Successful Trader Secreats
Linda Bradford Raschke is one of the successful Trader in America. Contrary to popular opinion, dreams can come true. For every dozen traders who’ve wished that they could just stay home and trade, there’s one trader who succeeds in doing so. Linda Bradford Raschke began her trading career in 1980 on the floor of the Pacific Coast Stock Exchange before moving to the Philadelphia Stock Exchange. After six years of trading on the floor there, Raschke left to trade her own account from home. Eventually, her focus evolved into developing reliable technical patterns for generating trades and then following 20 different markets and waiting for the most lucrative trades to come to her. She has always been a consistently profitable trader using her research. So how does she do it? She’ll tell you: Raschke shared her insights with STOCKS & COMMODITIES Editor Thom Hartle on June 18, 1993, via telephone.
I’ve been trading for about 13 years. For about the first three I hardly knew what I was doing, other than using simple oscillators. I didn’t know how to be discriminating in my choice of indicators. I was still finding my own way. It took three to four years before the pieces finally fell into place. —Linda Bradford Raschke How long have you been trading?
I’ve been trading for about . 13 years. For about the first three I hardly knew what I was doing, other than using simple oscillators. I was working in San Francisco and there was an excellent business library there that had all these wonderful books that are hard to get hold of nowadays, books on volume analysis and Equivolume and this and that. I was constantly working on becoming a better trader.
And you were studying up a storm?
I should clarify what I just said. When I say I didn’t know what I was doing, I didn’t know how to apply the knowledge I did have. I didn’t know how to be discriminating in my choice of indicators. I was still finding my own way. And even though I was trading full time and studying all this full time, it took three to four years before the pieces finally fell into place.
I learned to wait for specific setups. If there was a trend, I would wait for buying during a pullback to a moving average. I’d use an oscillator and time my trades. I’d also trade based on a certain divergence pattern that happened, with a double bottom occurring l0 days apart (Figure 1). So I didn’t try and immediately fit the whole market into a framework of technical analysis; rather, I would focus on specific patterns that occurred with a f air amount of regularity, and then try to take advantage of those patterns. And that’s pretty much what I do today with my technical analysis.
Were you active in the commodity markets?
No. For the first eight years of my trading, all I did was trade the stock market. Granted, I was looking at different stocks, but I didn’t have any familiarity with trading commodities.
How did you gravitate toward the other markets, then?
I got frustrated in the late 1980s when the equity options lost a lot of their liquidity, and I also got frustrated when the margin requirements on the OEXs changed, because I liked holding short premium positions. Trading equity options became an inefficient use of capital. So it just seemed like a more efficient use of margin to move toward the futures market. I’d always traded the financial futures, and it was a matter of seeing if my systems and methodologies worked on wheat and crude oil. It came down to a numbers game.
Well, meaning if you have only two great conditions setting up a month, if you look at 20 markets as opposed to six, you just might have that much more opportunity. But it truly does come down to a management function. I can look at 20 markets now because I’ve got it pretty systematized. But it took me five years to build up to that. At first I could handle six markets and then I could just handle 10. And I still can’t really handle it if I have more than seven or eight positions on at a time.
So you feel stress when you have multiple positions on?
No, I don’t feel stress when I have a trade on. I don’t feel any stress at all whether I have trades on or not. It’s just the amount of work involved — it’s just that much more to manage. What I’ve really focused on is the management part of trading, trying to do it as easily as possible.
Can you give me an example?
It’s easier to monitor three positions with an 85% probability of working out than 10 positions with a 65% probability of working out. That I have a trade on doesn’t cause me stress; I enjoy it. I enjoy having more positions working in the marketplace. I’m just trying to pace myself because I ran into a problem about three years ago when I hit the wall. I just don’t want to overextend myself.
Do you analyze the markets during non market hours?
Always. I always do my analysis the day before the markets open, so that’s either after the close or at night. I don’t want to be influenced by overnight market activity or reports coming out the next day. My basic thesis has always been that the market’s going to do whatever it wants to anyway. And I like to come into the trading day fresh, so I don’t have to scramble in the morning to get my game plan together.
What would be an example of your game plan?
Well, I have an order sheet. I write down the previous day’s high and low, because to me those are the most important reference points when I’m entering a position the next day; I like to know where the price today is relative to the previous day’s high/low. Sometimes I’ll be looking to sell around that high if I’m in a swing-trading mode. Other times I’ll be looking to buy a breakout of that high. So I program myself to either being in a breakout or countertrend mode. I need to determine the day before whether I’m going to trade that market in a breakout mode or in a swing-trading mode.
