally reached the beach. Even many of those brave enough to call a bear a bear when they saw it were prone to positive spin, arguing a market turnaround was probably two or three quarters away. More than two years later, the market has long since delivered the coup de grace to the last of the big-picture bulls. For the most part, public speculation is now focused on signs of general economic strength and reassurance the worst is over in the market — not hope that a new, sustained surge in equities will occur any time soon. The market has indeed changed over the past three or so years, but as Bollinger, 52, points out, from a broader perspective, it has morphed back to what it was before the bull move started. (And for the younger readers out there, we’re talking 1982, not 1995.) 40

He’s not just making an academic observation. Bollinger’s take on the current market is especially interesting because he started in the trading industry right around the time the last prolonged bear-market consolidation was ending and the market was gearing up for the 1980s bull move. At the time, Bollinger says, long-term investors had essentially given up on the stock market after years of frustrating non-returns, while traders, who had adjusted to the up and down character of the market, were still in the fray — and making money. In January, Bollinger spoke with us about market cycles and what they portend for the future of U.S. stocks, and how traders may actually be more in the cat-bird’s seat than they think — if they understand the nature of the market and apply the appropriate trading techniques. AT: Things have changed a lot since the last time we talked. What can traders learn from what’s happened since 2000?

JB: I think it’s important to understand that all stock markets — not only ours — go through cycles of expansion and consolidation. And these are fairly long-term cycles we’re talking about. In our market the cycle is typically 16 years. The expansion www.activetradermag.com • April 2003 • ACTIVE TRADER

FIGURE 1 AROUND AND AROUND WE GO

after World War II, which started in 1950 and continued into 1966, was followed by a consolidation phase that ran from 1966 to 1982 (see Figure 1). Then there was another big expansion phase from 1982 to summer 1998, which is when the broad market topped out. So, we’re now four-and-ahalf years into a consolidation phase.

A long-term chart of the Dow Jones Industrial Average (INDU) shows a pattern of alternating trend and consolidation/bear phases roughly 16 years in length. However, the intermediate-term swings within the consolidation periods are often quite substantial.

AT: Only 12 years or so left to go! Well, you’ve just depressed everybody, John.

Dow Jones Industrial Average ($INDU), monthly 9,059 6,069 4,066 2,724 1,825 1,223 819 549 368 246

JB: (Laughs.) I don’t think 165 that should be the case! 111 When I came into the market in 1980 there were no 1940 1950 1960 1970 1980 1990 2000 long-term investors left — Source: TradeStation Platform by TradeStation Group they had all departed the market in frustration. The JB: Yes, there are huge moves. And that spells opportunity. Just people still active were the ones we called swing traders. The definition of a swing trader then was a little diff erent to use round numbers, the stock market topped out at roughly than it is now. Back then, it meant something a bit longer term. Dow 10,000 in summer of 1998. Just counting swings of 10 perSwing traders then used daily and weekly charts and were cent or more, the stock market has traveled 20,000 points since interested in taking intermediate swings out of the big stocks then. that lasted anywhere from 10 or 20 days to six months or even Of course, it’s virtually impossible to pick the very bottom a year. These were the people making money, and they were or top of a move. That means traders trying to take advantage doing quite well. of these swings will be late getting in and late getting out. For argument’s sake, let’s assume you’re 25 percent late — that is, AT: Can you give the broad strokes of how you perceive the it takes one quarter of a price move to occur before you recogrotation of expansion and consolidation cycles in the market? nize the direction of the market has changed. But that means What are their earmarks? you can still capture half the move, or 10,000 of the 20,000 JB: At the beginning of a consolidation phase everybody points that occurred over the last four and a half years — a believes in buy and hold because they’ve just enjoyed a very period in which most people lost money. long bullish period, during which the market was very hard to The important thing to understand right now is that we’re beat and you did really well if you simply bought the broad list not in a buy-and-hold environment, we’re in a trader’s enviof stocks and held on to them. ronment. The keys to success in a trader’s environment are At the end of a consolidation phase, all the buy-and-holders market timing and industry group and market sector rotation have been totally frustrated. They’ve made no money for years — which is simply to say, market timing on both a macro and and years, so they’ve gone elsewhere. At this point, the swing micro level. traders are the kings of the market — the only ones making any AT: In contrast to when you started in the business, traders money. Market timing, which did not pay at all during the expan- now have access to different index and sector instruments — sion phase, pays tremendously well during the consolidation ETFs and HOLDRS, not to mention different stock index phase. Frankly, for the technical investor or trader, this is the futures and options. Does that make trading this kind of envi best of all possible worlds because there are big, fat, interme- ronment that much easier? diate swings to trade. Also, classic indicators such as the JB: I think the growth of these instruments has been a fantasadvance-decline line, new highs/new lows, the MACD, mov- tic development. I remember when the Value Line futures were ing averages, and overbought-oversold indicators work very the first index futures product launched in Kansas City (the well in this type of environment. Kansas City Board of Trade), and the S&P 500 futures came shortly thereafter, followed by OEX options. In a very short AT: The “swings” you’re talking about are very large, though, period of time we went from having no vehicles for market ar en’t they? If you look at, say, that 1966 to 1982 period, the timing to having several great vehicles for market timing. market was essentially in a trading range, but it was a very wide However, the interesting thing is that this was happening range and the major moves sometimes lasted more than a year. just when market timing was going to stop paying. The market ACTIVE TRADER • April 2003 • www.activetradermag.com

41

Indicator reference Advance-Decline line: A day-to-day running total of the number of stocks that have gained on the day minus the number of stocks that have fallen on the day. A version using the weekto-week figures can also be used as a longer-term indicator. The most commonly referenced A-D line is the one calculated on New York Stock Exchange (NYSE) stocks. Bollinger Bands: Price bands that are placed two standard deviations above and below a 20-bar moving average (see Figure 2 in main story). Moving average convergence-divergence (MACD) indicator: A hybrid trend-momentum indicator consisting of two lines: the first (the MACD line) is the difference between a 12- and 26-day exponential moving average (EMA); the second is a nineday EMA of the MACD line. On Balance Volume (OBV): A running sum of daily market volume weighted by whether the market closes up or down for the day. If today closed higher than yesterday, the OBV value is yesterday’s OBV + today’s volume; if today closed lower than yesterday, the OBV is yesterday’s OBV today’s volume; if today closed unchanged from yesterday, the OBV is the same as yesterday’s.

was just finishing a consolidation phase and entering a wild bull market, in which market timing offered very little added value. Today, by contrast, I think it’s a case of having the right instruments available at the right time. We have a plethora of tools to trade the markets: stock index futures of every type, stock and index options, and exchange-traded funds, which I think are a fantastic opportunity for investors and traders. Not only can you invest in the index of your choice, many indices are now even being brok en up into their growth components and value components, which allows you to make “style” bets within the indices. AT: With all these tools, do you see any advantage to trading individual equities in this kind of market? If you want to trade a sector or group, why bother building a portfolio of individual stocks and deal ing with things like shortselling restrictions? Sector and index derivatives would seem to simplify things.

JB: They do and they don’t, because traders are individuals. Some people are never going to be comfortable trading index and sector aggregate products. They’re creatures of the stock market, they’re interested in individual stocks, and they’re never going to change. Then there are the people on the other side of the equation who, frankly, will never trade an individual stock again. They’re going to trade these tools exclusively over the coming years. It’s really a mat ter of psychology — a matter of what you like, what suits you. AT: On a semi-related subject, do you think single-stock futures have a role to play in all this? JB: I’ve done quite a bit of research on them and I’m curious to 42

see how they’ll do. Europe has effectively had single-stock futures for a long time. In France, all stock transactions are conducted as forward transactions that are settled, I believe, in 30 days. Single-stock futures have worked in other situations, and other types of investors have found them viable. The question I have is whether U.S. investors are going to be interested in them. To work, they’re going to have to attract a large following and really develop volume and liquidity. American investors are very fickle, and you never know what they’ll take to. For instance, for years there were products called the Americus Trusts, which were traded on the AMEX. They were a form of exchange-traded fund — the predecessors of today’s iShares — and they languished. They were a little too complex for most people, and they just didn’t work. But on the other hand, iShares have gone great guns. Singlestock futures have the potential to be an important part of the puzzle, but we’ll have to wait and see. It’s very hard to figure out what the public will go for. AT: Let’s go over some specific ideas for trading the kind of consolidation environment you were describing before. Where do you suggest a trader conditioned to trade in a longterm bull trend get started? JB: In this type of market, the tools that work the best are relative tools — those that let you get a grip on what’s happening in relation to immediately prior history. Of course, different traders will have different ideas about what “immediately prior” means. For some it will mean what’s happened this morning, for others it will mean the past 20 days or the past six months or year. But the advantage of relative tools is that you can adapt them to your purposes. Bollinger Bands, for instance, provide a definition whether prices are high or low on a relative basis (see “Indicator ref erence,”). At the upper band prices are high, and at the lower band prices are low. If price tags the upper b and, you know prices are high, so you can consult another tool to determine whether you believe that “highness” is sustainable or whether it’s a potential reversal to be sold. One of the problems I’ve seen is that people treat Bollinger Bands in the simplest way possible. They automatically think a tag of the upper band is a sell and a tag of the lower band is a buy. Nothing could be further from the truth. Some tags of the upper and lower bands are sells and buys, respectively, but not all tags are action points. Price can, and does, walk up the upper band or walk down the lower band, and often when this happens you get some of the most profitable trades. With rare exceptions, it’s not enough to use Bollinger Bands alone. You have to combine them with something else that tells you about sustainability. For me, that something else is volume indicators. AT: Just indicators? Do you ever look at raw volume numbers? JB: Both can work. Some people are able to look at volume, relate it to the price bars and intuitively understand the supply-demand relationship. Other people need to parse volume into an indicator to clarify the picture. Older traders who grew up keeping charts by hand would probably be more comfortable with raw volume numbers. Traders who grew up with technology that could easily calcuwww.activetradermag.com • April 2003 • ACTIVE TRADER

late and plot complex indicators will likely be happier with volume indicators. AT: Can’t

volume be misleading, though? High volume can accompany reversal points or support trends, but it seems like many volume-watching traders conveniently overlook the fre quent occasions when volume gives “classic” signals and price does the opposite of what it’s supposed to. And you can also find plenty of turning points where volume wasn’t unusual one way or the other.

FIGURE 2 THE RELATIVE PICTURE In early and mid-January, Microsoft was “walking” up the upper Bollinger Band after bouncing off the lower band when it formed a “W” bottom. Below price is normalized volume, which is the 50-day average volume multiplied by 100 (the black horizontal line). Microsoft (MSFT), daily 57 55 53 51

JB: Well, first of all, I’m working with the relative definition of high and low price 49 levels, so that lets me know when to consult volume. 47 For example, if price has just tagged the upper band, I know this is a point at 45 which to see if volume is adding anything to the picture. I don’t scan volume continuously, trying to make an ongoing 160 stream of decisions based on the relationship between it and price. I only look at 110 critical junctures. 70 What I’ve found, in this regard, is that it pays to wait. In other words, after I get 10/30/02 11/11/02 11/21/02 12/4/02 12/16/02 12/27/02 1/9/03 a buy or sell signal, I wait for price action Source: www.equitytrader.com (Acme Analytics) to confirm that signal. If price tags the upper band and a volume indicator — say, 20-day on balance volume (OBV) — winners are two to three times the size of your losers, you’ll is in negative territory, you can treat that as a warning or alert find you’re making money pretty quickly. because the combination suggests this is a potentially unsusBy using this relative trading approach, you can address tainable situation. Then, if there’s evidence of a decline, you both those risk-reward dimensions. You address the size of the can act, because you have the knowledge a proper setup was in winners vs. losers by having entry points with logical [stop] place. points nearby that let you know your trade was wrong. You The other thing you have going for you in this kind of trade address the number of winners by finding the right volume situation is knowing whether there’s a good risk-reward rela- indicators to assess the type of trade and the vehicles you’re tionship. If price tags the upper band and turns down, you can using. place a stop just above the entry point, knowing if price goes back up and violates the stop, your setup is broken. And that AT: What about a trend component that’s independent from stop will be fairly close by, so the amount you’re risking is rel- what you’re discussing now — independent in that it would atively small, whereas the immediate target for the move is for probably be on a longer time frame? price to get back to the lower band, which is much farther JB: I think the idea of biasing your trading in the direction of away. the greatest probability of success is very important. In a sideways market, you’ll get fairly important intermediate-term buy AT: What kind of risk-reward numbers do you operate with? signals near the bottom of the range and sell signals near the JB: Here’s one way to look at it: There are only two ways to top. Those should absolutely dictate the direction of your tradimprove your trading performance. First, you can increase ing. Clearly, if you can bias your trading in favor of the interyour number of winning trades vs. your number of losing mediate swing direction of the market, you’ll go a long way trades. If you’re batting around .500, you can try to add differ- toward improving the two key components of success. ent timing information and indicators, and so on, and maybe In terms of time frame, if you’re using Bollinger Bands, for get your batting average up to around .600 or .650. I think example, rather than trying to adjust the time frame by changyou’re doing pretty well if you have 65 percent winners. ing the periods and width of the bands — 20 and 2 seem to Second, you can increase the size of your winners vs. the size work very well for most applications and are certainly a good of your losers. Say your winners are twice the size of your los- place to start — try using a different bar length. ers — that’s pretty good. If you get up to three times the size, I If you’re using daily bars and you want a shorter-term view, think you’ll find the mathematics work very much in your you might switch to hourly bars. If you want a longer-term view, favor. If you have 60 to 65 percent winning trades and your try switching to weekly bars. That’s a good way to get an idea of ACTIVE TRADER • April 2003 • www.activetradermag.com

43

what’s happening in diff erent time frames. AT: Can we pick out a chart and kick around a few of these ideas we’ve been discussing?

JB: What’s interesting about this market right now is that so many instruments have established interesting trading patterns. If you look at Microsoft (MSFT), you see a fairly classic picture (see Figure 2). The stock made a “W” bottom recently: The right side of the W was a bit lower than the left side. Both sides tagged the lower band, as you might expect, and then volume came in as the stock started rallying away from the right side of the W. The buy came on Jan. 2, the expansion day. Now (mid-January) the stock is making a nice little walk up the upper band. From my perspective, Microsoft has behaved exactly the way an intermediate-term swing trader would like:

It established a well-defined, easy-to-diagnose turning point, and then confirmation came in the form of volume expansion in the direction of the new rally.

level of the market. Aone-point move in a futures contract will have the same value whether that move is from 100 to 101 or 200 to 201. The gain-loss is always going to be the same for a given distance on the chart. But with stocks, you have to put up the value of your trade — or at least 50 percent of it — so a one-point move from 10 to 11 is quite different from a move from 100 to 101. The first is a 10-percent gain, the second is a 1-percent gain. Logarithmic charting shows those moves occupying the same physical space on the chart. Of course, I’d recommend using Bollinger Bands, which give you a relative definition of high and low price levels. Then, you can compare this to the action of a volume indicator, which I think you should also include. Use a volume “clip” — normalized volume, or at least a moving average of volume so you have some idea of whether volume is high or low on a relative basis. And this may sound less important, but it’s something that drives me nuts: Make the chart ( pause) pretty — make it something you can look at. I hate charts with black backgrounds and garish price-bar and indicator colors — they give me a headache. You’re going to spend a lot of time looking at a chart, so you want to make it as comfortable as possible to look at. You obviously don’t want to burn out your eyes or get a headache, but the more important aspect of this is that a chart is simply a tool to inform that fabulous pattern-recognition engine you were born with. You want to give it the best possi ble presentation to work with — the easiest to understand, the easiest to utilize. I’ve seen guys in their trading offices using nine monitors, with one garish display after another. I like a two-monitor setup. One for charts and the second for numerical things — portfolio information, spreadsheets and what not. After all, you can only look at one chart at a time.

AT: The volume is increasing there, but if you compare it to the last six months or so, it might be a little below average. Is that cause for concern?

JB: No, I don’t look at volume that way. I look at volume on a relative basis, as I do everything else. I divide volume by its 50day average and multiply by 100 to get “normalized volume.” Relative to that, the current volume is behaving very well indeed. When the market is up on the day, the volume bars are colored green; when it’s down, they’re red. Since that bottom, volume has tended to expand on up days and contract on down days.

AT: Working on the assumption the current market may be in a consolidation environment for an extended period, what time frame would you begin your analysis on?

JB: It depends upon what you’re trying to do. If you’re trying to get a handle on the market’s intermediate-term activity as background information, I think you can estimate the swings will be three to six months at a minimum and six months to a year maximum — that’s a typical pattern that’s occurred in the past. Weekly bars seem to be the appropriate way to visualize that information. When you get down to the shorter-term — actually executing trades in individual stocks or indices — I recommend daily AT: Some people tend to throw everything but the kitchen bars. That’s my bias. I’ve looked at charts for years and I’m sink onto a price chart. What do you think provides a solid comfortable in that time frame. For even shorter-term trades, foundation for effective chart analysis? hourly bars are quite useful. JB: Start with how you scale your chart. It’s very important to There’s obviously that subset of traders who are going to use logarithmic (percentage) scaling if you’re trading stocks. If trade within much tighter parameters — people who are using you’re trading futures, it’s very important to use arithmetic five-minute bars and tick charts. But the concepts really remain (regular) scaling. the same, regardless of the time frame: knowing what’s happening on the longer-term time frame so you can correctly bias AT: Why? your operations in the shorter time frame. JB: You want the depiction of the security to reflect its proper If you get a nice entry signal for a long trade and the market economic value within your portfolio. is in an upswing, you probably want to take that a little more Futures margin is generally the same regardless of the price seriously than you would a nice entry signal for a short. The 44


principles we’re talking about are fractal in nature — they exhibit the same kinds of patterns and characteristics at different levels of magnification, whether it’s 10 minutes and hourly, hourly and daily or daily and weekly. The same types of setups and trading patterns are evident. AT: It’s kind of surprising how many people don’t buy into that, because it seems pretty apparent if you just look at charts for a while. JB: I remember there was a fellow by the name of Sam Kachigan who designed a trading system called the Lennox system. One of the basic elements of the system was that trades had to be confirmed in three time frames. There was the longterm setup, then you looked for a similar setup on the intermediate time frame and, finally, the same thing on the shorter time frame, which is where you executed the trade. It’s the same old idea — having all the parts and pieces pulling in the same direction. AT: What about the different areas of the market being in sync?

JB: I really believe in the importance of industry groups and sectors, especially in a trading range market. It’s like a ro w boat: If you have four people rowing, one rower can be thought of as the stock, one as the industry group, one as the sector, and the last one as the overall market. When all four are pulling their oars in the same direction, the odds of success increase dramatically. But if two are paddling in one direction the other two in the opposite direction, they’ll go nowhere. It’s the same thing in the market: If you’re trying to short a stock in a weak group when the sector and the market are strong, the odds are against you. In this type of market, I’ve found that getting the group and the sector right is much more important than in a bull phase. During the bull market, getting the group and sector right may have added a couple of points to your performance. In this type of market environment, it’s the difference between negative and positive performance. When I started in 1980, certain things were clear. First, buy and hold didn’t work — it hadn’t worked for a very long time. If you bought a stock and prayed for the long-term trend to bail you out, you were dead. Second, relative strength worked, and worked well — you could buy strong stocks in a flat, choppy market and do well, no question about it. Nobody proved that more brilliantly at the time than George Chestnut at American Investors. It was his work that really got me focused. He understood relative strength wasn’t enough, and fully two-thirds of every issue of his [investment timing] service was devoted to analyzing industry groups and sectors. The final piece to the puzzle was market timing. You had to get your market timing right — period. Of course, though, after a 16-year bull market, everybody says, “No, no, no, market timing is terrible, you’re not supposed to do it — just buy and hold.” But now we’re back in market-timing mode again. It’s interesting. If you think about the environment that prevailed from 1966 to 1982, what was the big thing in terms of the investment advisory business? Newsletters. There were a ton of newsletters and investors subscribed to them. Why? Because they dispensed an invaluable service — market timing. ACTIVE TRADER • April 2003 • www.activetradermag.com

Almost all those newsletters died out between 1982 and 1998. People say, “Oh, it’s the Internet — nobody wants to buy a paper newsletter anymore.” That’s simply wrong. The newsletter business dried up the way it did because there was a huge bull market and investors, correctly, realized market timing wasn’t doing them any good! AT: Do you think this cycle is going to be any different from past cycles, like the 1970s, when many people abandoned the market altogether?