I also need to determine which markets I’m going to concentrate on buying or selling. Essentially, I analyze 20 markets every day, but it’s impossible to trade 20 markets every day. So if I concentrate on selecting the three or four or five best trades for the next day, then what will actually happen is maybe half of those trades will set themselves up beautifully and I’ll be able execute those.
What about the other half?
With the other half, maybe, I might be looking to buy a market, but say it doesn’t hold the previous day’s low and it breaks hard and has continuation. Then I’ll just let that trade go.
Your trading is separate from your analysis. Do you hold the view that there’s a difference between being a technician and being a trader?
I think there is a difference. You don’t have to be good at technical analysis to be a good trader and vice versa The two aren’t necessarily exclusive, but just because you’re good at one doesn’t mean that you’ll be successful at the other. They’re both helpful to each other, though. If you ‘re good at technical analysis, it will certainly help your trading. I always use technical analysis to outline the basic framework that I trade in.
Do you have a specific way of analyzing the markets, then?
Well, what technical analysis does is support the framework that gives you the statistical edge. First of all, parts of it can be quantified so you’re always working within a probability matrix. This is what I try to set up for myself. If I’m trading systematically, I know that I have a probability matrix that says this is 80% correct, this is 20% incorrect, this is a high probability trade, this is a low-probability trade.
And all that is based upon technical analysis and supportive research?
Right. It’s also a tool to draw you in closer to the markets, to get you more involved with the markets, which you have to be when you’re trading. You might just get that occasional insight or glimmer.
Essentially, I use technical analysis to arrive at my decisions on whether I’m going to buy or sell, and then my trading skills along with some common sense take over to manage the entries and the exits and the risks.
Has most of the technical analysis you use been backtested, then?
I would say that the majority of the technical analysis that I use is through observation and experience. Some things are difficult to backtest because in a pure sense, computers are going to pick up certain things that you may not be looking at or other things that you don’t feel is relevant to what you ‘re looking for. Technical analysis is really an art, combining several different indicators to confirm a conclusion. It’s hard to quantify that because one decision may be based on oscillators and volume and open-interest patterns, while another decision might be based on the degree of trend in oscillators. That’s where the art comes in: learning to combine different things for confirmation. If you’re backtesting, you can isolate specific patterns and backtest those. It’s a useful idea- but conditions are never the same in the real trading world.
Well, hardly ever. I could backtest a stochastic I crossover or an inside day or a certain price pattern, and what’s going to happen in the real world is never going to work out the same way. There might be a report in the morning, there might be different volume characteristics or there might be different intermarket factors in the real world that will make a difference.
Well, the markets are always evolving.
Sure ! Everything repeats itself to some degree, but nothing repeats itself exactly. You have to allow that conditions can change.
That’s where the common sense comes in then, the application of the analysis?
What are your favorite inidcators that you use on a regular basis?
That’s evolved over time too. I still use the very basic indicators that I used for the markets starting in the early 1980s. These are a start for the framework. I use other indicators for short-term timing, but for long-term market timing, I was originally trained to use the Arms index and 10-day advance-decline, the McClellan oscillator, 10-day closing ticks, up/down volume patterns, put/call ratios and specialist shorts. I still track these everyday, and I have for the past 13 years. So I’m real familiar with their nuances.
What else do you use?
When I first started trading I was using a charting service called Security Market Research (SMR), which essentially had a simple oscillator that was based on the difference between two moving averages and a longer-term oscillator. So right off I was very countertrend-focused. Oscillator systems tend to be that way, and I learned patterns within those oscillators — such as if there is a trend, buying the pullbacks to the moving average, confirmed with an oscillator — simple things like that. In addition, if there is a trend, don’t fade the trend until you have one or two divergences. A lot of my trading rules have come over observation and trading practice and learning the hard way. What works and what doesn’t.
So trade selection is the key?
Yes. I’ve found that over time I’ve become more and more selective about my trades.
Did you find that you hod your losing trades when you weren’t being selective?
Not necessarily. I used to be a much more aggressive trader, and I was extremely profitable. I still am, but it wears you down. Those extra trades are fun, but incrementally, they don’t add that much to the bottom line. If I’m taking 85% probability trades, maybe they’re making up 90% of my trading revenue. And if I took an equal amount of 65% probability trades, I would be more profitable but my profitability per trade would drop.
Let’s say in the morning you were coming in with a buy attitude, and the market set up for you to make the trade. You then find yourself long but the market is not rallying. What would be an indication to you that you should abandon the trade?
Well, if it’s a short-term trade, to me a short-term is one to two trading days, and I’ll give it a lot less leeway. If I have a strong scenario for a longer-term pattern setting up, though, I would definitely give it a lot more room. So it depends on my initial reasons for buying or selling in the first place. What’s important is you have to have this already determined before you enter the position.
No changing your mind after you start?
If you’re buying because there’s a one to two day pattern and it goes your way, you should be exiting in one to two days, not all of a sudden make a case of that turning into a one- to three-week scenario trade. Normally, if it’s a one- to two-day trade, I won’t give it very much room. If it closes with a profit I’ll carry it overnight. The last hour’s action is very important to me . I want to see how a market’s going to perform at the end of the day, and if I like how it acts at the end of the day I will continue to carry the trade.
Do you look at intraday charts?
Mostly, I watch price. I find that if I want to buy, I’m buying because of some daily setup, and if I start looking at five minute charts or 30-minute charts, it might influence me the wrong way.
You’re trading for a predetermined set of circumstances, so you have to concentrate on your plan.
Right! In other words, a five-minute chart might be correct for the next hour, but I’m not there to trade soybeans on a five-minute basis. I want a good entry for that day because I have a pattern that shows me they’ll close higher in three days, but if I start looking at five-minute charts I’ll either miss my entry by waiting for a pullback or I’ll want to get out too soon. So I try to stick with what I determine to be the “true” trend, as I call it.
In other words—
In other words, if there’s a cycle, a five-day cycle in the market where a market goes from highs to lows every two and half days . That’s a really crude model . But I try to keep all my models really simple and basic to trade within. The simpler the model, the easier it is to trade and recognize when the price is going inversely to your model. But essentially, I buy one day and hold for a day or two days, and then look to reverse.
So the intraday activity on the charts isn’t what you’re looking for.
Well, they might set up on hourly charts, nice hourly cycles, but if I just think about buying something and holding it for two days, I don’t get stressed out by all that intraday noise. And it does take a little bit of confidence. That may come with experience, and if any of your readers are interested in trying this, it’s always useful to paper trade for about 20 trades, so that you can get the feel that, yes, the price will hold together for two days.
So let me get this straight. For a three day trade, you might have a trend one day, the next day there’d be a consolidation of that trend, and then a follow through trend the next day?
Most likely what’ll happen is that the first day the market makes a test of the previous day’s low. Then the middle day is when it’ll have a trend day, and then the third day it peters out. Usually the middle day tends to be the big day, so if you ever get a middle, or a good trend day, or a good movement that day you should always carry that trade overnight because the odds are that you will get follow through the next day.
After you’ve had a trade like that, you go back and continue to do your analysis looking for the next trade?
Exactly! It’s a lot of work each night, but when you trade on a short-term basis, you’re constantly having to look for new setups. I’m always taking profits or losses and the next day I always have to look for anew setup. So if I haven’t looked for my setups the night before, I come into the trading day and I end up not making any trades because I’m really not prepared. Sometimes I do have longer-term positions on—you know, two- to three-week positions, where I sit back and let them work for me. But that’s the exception.
What would be a setup for a trade?
I like to have what I call a library of patterns. Each of those patterns is a specific trade with specific rules. A good example might be buying the breakout of an NR7 day, which stands for a narrow range seven-day pattern (Figure 2). That trade would be going with the volatility in that case.
What about another pattern?
Another pattern might be looking for a retest of a previous price level that was climactic and looking for a trend reversal, and that would have specific rules both for stops, exit and entry. What happens over time and trading a number of years is that you start to build up a library of patterns and each one of those has some specific rules.
Other patterns that you like to look at are based on your indicators. Can you expound on that?
I use different measurements for the amount of contraction in the market. There’s a number of different proxies you can use, but everybody should have some method of measuring the amount of contraction in the market. Many times, oscillators or momentum indicators will lose their oscillations or extreme readings . This is what I like to measure, and that tells me to switch to a breakout mode.
How do you do that?
It’s an observation thing. I’ve just conditioned myself to pull up a chart and glance over it in two minutes. I do it by visually inspecting the oscillators; you’ll notice that you can almost do it by visually inspecting the chart pattern. You could study a chart and you’d notice a triangle.