JB: That’s not going to be any different whatsoever. Some people will leave the market shrieking in frustration. It’s part of the puzzle, part of the development of a particular cycle. You can already see some of the other mechanisms start falling into place: The Bush administration proposes to end the double taxation of dividends, which makes dividends more attractive. Gradually, payout ratios will rise again and investors will get more interested in investing in stocks that are paying dividends, and over the y ears that will p lay into other aspects of the market that get long-term investors back into stocks. Eventually, we’ll have served our time in this consolidation and the base will be built for the next great bull market. When it starts running, there will be a bunch of naysayers shouting, “No, this isn’t for real, watch out!” and meanwhile the market will continue to go up and up and up. And when the last doubter is finally convinced, it will be time for the next consolidation! The three keys to success here are going to be relative strength — positive or negative; market timing, so you know which type of environment you’re operating in; and being right about groups and sectors, so you have the odds in your favor. That’s pretty much the agenda that prevailed in the ’70s and early ’80s, and it’s the agenda that’s going to prevail now. Ý

Bollinger resources John Bollinger, CFA, CMT, is the president and founder of Bollinger Capital Management. His book Bollinger on Bollinger Bands was published in 2002 by McGraw Hill. He has developed a new Web site devoted to industrygroup analytics, www.GroupPower.com, and is also the founder of the following sites: www.EquityTrader.com www.FundsTrader.com www.BollingerBands.com www.BollingerOnBollingerBands.com www.PatternPower.com

Additional reading: “Volume indicators revisited,” by John Bollinger, Active Trader , March 2002, p. 32. “John Bollinger: Focus on the markets,” by Mark Etzkorn, Active Trader , January 2001, p. 74.

45

The Face of TRADING

From pennies to patterns BY KIRA McCAFFREY BRECHT Name: Daryn Collier Age: 30 Lives and works in: Vancouver, British Columbia

D

aryn Collier has always had an independent streak. Over the last decade, he’s tried his hand at various forms of self-employment, including stints in the restaurant, landscaping and drycleaning businesses. After eventually tiring of those endeavors, Collier borrowed some money from a relative and turned his hobby of trading penny stocks into full-time trading. In his teens, Collier read One Up on Wall Street , written by Fidelity mutual fund legend Peter Lynch. He has been interested in the markets even since. “I got started in highly speculative penny stocks and basically just gave my money away,” Collier says. “I’d be at work at the dry cleaners and on the phone with the automated broker.” Living in Canada, Collier faced challenges in merely opening an account. Until recently, government regulations made it difficult for Canadians to open up an account with a U.S. discount brokerage firm. Collier was using a Canadian broker and paying prohibitively high commission rates for NYSE stock trades. However, a U.S.-based proprietary trading firm — Bright Trading — opened an office in Vancouver. Collier put up $25,000 and began trading full-time in May 2001. Trading method: When he first began, Collier traded the spread between two NYSE companies that were merging. “If you knew GE was acquiring Honeywell, offering 1.05 shares for each share of Honeywell, you’d take the last price traded on GE and multiply it by 1.05,” Collier says. “Subtract the last price of Honeywell and you have the spread. You could either be long or short.” At first, this method was successful for Collier, and it enabled him to build up his trading account. “There was good movement in the spreads,” Collier says. “It was easy to put them on and take them off — it was almost like free money.” However, the spreads eventually became tighter, and making money became more difficult. Undaunted, Collier switched to pair trading — buying one stock and selling a different stock Trading setup Hardware: PC with 1.3 MHz processor and 512 MB RAM. Five monitors: three 17-inch for charts, two 21-inch for executions. Software: TradeStation. Internet connection: Cable modem. Brokerage: RediPlus and DirectFutures.

46

in the same sector (i.e., buying Citibank and selling J.P. Morgan Chase). That well also ran dry after pairs ceased trading in predictable ranges, causing Collier to search for yet another trading method. “You have to know when to cut your losses and get out,” he says. “You have to be willing to look at what’s working and what’s not.” Collier currently trades stocks that are in short-term consolidation or flag patterns. He primarily uses intraday charts — one-, five- and 15-minute charts of big-cap, liquid NYSE stocks such as Dupont, Citigroup, J.P. Morgan Chase, Boeing, Johnson & Johnson, Home Depot and Wal-Mart. He uses three monitors just for charts, and he scans them for short-term bull or bear flags that are just forming. However, he doesn’t have specific rules for entry, noting, “It has to be done by feel.” “I’ll try to wait for a flag to [form] on the 15-minute chart and then look for an entry point on the five- or one-minute chart to keep my risk down,” Collier says. Collier enters the trade while the flag is forming, hoping to catch a breakout in the direction of the longer-term trend. He trades roughly 500 to 1,000 shares at a time. As far as exiting is concerned, he will take partial or full profits on a retest of the breakout point of the flag. If he keeps a portion of the position on, he moves his stop to breakeven. He completes six to eight trades a day, none lasting more than 20 minutes. Worst trading experience: Collier experienced his worst trade when he used his merger arbitrage strategy to trade GE and Honeywell. “It was the day bad news came out from European regulators and the completion of the merger was in question,” he says. At the worst point, Collier lost 20 percent of his trading account, although he cut his losses to 10 percent by the final bell. When not trading: Collier enjoys running and spending time with his family. The best thing about trading: “Trading is the only thing I’ve found that has a constant supply of challenges,” he says. “Every day is different.” Recommended reading: Street Smarts by Linda Bradford Raschke and Laurence Connors; Pit Bull by Marty Schwartz; Market Wizards by Jack Schwager; Linda Bradford Raschke’s Web site (www.lbrcapital.com). Ý www.activetradermag.com • April 2003 • ACTIVE TRADER

RISK Control and MONEY Management

Road to RUIN

O

bjectively, trading can be thought of in terms of probabilities: the probability a price move of a certain magnitude will follow a particular trade signal, the percentage of winning trades a strategy generates over time, and so on. Although it is human nature to think almost exclusively about the chances of making money, a more practical mindset is to appreciate the odds of losing your money by using a particular trading strategy or money management technique. “Take care of the losses first,” the trading axiom says, “and the winners will take care of themselves.” “Risk of ruin” is a concept sometimes used to determine the probability you will “blow out” using a particular trading approach (or alternately, the odds you will reach an equity goal before losing a certain percentage of your equity). However, risk-of-ruin calculations typically are based on simplifications and assumptions not possible or practical in real trading. It should come as no surprise that many of these concepts have been applied more commonly to blackjack and other games of chance. There are, though, important differences between a hand of blackjack and the infinitely more complex “game” of trading. We’ll look at what a few simple calculations can tell us about the probabilities of trading.

Risk of ruin is a mathematical estimate of the odds you will lose your money using a trading approach. But is this concept really helpful, or does it give traders a false impression of a trading system’s risk profile?

More than one road There are several ways to calculate risk of ruin, but you need to have at least three pieces of information to start: 1) the account size; 2) the financial goal, or target, for that account; and 3) the “ruin point.” “Ruin” in this context does not have to mean losing all your money. It could mean losing half your current account, 75 percent of it, 90 percent of it — whatever point would cause you to shut down your trading account or consider your trading plan a complete failure. The financial goal is similarly expressed in terms of the account size — two times the account size, three times the account size, etc. From this perspective, risk of ruin is the likelihood you will reach your ruin point before reaching your financial goal, or vice versa — e.g., what are the odds you’ll double your money before losing half of it. The following calculations illustrate the basic mathematics of estimating risk of ruin. Unless otherwise noted, they are based on the assumption the strategy in question is risking a predefined percentage of equity — e.g., 2 percent, 5 percent, etc. — on a trade.

Simple risk of ruin formula

BY ACTIVE TRADER STAFF ACTIVE TRADER • April 2003 • www.activetradermag.com

This formula is the most basic — and the most limited. It calculates the odds of hitting a ruin point before a financial goal when betting 100 percent of the predetermined optimal 47

equity risk amount. The formula assumes an equal likelihood of a winning bet and a losing bet, and that wins and losses are the same size. Probability [A reaches G*A before reaching R*A] = G*(R-1)/(R-G) where A = the Account size G = the multiple of the initial account size that results in

the financial Goal. R = the multiple of the initial account size that results in the Ruin point. For example, the probability of tripling the account size (G = 3) before losing half of it (R = .5) when betting an optimal amount is: 3*(.5 - 1)/(.5 - 3) = .6, or 60 percent. The probability of doubling the account size (G = 2) before losing half of it (R = .5) is: 2*(.5 - 1)/(.5 - 2) = .666, or 67 percent. Despite its limitations, this formula shows that the more modest a financial goal is, the higher the chances of reaching it — no surprise there.

Fractional risk of ruin This formula estimates the odds of reaching a financial goal before a ruin point when betting a certain fraction of the predetermined risk amount. Like before, the odds of winning or losing are assumed to be equal, and the winning trades are assumed to be the same size as losing trades. Probability [A reaches G*A before R*A] = [R((2/f) -1) - 1]/[(R/G)((2/f) -1) -1] where A = the Account size G = the multiple of the initial account size that results in

the financial Goal f = Fraction of the predetermined risk percentage R = the multiple of the initial account size that results in the Ruin point If risking 50 percent of the risk percentage, the probability of doubling the account before losing half of it is: [.5((2/.5) -1) – 1]/[(.5/2)((2/.5) -1) – 1] = (.53 – 1)/(.253 – 1) = (.125 – 1)/(.015625 – 1) = -.875/-.984375 = .89, or 89 percent. If risking 80 percent, the probability of doubling the account before losing half of it is: [.5((2/.80) -1) – 1]/[(.5/2) ((2/.80) -1) – 1] = (.51.5 – 1)/(.25 1.5 – 1) = .74, or 74 percent. If risking 10 percent, the probability of doubling the account before losing half of it is: 48

[.5((2/.10) -1) – 1]/[(.5/2) ((2/.10) -1) – 1] = (.519 – 1)/(.25 19 – 1) = 99 percent. However, neither of the two formulas we’ve reviewed so far uses a strategy’s winning percentage as an input.