So you prefer the indicator over the chart?
Looking at the oscillators instead of the patterns tends to be a little more objective for me. I’m one of those people who can look at a chart and see what I want to see. If I’m long, I can see all the patterns as to why that chart is going to go higher, and vice versa if I’m short. If I just look at the hard-core oscillators or technical indicators, it’ll keep me more objective.
Which oscillators would that be?
Well, there isn’t a right one or a wrong one. I use the same generic parameters for everything. There isn’t one optimal set of parameters. I use a moving average convergence/divergence (MACD) histogram, for example; I like using a particular combination- some of your readers may be interested in this if they’re into stochastics- I use a %K with seven periods and a %D with 10 periods, so that I have a really long-term stable oscillator.
What do you use for the short term?
I like short-term rates of change — you know, two-, three-, five-day rates of change. Just simple rates of change, not smoothed. I like an oscillator that’s a difference between a three- and a 10-day moving average, with a 16-period moving average. That one’s been around for 20 or 30 years; in fact, I know it’s been around for 30 years. Any of these will show the same type of price action. I can do equally well with one as I can with the other. It’s just a matter of getting used to looking at these oscillators to highlight whether there are cycles in the market. And you can get a feeling about that by looking at the slope of these indicators, whether there’s market acceleration.
Do you combine your short-term indicators with the long-term indicators?
I strongly believe that short-term indicators precede long-term indicators. So I look at the market mostly on a short-term timeframe, because short-term momentum precedes long-term momentum. And momentum precedes price. Usually, trend reversals or the beginning of new moves start with an increase in the short-term momentum. You’ll notice a lot of times there’ll be a big break or two or three big price bars, or a break to the downside with a couple of big price bars. Being aware of this makes you realize that you can wait for the market to tip its hand and have that short-term momentum indicate what the market’s going to do, and then wait for the first two- to three-day retracement to enter the market.
How often do you see setups based on the oscillators?
About 70% of the time I don’t have a clue as to what those oscillators are telling me.
A lot of the time I really don’t know which way the market’s going to go, but I just wait for something I recognize, and then I’ll act on it. It’s really a patience game, waiting until you see something that you’re familiar with. It doesn’t necessarily mean that you have to be trading in the market for 10 years; it means you’ve got to study charts or previous oscillator patterns to get your own feel for it.
Oscillators can steer you wrong in a strong trend. What do you do about that?
That is true. One thing that you have to be careful of with momentum oscillators is quantifying the degree of trend in the market. I use the ADX indicator to quantify that.
What’s your rule of thumb there?
I avoid the markets where the ADX falls below 20 and is declining, because there’s just not enough movement between the swings for me. Likewise, I have something of a golden rule: I will not fade a market. I will not try to trade against a market where the ADX is above 32 and rising. If I just keep it to a very simple little rule like that, I know I won’t try and fight too strong of a trend. What do you look for as a warning that a trend may be stalling?
If there is a trend, I never try and go against it unless I have a clearly defined double top, or what I call a
double stop point. I have to see that the market has twice failed at a certain point before I will go against it. If there is a trend, I’ll try to always look for buying a pullback to a moving average (Figure 3). I personally use a 20-period exponential moving average on all my markets. But what you’ll find is that there really is no right or wrong parameter for a moving average.
Do you use orders with price limits, or do you trade with market orders?
I use market orders most of the time.
So you don’t quibble about paying the offer or hitting the bid?
No. I’m more interested in making sure that I get my position established. I believe I’ve already got a couple of minutes’ delay against me anyway by the time my order gets to the floor. I ‘m never going to know exactly where the price is. I actually did a study once. I wrote down all the prices, how much money I saved off of, let’s say, for example 10 transactions, and nine out of 10 times I found I could get a better price if I waited a little bit.
What about the 10th time?
The 10th time that you miss the trade is the one that could make your month. So it’s very important not to miss the big ones. You should have a setup that’s strong enough that it really doesn’t matter if you get in plus or minus a couple of ticks. What you should be doing is seeing more in a trade when you’re trading from upstairs. Even though I’m on a short-term timeframe, two to three days, I try to look for something where I can see a couple points’ move, or I just won’t play it.
We’ve only been talking about your work with classic analytical methods. Have you looked at some of the more current topics?