Using winning percentage This formula, adapated from Robert P. Rotella’s The Elements of Successful Trading (New York Institute of Finance, 1992), demonstrates how risk of ruin depends on a strategy’s percentage of winning trades and the fraction of account equity risked. Risk of ruin = [(1 - PD) / (1 + PD)] f where PD = Percentage Difference, or the winning percentage

minus the losing percentage ( e.g., .40 for a winning percentage of .70). f = Fraction of account equity risked on a trade, expressed in “units.” If you have a $50,000 trading account and you risk $1,000 per trade, you have 50 trade units. Units can be also determined by dividing one (1) by the percentage of account equity risked per trade. For example, if you risk 5 percent of equity per trade, you have 1/.05 = 20 units. Table 1 shows the risk of ruin across several PD and account equity risk levels. TABLE 1 RISK OF RUIN Win % PD 5% 55 60 70 80 90

10%

Risk % 20%

33%

50%

0.1

1.81% 13.44%

36.66%

54.77%

66.94%

0.2 0.4 0.6 0.8

0.03% — — —

13.17% 1.45% 0.10% 0.00%

29.63% 7.87% 1.56% 0.14%

44.44% 18.37% 6.25% 1.23%

1.73% 0.02% — —

Practical value Because of their inherent limitations, these formulas can give traders only a rough feel for how risky a strategy might be. The assumptions on which they are based (neglecting to factor in winning percentage, and/or requiring wins and losses always to be the same size, etc.) are too rigid for the complexities of the trading world. The most beneficial aspect of the risk-of-ruin concept is arguably its clarification of the dangers of risking too much money on a trade. All the math in the world cannot predict the f uture. But you don’t need to know the future to understand the two simple rules that are the foundation of conservative trading: The less you risk, the less likely you are to get wiped out. And the more (and larger) winning trades you have, the more likely you are to make money. Pretty simple stuff, when you think about it.Ý Some of the formulas in this article were adapted from those at www.bjmath.com. www.activetradermag.com • April 2003 • ACTIVE TRADER

TRADING Basics

Taking a DIFFERENT ROUTE ECNs have changed the face of short-term trading. Here’s a look at what they do and why they’ve become so popular.

BY JEFF PONCZAK

A

lthough they didn’t exist just a few years ago, Electronic Communications Networks (ECNs) have become an integral part of the equity-trading landscape. While swing traders and those with a longer-term time frame may not be over-

Glossary: Market makers: Professional traders who provide liquidity and execute orders in Nasdaq listed stocks. Spreads: The difference between a stock’s bid and ask price. Order book: A list of open buy and sell orders. Slippage: The difference between the price you expect to get and the price at which you actually get filled. Specialists: The NYSE equivalent of market makers. They provide liquidity and execute orders in NYSE stocks.

ly reliant on them, they are the lifeblood of day traders and scalpers. Without ECNs, short-term trading would be much diff erent, and likely much more difficult, than it is today. The trading opportunities provided by ECNs are so important that brokerage firms (especially direct-access firms) tout their connectivity to various ECNs; the Nasdaq recently revamped its entire trading platform in an effort to regain volume lost to ECNs. However, it’s likely that while many new traders (and possibly even some veteran traders) appreciate the importance of ECNs, they have no idea how they actually function.

Back in the last century… In late 1996, the Securities and Exchange Commission (SEC) found that Nasdaq market makers were illegally inflating spreads to increase their profits. Among several rule changes put in place was one that allowed the creation of alternative trading systems — non-exchange organizations that matched buy and sell orders in stocks. In early 1997, four alternative trading systems — the first modern ECNs – opened their virtual doors and began matching stock trades. Instinet, which today is still one of the leading ECNs, has been around since the 1960s. Up until very recently, Instinet served only institutional customers, providing a non-exchange venue for execution of stock trades. So, while Instinet


was truly the “first” ECN in terms of the services it provided, the term ECN was not created until 1997. Since 1997, several other ECNs have launched operations. At the peak of the late ’90s bull market, there were 11 ECNs and more being planned. Although consolidation has changed the look of the ECN landscape over the last six years (there are currently nine ECNs), one thing hasn’t changed: ECNs deal almost exclusively with Nasdaq stocks (about 95 percent of total ECN volume is on the Nasdaq). When ECNs first came into existence, the New York Stock Exchange (NYSE) had a rule prohibiting trading of many of its listed stocks anywhere except on the floor of the exchange. Although that rule has since been repealed, very little NYSE volume has moved to ECNs. Although the ECNs hope to change that, it will likely be a while before significant non-exchange NYSE volume develops.

So, what do they do? All ECNs have an “order book” no diff erent than the order book kept by Nasdaq market makers or NYSE specialists. There are two kinds of ECNs: “execution” ECNs and “routing” ECNs. Both types try to match buyers and sellers, but if a routing ECN can’t make an immediate match within its own order book, it will search for another match (i.e., with another ECN, a market maker, etc.) and send 49

the order there. Most importantly, ECNs have the ability to perform these functions in less than a second. In terms of volume, the No. 1 execution ECN is Island and the No. 1 routing ECN is Archipelago, which last year acquired REDIBook, another routing ECN. Archipelago recently became an exchange, although its ECN business remains the same. Traders do not set up accounts with ECNs. Instead, traders gain access to ECNs through a broker. All brokers that use direct-access technology have the ability to trade with ECNs, and an increasing (but still relatively small)

number of non-direct-access platforms and both can see the lowest current ask catering to active traders are making use price is 49.33. of ECNs. After the non-ECN trader enters the To fully understand the benefits of trade information and clicks the “buy” using ECNs, we’ll look at the same trade button, the trade gets sent to the order made by two traders — one who uses desk at the trader’s brokerage firm. ECNs, and one who doesn’t. (When trading through a standard “Web-based” broker, hitting the buy butThe need for speed ton is generally no different than sendLet’s assume both traders have access to ing an e-mail.) The order desk then the Nasdaq Level II screen (which shows sends the order to a market maker, who multiple levels of bids and offers) and fills the order if a match is available. can see the best bid and best offer in a Of course, the market maker has five particular stock. Figure 1 shows a Level seconds to decide what to do with the II screen for Microsoft (MSFT). trade, and in that time he may take the Both traders want to buy 100 shares other side of it, send it elsewhere or reject it outright. If an order is filled, the entire process FIGURE 1 LEVEL II SCREEN often takes less than 30 seconds. However, in a fast-moving or illiquid A trader with ECN connectivity could look at a Level II screen, see which market, 30 seconds is enough to change market participant has the best price and send the order there. Traders with the price of a stock dramatically. Market out access to ECNs are at the mercy of their brokerage. orders may get filled at a price much worse than where the stock was trading when the trade was entered, and limit orders might not get filled at all. On the other hand, the trader using ECNs has many options. The Level II screen shows that four ECNs — Brut (shown on the Level II screen as BRUT), Instinet (INCA), Island (CINN — Island posts all its quotes through the Cincinnati Stock Exchange), and Track (TRAC) — are sellers at 49.33. Directaccess software allows traders to choose a specific ECN to send an order to, and a limit order of 49.33 sent to any of these four ECNs will arrive in their ord er books almost instantaneously and stands a good chance of being immediately executed. In any event, the large majority of ECN trades are executed without that trade ever going into the Nasdaq order book, where it can be manipulated by market makers. To put things in perspective, one should consider what trading was like in the pre-ECN era. Market makers were very poorly regulated, and if you entered an order (especially in the preInternet days, when orders were placed over the phone), you could only hope that it got filled (and if it did, you could only hope that someone would call you Source: eSignal back and confirm it).

50


Because time is of the essence, directaccess brokerages typically allow traders to use preset “one-click” trading buttons to execute trades a certain way. In other words, a trader can designate a “hotkey” (e.g., the F4 button, or CTRL+ I) that will automatically send an order to Island or any other ECN. This eliminates the need to manually enter an ECN to send an order to, which would cost precious seconds. Some traders only look at best bid and offer prices and do not concern themselves with the depth of the Level II screen. These traders often designate all FIGURE 2 ISLAND ORDER BOOK The Island’s order book in Microsoft (MSFT) shows bids far below the highest bid price and far above the lowest offer price. This depth makes Island a popular choice for traders using ECNs.

Keep in mind, there are other execution venues besides ECNs. High-end trading platforms also give you the choice to send orders to the Nasdaq, and non-ECN alternative trading systems also are an option. trades to go through Archipelago. In the previous example, if a trader sent a limit order of 49.33 to Archipelago (ARCA on the Level II screen), Archipelago would “scan” the market and send the order to one of the four ECNs for automatic execution. Market makers on the best bid/offer could also have trades sent to them thro ugh Archipelago.

Nobody’s perfect

Source: The Island ECN

Of course, while automatic execution is the ideal, it doesn’t happen all the time. Limit orders can sometimes linger, and slippage does occur. While a routing ECN may seem to be the ideal situation (i.e., if a trade isn’t matched in the order book, it’s sent out), that’s not always the case. Although some traders are content to send all their orders through Archipelago, others trade exclusively (or at least primarily) on Island just because of the sheer depth of the Island order book. In some stocks, Island accounts for almost one out of every five trades. And since Island merged with Instinet, the leading institutional ECN, in 2002, its liquidity pool has become even bigger. Say the ECN trader mentioned previously wanted to exit a trade rather than enter one. Three firms — Archipelago, Instinet and market-maker J.P. Morgan (JPHQ) — are showing the best bid. Island is a penny below them. However, a look at the Island order book (Figure 2) shows numerous orders at many price levels (some just .001 away from the next level). Some traders find Island to be the most reliable ECN simply because of the huge amount of volume it does, and they choose to route


all orders there.

The big two There are nine ECNs — Island, Archipelago, Instinet, Bloomberg Tradebook, Brut, MarketXT, Attain, NexTrade and Track — but the reality is that Island, Archipelago and Instinet are responsible for a disproportionate amount — more than 85 percent — of all ECN volume. While most direct-access trading platforms allow traders to route an order to any ECN, many successful traders never use anything other than Island or Archipelago. And since liquidity begets liquidity, it’s rather unlikely that the pecking order among ECNs will change in the foreseeable future. Keep in mind, there are other execution venues besides ECNs. High-end trading platforms also give you the choice to send orders to the Nasdaq, and non-ECN alternative trading systems also are an option. And, orders executed through an ECN have additional fees (no more than a penny a share, usually less) not associated with other market participants.

There’s more to trading than ECNs. In many ways, an ECN is a short-term trader’s best friend. However, using them won’t make up for a flawed trading strategy, poor money management or lack of discipline. Plus, configuring your trading software to route orders in the most efficient manner requires a more thorough understanding of each ECN than is presented here. ECNs may be the fastest car on the highway, but if you don’t have a driver’s license you’re doomed to wind up skidding off the side of the road. Ý 51

TRADING Basics

Indicator Insight: True

Strength Index and Ergodic oscillator

BY ACTIVE TRADER STAFF

T

he True Strength Index (TSI) is an indicator that contains elements of both trend-defining and oscillator tools (an oscillator is a momentum indicator that highlights shorter-term overbought and oversold levels). The indicator was developed by William Blau and explained in his book, Momentum, Direction and Divergence (John Wiley & Sons, 1995). The TSI is based on a concept called “double smoothing,” by which Blau applied two exponential moving averages (EMAs; see sidebar, opposite page) to a one-day momentum calculation (defined as the difference between yesterday’s closing price and today’s closing price).

Calculation The TSI is defined as: TSI = 100*[{EMA 2(EMA1(mtm))}/{EMA2(EMA1(|mtm|))}] where mtm = the one-day momentum (today’s close – yesterday’s

close) |mtm| = absolute value of one-day momentum. EMA1 = a longer-term exponential moving average (e.g., 60 bars). EMA2 = a shorter-term exponential moving average (e.g., five bars).

FIGURE 1 THE TRUE STRENGTH INDEX/ERGODIC OSCILLATOR The TSI line (blue) is a simple momentum calculation “double-smoothed” by two exponential moving averages (EMAs). The addition of a “signal line” (red), which in this case is a 10-day EMA of the TSI line, results in the Ergodic oscillator, which ranges from +100 to -100 and indicates both trend direction (when the TSI crosses above and below the signal line) and momentum strength (whether the indicator is at a relative high level or low level). Oracle Corp. (ORCL), daily 13 12 11 10 9 Divergence between rising price peaks and flat TSI peaks

8

True strength index (TSI)/Ergodic oscillator

14 6

TSI line

2

Signal line September

October

November


52

Note the numerator and denominator of the indicator calculation differ in only one respect: the numerator uses one-day momentum, which can be positive or negative; the denominator uses the absolute value of oneday momentum, which is always positive. This results in a figure that is always between –1 and +1. Multiplying by 100 sets the indicator’s range between –100 and +100. The addition of a moving average of the TSI (a “signal line”), which Blau originally designated as a fiveday EMA, created an indicator called the Ergodic oscillator. Figure 1 shows the TSI/Ergodic oscillator on a daily chart. The thick blue line is the TSI line, which is the double-smoothed momentum calculation; the thin red line is the signal line. In this example, the longer-term average is a 60-day EMA and the shorter-term average is a 20-day EMA. The signal line is a 10day EMAof the TSI line.

Interpretation and use December

January 2003

As a trend indicator, the Ergodic oscillator signals an uptrend when the TSI line is above the signal line, as it was


FIGURE 2 INTRADAY TREND Here the TSI/Ergodic oscillator highlights a choppy uptrend on an intraday chart. Notice the indicator reflects the price trend but removes most of the choppiness, or “noise.” At the beginning of the chart, the TSI crossed above its signal line when the indicator was at a relatively low level, indicating an uptrend. Near the end of the chart, the TSI crossed below its signal line when the indicator was at a relatively high level, indicating a downside reversal. Oracle Corp. (ORCL), 10-minute

12.6 12.4 12.2 12.0

f rom early October to early December in Figure 1; it signals a downtre nd when the TSI line is below the signal line, as it was in September. This application will be recognizable to anyone familiar with the equivalent interpretation of the moving average convergence-divergence (MACD) indicator. Because the TSI is bound within a fixed range between -100 to +100, it is possible to use the indicator to determine when price is potentially overextended — i.e., overbought or oversold. Higher TSI values reflect increasingly overbought prices and lower values reflect the opposite condition. In Blau’s writings, over bought and oversold levels for the indicator are typically set equidistant from the midpoint — for example, +25 and -25, respectively, although these levels will vary depending on the length of the moving averages used to calculate the indicator. An interesting aspect of these parameters is that the indicator can move much higher and lower than these representative overbought and oversold levels, which highlights the fact that markets with strong momentum can remain, technically, “over bought” or “oversold” for extended periods of time. A trend reversal signal occurs when the indicator is at a relatively high or low extreme. In Figure 2, which applies the TSI/Ergodic to a 10-minute chart, note the buy and sell signals that occurred when the indicator was at a low level (near the beginning of the chart) and at a high level (near the end of the chart).

Key points Blau’s double-smoothing technique was based on his observation that a very long-term moving average of momentum was an accurate — and smooth — approximation of price movement. Further, applying a sec-

11.8 11.6

True strength index (TSI)/Ergodic oscillator

20 0

Crossover at high indicator level

-20

Crossover at low indicator level 13:00

1/21

10:50

13:00

1/22

10:50

13:00

1/23

10:50

13:00

1/24

10:50


Exponential moving average (EMA) The simple moving average (SMA) is the standard moving average calculation that gives every price point in the average equal emphasis, or weight. For example, a five-day SMA is the sum of the most recent five closing prices divided by five. Weighted moving averages give extra emphasis to more recent price action. The exponential moving average (EMA) weights prices using the following formula:

EMA = SC * Price + (1 - SC) * EMA(yesterday) where

SC = a “smoothing constant” between 0 and 1, and EMA(yesterday) = the previous day’s EMA value. You can approximate a particular SMA length for an EMA by using the following formula to calculate the equivalent smoothing constant:

SC = 2/(n + 1) where

n = the number of days in a simple moving average of approximately equivalent length. For example, a smoothing constant of .095 creates an exponential moving average equivalent to a 20-day SMA (2/(20 + 1) = .095). The larger n is, the smaller the constant, and the smaller the constant, the less impact the most recent price action will have on the EMA. In practice, most software programs allow you to simply choose how many days you want in your moving average and select either simple, weighted or exponential calculations.


53

FIGURE 3 SIMILAR INDICATORS The TSI/Ergodic oscillator from Figure 1 is shown along with the MACD and sto chastic oscillator. All three of the indicators use double smoothing, but the TSI double smoothes a one-day momentum calculation rather than price. This exam ple shows the TSI and its signal line are a little smoother than their MACD and stochastic counterparts. Oracle Corp (ORCL), 10-minute

True strength index (TSI)

MACD

Stochastic Slow

September

October

November


December

oscillator. The similarities between the indicators are obvious. One distinguishing characteristic of the 12.0 TSI/Ergodic is that the primary indi9.5 cator line (the TSI) is slightly smoother than the other indicators, and its signal line “whipsaws” above 14 and below the main indicator line less than those of the other indica2 tors. Besides identifying price direction and overbought and oversold levels, 0.5 divergence between TSI peaks and 0 bottoms and price peaks and bottoms can also signal potential reversals. For 100 example, in Figure 1, the stock made consecutive higher price peaks in 50 December and January. The TSI/Ergodic’s January peak, howevJanuary 2003 er, failed to exceed its December peak, suggesting the January price high was established on weakening momentum. The TSI line subsequently dropped below the signal line, indicating a downtrend.

ond, shorter moving average to this first calculation resulted in an indicator that removed most market “noise” while lagging very little. (Moving averages of price, by contrast, lag the price action relative to the number of days used to calculate the aver- Bottom line age; the longer the average, the greater the lag.) The MACD and The True Strength Index is similar to other indicators that have the stochastic oscillator are two common indicators also based both trend-following and momentum characteristics, such as on calculating moving averages of moving averages. the MACD. The unique aspect of the TSI is the concept of douThe TSI reflects longer and shorter price action, or trends, ble-smoothed momentum used to create it. This concept — that depending on whether longer or shorter EMAs are used. In his double-smoothed momentum creates an indicator with mini book, Blau illustrated a number of different EMA lengths. mum noise and lag — can be used in a variety of situations to Figure 3 shows the same TSI/Ergodic oscillator from Figure 1 determine both trend direction and overbought and oversold along with the standard MACD and a longer-term stochastic conditions.Ý

54


TRADING Basics

Technical Tool Insight:

Gaps

BY KIRA MCCAFFREY BRECHT FIGURE 1 PRICE GAPS

A

price gap occurs when a price bar’s low is above the previous bar’s high (an “up gap” or “gap higher”) or its high is below the previous bar’s low (a “down gap” or “gap lower”). Figure 1, shows examples of

both. Gaps reveal a quick shift in supply-demand dynamics. They appear when there is a sudden change in market sentiment or outlook, which causes buyers or sellers to quickly accelerate their buying or selling. Generally, up gaps in uptrending markets are considered signs of strength, while down gaps in downtrending markets are viewed as evidence of weakness. Gaps are often referred to by specific names — e.g., common gap, breakaway gap, running or measuring gap, and exhaustion gap — depending on the market context in which they appear. On candlestick charts, gaps are called “windows.” A variation of the basic gap is the “opening gap,” which occurs when price opens outside the range (above the high or below the low) of the previous bar. However, opening gaps can close very quickly, after which no true bar-to-bar gap will exist.