I use a chaos model. This model can explain some situations that occur in the market. For example, say there’s positive feedback in the market. That may be a market that’s making new highs or new lows in a downtrend (Figure 4). These are situations where professionals cannot set price limits. In a runaway market, or a market where there’s deterministic chaos, where you have an incremental gain in each stage, nobody can say how far that market is going to go . I don ‘ t care if you ‘ re using Elliott wave or Gann analysis or whatever the measuring method may be, nobody can say with any statistical accuracy how far that price is going to go. And things will usually go farther than you think; they usually overshoot targets. So you have to be aware of those situations in the market. This chaos model helps me understand that I’ve got to be on board those situations and not fight them.
The chaos model — is this a kind of generic example of what can unfold in the market?
It’s almost as if you were to look at a chart and be able to normalize the price movements, or normalize the cycles in that. That’s what it would look like. It’s a mathematical way of representing the degree of contraction in the market, or the lack of oscillation in the market.
So how does that work?
An example from a classic analysis approach would be a group of indicators of rates of change, and the absolute mean deviation starts to approach zero where you have a neutral point: those are the best conditions that precede “a chaotic event” (Figure 5). Those are the conditions where you have this deterministic chaos, or these severe breaks. This happens because all these cycles have reached such neutral points that wherever the market goes, it’s going to pull all the cycles up or down together.
Do you actually in a sense trade off this knowledge, or do you think this knowledge just gives more conviction about taking the trade when you know these events can occur?
It gives me more conviction and more confirmation and, at the very least but probably most important, it serves as a filter for when not to fight a move. If we can just filter out those losing trades, you know, we’d be ahead of the game there.
Then this is the kind of ideal trade you ‘d be looking for. One example of a good trade setting up is when the market’s all settled down into—
Ideally, actually, the nicest thing is to be on these moves before they occur. It’s difficult to get on board because the markets move very quickly. But there’s always a tradeoff here, which is that it’s very difficult to predict the movement out of these points.
Exactly! It’s like a triangle. It really becomes a 50/50 bet. So you either have to have some stronger supporting longterm structure or you have to use extremely good risk management. If you are wrong you cannot be stubborn, you must learn to immediately recognize that you ‘ re wrong and get out. How will you recognize that you’re wrong? Because you’ll take a loss. That’s how I tell if I’m wrong or right, either the price is moving in my favor or it’s moving against me.
Well, do you place, say, resting stop-loss orders, or do you pick up the phone and initiate the action?
I pick up the phone and I say, “Get me out of the market!”
And you go look for your next trade. What are your thoughts about using statistical analysis of price for selecting trades?
What I’ve found in trading is that a lot of the people create statistical models, and they take a regular, standard deviation, bell-shaped curve. Fine. But I think people underestimate the tails. The tails are a lot fatter in real life than people realize, especially with their risk management. The odds of that 2% occurring in real life actually ends up being about 5%.
Something along the lines that Murphy was an optimist?
People underestimate the frequency with which the worst-case scenarios can occur. And they have to account for that in their trading models. I trade very conservatively; that’s probably why I’ve been around for so long.
What can go wrong, will.
Right! You have to always be determining what would happen under the worst-case scenario: Can I tolerate that? The same is true with a lot of short-term trading systems; you’ll find that maybe 90% of your profits are made from 10% of the trades. See? So a lot of times with these systems you’re treading water, taking small gains, taking small losses, waiting for really that one or two times a month when you get lucky.
Surely it’s not all luck!
Well, it’s just a matter of getting lucky by playing the game, and luck is a residue of effort. So you have to continue to play that game. A classic example would be soybeans. Two days ago I got lucky, because I bought those beans because there was a really nice 10-day divergence pattern — the market sold off for four days, and then tested the low and held and had the first day up. I just bought a position because that was a pattern I’d seen a lot of times, but I really got lucky in that the beans happened to open up the next day 15 to 20 cents higher. So you keep playing the game, and occasionally you catch a big ride. In trading systems, traders shouldn’t look for an equal distribution of wins and losses.
So luck can be a residue of effort.
That’s the point I would like to make. Trading is work, and if you work at it, learn the basics and apply yourself, then success will follow.
Thanks for your time, Linda.
Sure! I really got lucky in that the beans happened to open up the next day 15 to 20 cents higher. So you keep playing the game, and occasionally you catch a big ride. In trading systems, traders shouldn’t look for an equal distribution of wins and losses.