Gaps defined The relationship between price bars that form a gap is simple to define: Up gap = today’s low > yesterday’s high or Down gap = today’s high < yesterday’s low.

A gap occurs when a price bar’s low is higher than the preceding bar’s high (up gap) or its high is lower than the preceding bar’s low (down gap).

ple, an up gap suggests highly motivated buyers are willing to pay more for a security than they were the previous trading up gap session. On the flip down gap side, a down gap means sellers are eager to exit long positions or go short below the lowest price of the previous session. Gaps often occur when surprising news hits a market — e.g., an unexpected earnings drop for a stock, or an unforeseen frost or disease that decimates an agricultural crop. Gaps are more meaningful (reflecting greater buying or selling pressure) when they occur within trends rather than trading ranges. Of the four types of gaps mentioned in the preced-

FIGURE 2 INTRADAY GAPS Intraday price gaps, such as those on this one-minute chart, are usually natural reflections of random short-term volatility (“noise”) or a sign of a thinly traded market. Chicago Mercantile Exchange (CME), one-minute

43.20 43.10

Although gaps can occur on all time frames, the longer the time frame, the more significant the gap. Price gaps occur frequently on intraday time frames (see Figure 2), especially in thinly traded markets. Such gaps are either simply part of the natural intraday volatility in a market or evidence of a lack of liquidity — not meaningful reflections of supply and demand dynamics. The remainder of the discussion addresses gaps on the daily (or longer) time frame.

43.00 42.90 42.80 42.70 42.60 42.50 42.40

Interpretation and uses A gap reveals a dramatic shift in the perceived value of a market from one day (or week, etc.) to the next. For exam-

14:23

14:28

14:32

14:36

14:40

14:44

14:48

14:52

14:56



55

ing section, common gaps (also known as “pattern” gaps), which occur within trading ranges or other trendless periods,

are the least significant and are often filled quickly — generally within the next few days. Figure 3, left, shows how “common” common gaps can be. Throughout spring FIGURE 3 COMMON GAPS and much of summer 2002, heating oil futures traded in a sideways range. Common gaps, which are also called pattern gaps, are those that appear in Many price gaps — both to the upside non-trending, sideways markets. They generally reveal little about price and the downside — occurred during momentum or supply and demand. this period. However, they were generally filled within a week or so, and January heating oil (HOF03), daily 0.7500 revealed little about price behavior. 0.7400 “Breakaway” gaps are more significant. This type of gap occurs when price 0.7300 moves forcefully out of a consolidation pattern (such as a trading range or a tri0.7200 angle) or some other low-volatility envi0.7100 ronment; such gaps can initiate significant price moves, as shown in Figure 4,. 0.7000 Throughout the month of August, the 0.6900 market traded sideways. On Sept. 4, 2002, it gapped higher and closed above 0.6800 the top of the August trading range (and also gapped completely above a narrow0.6700 er trading range that formed within the 0.6600 larger range). This breakaway gap ushered in a strong new uptrend. The third type of gap is a running (or April May June July “runaway”) gap, which generally occurs Source: TradeStation Platform by TradeStation Group in the midst of an established trend and is considered a sign of acceleration of bullish or bearish momentum (see Figure 5). FIGURE 4 BREAKAWAY GAP Sometimes, several running gaps will appear around the midpoint of a trend. A breakaway gap occurs when price gaps higher or lower out of The final type of gap is the exhaustion gap, which as a market consolidation, such as the trading range(s) shown here. the name implies, occurs toward the end of a trend Such gaps indicate the breakout occurred on strong momentum, when the buyers or sellers are running out of steam. and so might be a sign of a healthy new trend. The gap represents a last gasp of momentum and is followed by a trend reversal. For example, at the end February porkbellies (PBG03), daily of an uptrend, prices will leap higher, but the buyers 76.000 will fail to sustain the move. A quick retreat below the 74.000 gap confirms a negative turn in the market. Figure 6, shows an example of an exhaustion gap at 72.000 the top of an uptrend. A long uptrend accelerates in the 70.000 final three bars, the last of which gaps higher. The next Breakout bar closes above the wider bar reverses dramatically, filling the exhaustion gap 68.000 trading range and beginning a downtrend. 66.000 64.000 62.000

Breakaway gap out of a narrow trading range September Source: TradeStation Platform by TradeStation Group

56

60.000

Key points There is an old market adage that “gaps are meant to be filled.” This is not always the case, but gaps do typically represent support or resistance levels (floors or ceilings that repel price movement in that direction). These levels have more significance in trending markets than non-trending ones. Heavy volume on breakaway gap days helps confirm the buying or selling power behind the new direcwww.activetradermag.com • April 2003 • ACTIVE TRADER

FIGURE 5 RUNNING GAP tional move. In the case of upside breakaway and running gaps, price will often Like breakaway gaps, running gaps reflect strong bullish or bearish momen retreat to test and possibly fill the gap. As tum. Running gaps, however, occur in markets that are already trending, and long as the far side of the gap range is not so reflect accelerated trend momentum. Gaps represent support and resist penetrated significantly, the gap is conance levels until they are penetrated significantly. Here, price soon pulled sidered a valid support level. back to the gap (“testing” the support level) before reversing to continue Notice in Figure 5, price pulled back to the uptrend. the gap five days later — penetrating but not filling it — before turning back to the Rambus (RMBS), daily upside. However, it also is not uncom9.50 mon for price to return to a gap after a few days, fill it (penetrating its far boundary only slightly), and then 8.50 reverse in the direction of the preceding Running gap trend. 7.50 The longer a gap remains unfilled, the higher the likelihood it represents a Test of support at meaningful support or resistance level. 6.50 top of gap Conversely, gaps that are filled quickly and/or are penetrated significantly 5.50 should be interpreted as reversal signals (as was the case of the exhaustion gap in Figure 6). 4.50 One “problem” with exhaustion gaps is that there is no way to diff erentiate October November December them from running gaps until after the fact. If the market in Figure 6 had continSource: TradeStation Platform by TradeStation Group ued to rally higher instead of reverse, the exhaustion gap would have been a running gap. However, a series of running gaps, is more likely to be an exhaustion gap. The important development to look for is whether or not price gap that occurs after a long-term trend, and especially after a violates the gap level. Gaps are continuation patterns that imply momentum strength in the direcFIGURE 6 EXHAUSTION GAP tion they occur — until they are filled. The exhaustion gap represents a final outburst of momentum before a trend loses its final strength and reverses. The fact that this gap occurred after an extended uptrend and the quick — and significant — downside reversal of the gap on the next bar were indications the uptrend had ended. Continuous coffee (KC), daily 300.00

Exhaustion gap

280.00 260.00 240.00 220.00 200.00 180.00 May

June



Bottom line In trending situations, price gaps can represent accelerated buying or selling strength that can establish important support and resistance levels, but in non-trending conditions (or in thinly traded markets), they are generally not meaningful, other than as a sign of illiquidity or choppiness. Also, intraday gaps are very common and reveal little about price action. Gaps that are not filled quickly (say, two to 10 bars) tend to represent more significant support and resistance levels. Ý Neither Technical Tool Insight nor Indicator Insight is intended to recommend (or disavow) any indicator, trading tool, methodology or school of thought. The goal of each is simply to explain the background and mechanics of well-known trading concepts and their most common uses, and to give the reader the basis for further research, through which he or she should deter mine the value of the concept. For information on the author see p. 3. 57

TRADING & Investing

Give me land, LOTS OF LAND… If you think booms or busts in the housing market provide an accurate indication of how housing stocks will perform, you may be in for a surprise. A smart trader with a keen understanding of macroeconomics knows where to look to find the real story.

BY PETER NAVARRO

A

n important question hung in the air as 2002 rolled into 2003: Is the housing market a “bubble” about to burst? On the one hand, a gaggle of experts — from academics and homebuilder lob byists to the Fed Chairman himself — repeatedly insisted at the turn of the new year that there would be no repeat of the housing market collapse that occurred in the late 1980s. The argument was straightforward: The builders had learned their lesson from the last bubble and were not flooding the market with “spec homes” and oversupply. Interest rates and mortgage rates would likely remain low for the foreseeable future, helping fuel demand. The demographics of an aging population likewise favored a continued strength in demand, as a bulge of 3058

somethings finally began to nest in earnest while a horde of recent retirees were now buying “second homes.” That’s all well and good. However, even if all this were true, does it still mean that housing stocks are good instruments to buy? Answering this question is precisely where a deeper understanding of economics comes in to play.

The media perspective The first level of analysis is the one you will likely have encountered in the popular media. As frequently quoted in the press, bubble advocates were quick to point out several ominous factors as 2002 came to a close. The interest rate cycle was at a bot-

tom. With soaring budget deficits and an economic recovery on the horizon, it would not be long before mortgage and interest rates spiked. If the recovery did not arrive as forecast, it would still be true that consumers had become exhausted — at least when it came to both the auto and housing sectors. Indeed, with their credit extended and geo-political uncertainties casting a longer-term pall over their confidence, it wouldn’t be long before consumers stopped splurging on big-ticket items. Perhaps worst of all, housing prices appeared to have already peaked in many major markets. Their fall would eat right into the bottom line of housing sector stocks.


FIGURE 1 HOUSES GO UP, STOCK PRICES GO DOWN While single-family home sales have risen steadily since the beginning of 2002, the same cannot be said of three big housing stocks. This is an indication rising sales are making vacant land harder to come by, and builders are losing profits because of the need to “restock” their land supply. Beazer Homes (BZH),

12/31/02

daily

2,000

60%

Single-family homes sales

1,075

50%

1,050

40%

1,025

30%

1,000

20%

975

10%

950

0%

925

-10%

900

-20%

875

-30%

850

-40%

Volume 3 2 1

2002

Feb.

Mar.

Apr.

May

June

July

Aug.

Sept.

Oct.

Nov.

Dec.

0 2003

Source: BigCharts.com

So much for the first level of bubble analysis. Where the active trader plot really thickens, however, is with the second level — an economic analysis of the underlying structure of the industry.

The macrowave perspective Table 1 illustrates the four major stages

of the development process of a house, prior to the actual sale. The major players are listed, as is the profit potential for each. Let’s work backwards and start with the last stage — home construction by the building contractor. Here we see the pro fit potential in this stage is very low. Why?

TABLE 1 THE FOUR STAGES OF DEVELOPMENT When it comes to housing development, the real money is in owning the land, not building the house.

Stages of development

Major player

Acquiring the raw land

Land owner

Very high

Sub-division and home design

Developer

Low

Obtaining the various building permits

Developer

Medium

Construction of the house

Contractors

Very low


Profit potential

Simply because building contractors operate in a highly competitive industry and are generally forced to bid on projects for construction. Because projects typically go to the low bid, there is little money to be earned in this segment of the business — in the language of economics, contractors merely earn a “normal” rate of return. That’s why most of the big housing developers tend to put out the actual construction of their homes to bid rather than try to make money in this segment of the business. Instead, they focus on the more profitable second and third stages of the process — designing the homes and subdivisions and then obtaining the various building permits. In this area, it is not in the actual design of a sub-division where much of the profits are made. For the most part, this is cookie-cutter stuff where plans are continually recycled. Instead, the big money gets 59

made in the securing of development hundred million windfall. as housing prices rise, so, too, does the rights and building permits. Note, however, these kinds of wind- price of the available raw land necessary Because of this, in most urban areas falls are always one-time deals. That’s to continue to fuel the boom. where potential traffic congestion and why what ultimately constrains the profIt follows that in every housing boom, school overcrowding may be generating its and stock prices of homebuilders is developers inevitably reach a point where significant anti-development sentiment, the first stage in the development most of their potential profits for future developers can face severe political oppo- process — namely, the acquisition of the projects will not be realized. Instead, they sition to gaining approval for new subdi- raw land. will have to dissipate these profits upfront visions. That’s why virtually every major This first stage is the most important through payments to the owners of raw homebuilder has a powerful lobbying to understand from a macrowave per- land. f orce to push projects through local (and spective because it is here where homePerhaps the best way to understand sometimes state) jurisdictions. this basic economic principle is Of course, in this political to think about a shack — or TABLE 2 PRICE/EARNINGS RATIOS OF SELECTED battle, the game is always to even a nice 2,000 square foot HOMEBUILDERS AND TECHNOLOGY COMPANIES increase the number of houses home — sitting on a coveted one can pack on to a parcel of piece of oceanfront property. While tech stocks use the promise of new innovation land while minimizing any The property might go for $5 to justify high P/E ratios, housing stocks are destined fees or costs associated with million with the shack and to have P/E ratios in the single digits. traffic mitigation and other maybe $5.25 million with the forms of public services such nice home. But the land is Homebuilders Price/earnings ratios as parks and schools. Indeed, where all the value — and every extra house becomes profits — ultimately reside. Beazer Homes (BZH) 5.60 pure profit. For this reason, the biggest Centex (CTX) 7.70 Now here’s the broader long-term beneficiary of the point: The developer who is housing boom in the developRyland (RYL) 5.86 able to swiftly obtain ment sector tends to be Technology companies approvals for his projects has landowners — not the actual basically been given the housing companies. Moreover, Dell (DELL) 36.14 development rights to signifiif a developer winds up buying IBM (IBM) 23.98 cantly boost his bottom line. land at exorbitant prices at a And the profits will be all the market peak and then the marMicrosoft (MSFT) 30.77 greater if these rights are ket softens, he’s just doomed obtained early in the boom his company to heavy losses. Source: www.finance.yahoo.com cycle before housing prices It follows that if you come to have started to rise significompletely understand the cantly. builders at the end of a boom are likely economics of the housing market, you For example, a typical developer in to see their bubble burst. Here’s the quickly realize the point illustrated in California might get approval for a problem: Table 2: Unlike the technology sector, 2,000-home subdivision. His business During a prolonged housing boom, where new innovations can make the sky plan may have originally projected an housing prices do indeed rise substantial- the limit for stock prices, housing sector average price of $250,000 for each home. ly, adding to the current profits of the stocks are forever captives of the owners However, by the time the homes actually developers. During such a period, howev- of raw land. They are eternally doomed hit the market, he winds up with an er, most homebuilders are forced to draw to earning merely a decent level of profaverage price of $350,000 per house. down heavily on their own purchased its in boom times and just as eternally That’s a difference of $100,000 and a $2 stocks of available raw land. Meanwhile, damned to facing huge risks every time their markets collapse with rising interest rates and geo-political uncertainty. That’s precisely why the price-earnings ratios of the homebuilders in Table 2 are laughably low compared to the leaders in technology. It’s also why home building stocks can be highly risky to trade on the long side if you get caught up in the boom rhetoric. Ý

During a prolonged housing boom, housing prices do indeed rise substantially, adding to the current profits of the developers. 60

For information on the author see p. 3.


The Business of TRADING

The PROOF is in the

RETURN

Traders have distinct tax advantages over non-traders. By comparing tax returns, you’ll see exactly how much those advantages are worth. And, you’ll see how much extra money you can save

BY ROBERT A. GREEN, CPA

by forming a single-member LLC and contributing to a Mini 401(k) plan. TABLE 1 COMPARING TAX RETURNS The numbers don’t lie: A trader who forms a single-member LLC enjoys greater tax savings than a sole proprietor trader, who enjoys greater tax savings than an investor lacking trader tax status.

Tax returns for 2002

Income (expense)

Investor (no trader tax status)

Trading gains $248,558 Schedule C - Trading business $0 Net trading gains reported Form 4797 Net trading gains reported on Schedule D $248,558 Schedule C - Administration Interest & dividend income $894 Gross income $249,452 Adjusted gross income deductions: One-half of self-employment tax $0 Self-employed health insurance deduction $0 Mini 401k plan retirement plan - maximum $0 Total adjusted gross income deductions $0 Adjusted gross income $249,452 Itemized or standard deductions Exemptions Taxable income Federal income tax Federal self-employment tax California total tax California single-member LLC tax Total tax Tax savings Total tax savings

($21,789) ($300) $227,363 $66,830 $0 $19,636 $0 $86,466 $0


Sole Proprietor Trader

Single Member LLC Trader

$248,558 ($43,900) $204,658

$248,558 ($197,020) $51,538

$894 $205,552

$152,320 $894 $204,752

$0 $0 $0 $0 $205,552

($7,304) ($2,415) ($40,000) ($49,719) $155,033

($4,700) ($1,320) $199,532 $57,089 $0 $16,962 $0 $74,051 $12,415

($4,700) ($2,520) $147,813 $38,987 $14,607 $12,312 $800 $66,706 $7,345 $19,760

W

hile there are a variety of ways to reduce your tax liability, there is one underlying truth that is essential come tax time: Those who qualify for trader tax status enjoy significant tax benefits over those who don’t qualify (i.e., investors). This month, we’ll use sample tax returns to show just how much better off the trader is. And, we’ll also show how a trader who forms a single-member Limited Liability Company (LLC) in conjunction with a Mini 401(k) plan can save even more. Before we get into details, let’s first review the requirements for trader tax status. The IRS Web site (www.irs.gov) details this information in Chapter 4 (Special Rules for Traders) of Publication 550. If you trade all day, every day, you should not have a problem attaining trader tax status. However, if you don’t, the IRS rules are vague and subjective. If you have less than 1,000 round-turn trades, you should seek the further help of a trader tax professional. If you qualify as a trader, the IRS considers your trading a “business activity.” This allows you to file a “sole proprietor” tax return and report trading business expenses as “ordinary” on Schedule C, where they are deducted in full against 61

FIGURE 1 INVESTOR TAX RETURN Because investors get minimal tax breaks, their adjusted gross income (line 35) is their standard income. Deductions (line 38) are limited.

gross income. If you fail to qualify as a trader, you must file as an investor, which means your trading expenses are considered investment expenses reported on Schedule A. These expenses may be limited, and some of them (e.g., home-off ice expenses and seminars) are not deductible at all. Trader status is not something you need to elect. You need to do nothing more than indicate it on your tax return, providing you quali fy (and the IRS agrees). Many traders confuse trader tax status with “mark-to-market accounting” (MTM). MTM does need to be elected (via a statement filed with the IRS) by April 15 of the tax year you wish to have MTM. For more information on MTM, see “Take control of your taxes,” ActiveTrader, February 2002, p. 96. MTM has no effect on the examples that follow, because the trader is profitable and did not hold any open positions at year-end. If the trader had lost money, MTM would be a huge tax benefit, as it would convert capital loss limitations and carryovers into ordinary loss deductions, generating immediate tax relief. Our scenarios involve Mike, a trader from California. The first example shows how he saved $12,415 by filing as a sole proprietor as opposed to an investor. The second shows how he could have saved an additional $7,345 had he formed a single-member LLC and funded a Mini 401(k) retirement plan. Each sample tax return has the same income and expenses for the taxpayer. For comparison purposes, it’s apples to apples. Table 1 (opposite page) breaks down the differences between the three returns.

Sole proprietor vs. investor Mike is a single 26-year-old resident of California. He trades securities (no commodities) every day, all day from his one-bedroom office (located in his two-bedroom apartment). He qualifies for trader tax status. Mike has typical business expenses for margin interest ($15,752), depreciation on equipment ($6,413), amortization on software ($646), supplies, travel, meals and entertainment, postage, telephone, tax and accounting services, chat rooms, Internet service providers, online information services, publications and books, and seminars. 62


FIGURE 2 SOLE PROPRIETOR TAX RETURN Sole proprietors have less adjusted gross income than investors (line 35), and they are able to claim a standard deduction (line 38).

Mike has a trading account at a direct-access brokerage, and in 2002 his realized trading gains on securities were $248,558. He did not keep any open positions at year-end. Mike had previously elected MTM, so he reports his trading gains on Form 4797, Part II (ordinary gain or loss). His trading account had intere st income of $795 and dividend income of $99. Taking a look at the Form 1040, we see some big differences between the investor tax return (Figure 1) and the sole proprietor return (Figure 2). The investor must report his full income on Line 13, Capital gain or (loss). However, because the trader qualifies as a business owner, he is able to deduct $43,900 in business expenses on Schedule C and transfer the same amount of trading gains from Form 4797. The overall effect is zero net business income on Schedule C (line 12) and lower Form 4797 trading gain income (line 14). On Line 38 (Itemized or standard deductions), the investor is entitled to limited business expenses and produces a taxable income (Line 41) of $227,363. The trader has already deducted (in full) these expenses elsewhere, and because he does not need to itemize deductions, he is also entitled to the standard deduction of $4,700. His taxable income is $199,532. The trader also has lower taxable income on his California state tax return. All told, the trader will pay $12,415 less than the investor.

Even more savings Had Mike formed a single-member LLC (SMLLC) in his home state in 2002, he could have saved an additional $7,345 if he followed through on the main reason to form a business entity — funding a retirement plan. Forming a single-member LLC (SMLLC) in California costs around $210 and is simple to do. The only entity level tax is a fixed state tax of $800 and a fee depending on the individual’s level of income. In many states, there are no entity level taxes. In any event, these taxes are usually deductible. An SMLLC trader and a sole proprietor have identical tax treatment in many ways: all trading business expenses are reported on a Schedule C; all trading gains are reported on a Form 4797; and all interest and dividend income is reported on Schedule B. ACTIVE TRADER • April 2003 • www.activetradermag.com

63

FIGURE 3 SINGLE-MEMBER LLC TAX RETURN Deductions on adjusted gross income are critical to tax savings for SMLLC traders is the . Deductions for selfemployment tax (line 29), health insurance (line 30) and a Mini 401(k) (line 31) provide significant tax savings.

Providing the retirement plan strategy is properly executed, the SMLLC trader enjoys several tax benefits over the sole proprietor. Figure 3 shows the SMLLC trader Form 1040. In an LLC, the entity pays a “fee” to the individual for administration. The fee — in this instance, $152,320 (the minimum amount necessary for the maximum Mini 401(k) deduction) — is reported on Line 12. This fee, along with $44,700 of trading business expenses is transferred from Form 4797 to the SMLLC Schedule C (both the entity and the individual must file a Schedule C), leaving just $51,538 on form 4797 (line 14), and gross income (line 22) of $204,752. While that’s about the same gross income as the sole proprietor trader (the only difference is the $800 California SMLLC tax deduction), the real savings occur when determining adjusted gross income. While an SMLLC trader is sub ject to self-employment taxes (on the individual Schedule C), half those taxes can be deducted from gross income (Line 29). Plus, the SMLLC trader can deduct health-insurance costs (Line 30) and — this is the big one — up to $40,000 from forming a Mini 401(k). When these three items are deducted, total adjusted gross income is $155,033. These deductions are not available to sole proprietor traders. Taxable income winds up at $147,813, and the SMLLC trader pays $7,345 less in taxes than the sole proprietor — $19,760 less than the investor. Furthermore, those with a Mini 401(k) can actively trade up to 25 percent of the plan’s assets, and all gains are tax-free until the trader retires. Mini 401(k) traders can also borrow money from the plan. For more on Mini 401(k) plans, see “A special ‘K’,” Active Trader, February 2003, p. 96.

The bottom line Trader tax status will save you big money. If you qualify, consider whether an SMLLC and Mini 401(k) plan are wise for your tax and financial planning. Ý Note that the sample tax returns in this article are incomplete. Full examples are available at www. greentradertax.com. For information on the author see p. 3.

64


AFTER

Hours

Mad(den) TV

A way to — BOOM! — improve financial TV

W

hile CNBC, Bloomberg and other trading-related networks still have a significant audience base, it’s possible the broadcasts have lost a little “zing.” After all, reporting on the latest new low set by the Nasdaq, or another batch of disappointing corporate earnings has a tendency to take the pep out of your step. So what’s the solution? Well, these programs need a personality — someone with instant name recognition, someone who’s not afraid to express his opinions, someone who can provide the kind of daily analysis that’s often lacking. Someone like John Madden. OK, so America’s most famous football analyst may not know that much about the markets. But he’s comfortable in front of a camera, and people will tune in just to watch him. Besides, when has knowing anything about the markets been a prerequisite for being on financial TV? We can almost picture it now….

don’t think Alan Greenspan can be called a superstar anymore. But he’s still a student of the game and can definitely make a big impact. GTH: OK, now let’s turn our attention to the Nasdaq. John, this market has been struggling to get back on track recently. What can the Nasdaq do to turn things around? JM: The Nasdaq has had no success in its battles with resistance. That resistance line has slowed Nasdaq’s progress more than once. (Achart of the Nasdaq is shown, and Madden grabs the telestra tor). See, right here is what I’m talking about (Madden draws a resistance line.) Here’s what the Nasdaq needs: The technology sector has to avoid any more bankruptcies, and the telecom stocks need a favorable report on new cell phone sales. If both those happen, BOOM! (Madden draws a line through the resistance line, then draws price bars above it.) The Nasdaq will begin to trade up here and might even be able to string some winning days together.

Generic TV Host: John, as we ready for the opening bell, what should we look for today? John Madden: There’s no question that Alan Greenspan is the Madden continues his analysis throughout the day. We join man to watch. Greenspan is a veteran economist out of him again after the closing bell. Washington, D.C., and he’ll face a tough Senate panel today that will be looking for a little payback. Last time these two met, GTH: Wow, what a performance by the Nasdaq, John! A gain of Greenspan had the Senators scrambling for their economic text- more than 3 percent. Who could have seen that coming? books. This time around, you’ve got to think the Senators will be JM : And what a spectacular performance by ABC Dru g Company! These guys have been practically unknown, but more prepared. when it had that cancer drug approved by the FDA, BOOM! ABC performed so well it lifted the performance of the entire GTH: What makes Greenspan so dangerous? JM: He’s really a triple threat. His testimony has the ability to index! make the market go higher, make the market go lower or just That just goes to show — you never know where your big confuse most people with esoteric ramblings. Plus, what he says contributions are going to come from. The Dells and Microsofts affects short-term traders as well as people with a longer time and Oracles get all the attention, but it’s often a little-known perframe, so he demands double coverage. former like ABC that comes through when it matters most. Plus, he’s 76 years old. He’s played this game a long time. He’s had to work under a few different bosses, but he’s always GTH: Any thoughts about tomorrow, John? JM: You’re only as good as your last performance, so it’s imporstuck to his gameplan. tant that Nasdaq keep this momentum going. And it can’t GTH: But hasn’t time caught up to him, John? Isn’t it fair to say expect to get unexpected help like it did today — everybody has he can’t single-handedly dominate like he did a few years ago? to keep their earnings up and their noses clean. That’s what JM: That’s a good point. A superstar’s time is limited, and I makes a winning index.

ON THE JOB

What does he think he’s doing?

He doesn’t like to sell short, so when he looks at his screen upside-down, it looks like a long trade.

That doesn’t make sense — his trades would be going back in time.

Wait a minute — aren’t we in the wrong cartoon?

TRADE

Diary

A timely entry makes the most of market conditions. S&P 500 index trust (SPY), daily

94.00

Trade

93.00 Short at 92.47

Date: Jan. 16, 2003

92.00

Entry: Short the S&P 500 index trust (SPY) at 92.47.

91.00 90.00 Exit second half

Reasons for trade/setup: The idea behind this trade is related to concepts about outside bars explored in “Outside bars: From concept to strategy,” p. 40. After rallying in early January, SPY moved sideways. On Jan. 15, it sold off intraday after opening at what would be the day’s high price (and above the prior day’s high); and closing below the previous two days’ closes (and making the lowest low in three days). This indicated the consolidation was resolving to the downside. A short sale was made the next day on the open.

89.00 88.00 87.00

Take profits on half the position

86.00 85.00

30

2003 6

13

27

February

Profit/loss: +5.95 (first half); +5.82 (second half). Initial stop: 93.82, 25 cents above the high of the outside bar (entry day). Target: 87.11, the Dec. 31 swing low. We will take partial profits at this level and then trail a stop no farther than above the previous day’s high for the remainder of the trade. Pluses: The long-term trend is down and the market has lost upside momentum over the last week or so. Minuses: The short-term trend is still up and the stop level is relatively far away (1.35 points), compared to the typical Trade Diary position. The market is in its bullish new-year phase; the current consolidation could be a pause before a continuation of the uptrend. Update (Jan. 24): Took profits on half the position at 87.12; stop

lowered to 89.58, 20 cents above the J an. 23 high.

Reason for exit: The trailing stop was hit.

66

Lesson(s): This trade illustrates an important point: Any trade plan or strategy is ultimately subject to the conditions in place at the time of a given trade. The entry, stop and profit target approach used here, which is designed primarily to capture short-term price moves, is very similar to the method used in the many Trade Diary trades. However, in this case, the market moved immediately — and significantly — in the trade’s direction, which allowed the position to become much more profitable than others executed in a similar fashion. There was no way to predict that such a sell-off would occur; the trade rules, however, were able to capitalize on it when it came to pass. This trade could have been slightly unprofitable or moderately profitable, depending on circumstances. Most trades occur in that unspectacular middle ground. Trades such as this are the exception, but it’s still necessary to structure your strategy to capture them effectively when they occur. Ý For a second installment of the Trade Diary, turn to next page.

Result Exit: 86.65 (second half of trade).

Trade executed according to plan? Yes.

Trade summary Date

Stock

Entry

Initial stop

Initial target

Initial reward/ risk

1/16/03

SPY

92.47

93.82

87.11

3.97

Exit

Date

P/L

87.12 (1/2 pos.) 1/24/03 +5.35 (1/2) 86.65 (1/2 pos.) 1/29/03 +5.82 (1/2) +5.58 total

Actual reward/ risk 3.96 4.31


Active Traders - Active Trader Magazine 2003 April Issue

Recommend Documents