Tuesday, May 08, 2007

Why Short-Term Traders Lose Money

Chris (fictitious example, but real trading results) is a trader of the S&P 500 market. He is disciplined and selective in his trades. He limits his trades to those that go with the longer-term and shorter-term trends. If the market is not trending in his direction, he does not trade.

Seeing a long bias to the recent market, Chris has developed a trading method that is long only. He looks at the Bollinger Bands (volatility envelopes) surrounding the 20-day moving average for the S&P 500 Index (SPY). If we close above the upper band/envelope, he buys the market on close. If we return to the area between the bands, he exits the position to limit losses. His idea is to participate in strong trending markets.

Chris averages 3 days per trade as his holding period. Over the past two years, his method has provided him with 18 trades. He has never deviated from this approach during that time.

Yesterday Chris tallied up the results from two years of trading. He was profitable on 3 trades, unprofitable on 15. He made a total of 1.34 SPY points (13.4 ES points) from his profitable trades. He lost 10.30 SPY points (100.30 ES points) from his losers. Overall, Chris lost about 90 S&P points trading long positions in a bull market with perfect market discipline and psychology.

He had an effective system--if he had traded it in reverse. Traders need to understand how markets move on *their* time frame. Much losing of money among short-term traders can be traced to the mistake made by Chris. What occurs on the longer time frame and at shorter time frames can be quite, quite different.


aspTrader said...

This is an extremely important point Brett.

In recognizing the importance of trading with the trend, it is essential to understand which "larger trends" are important to pay attention to and which ones should be ignored given the timeframe being traded.

It is extremely easy to become enchanted with the trend of a larger timeframe without realizing that THAT timeframe trend is too large to be helpful and may in fact be harmful to one's trading.

If only I could have the dollars back that this mistake has cost me...

Dr Bruce Hong said...

Hi Dr Brett
I have a question about your trader. Aren't Bollinger bands frequently used to play "reversion to the mean" trades. In your theoretical trader's example, he would wait until the SPY moved to the lower BB AND THEN STARTED TO REBOUND, before entering a long position. He would wait until it crossed the middle band, move his stop there and exit at the upper BB.

Looked at it this way, and it appears that your trader may have been disciplined in following his trading plan - but he had misinterpreted the theory behind the BB's and had chosen the wrong strategy or application.

As you point out, reversion to the mean is much more prevalent that out-and-out breakouts. As your trader found, to his misfortume, 83% of the time, he experienced a reversion to the mean.

"He had an effective system--if he had traded it in reverse".

weightoftheevidence said...

Bollinger bands are made to come back into band ...all he had to do was backtest it .... he never would have traded

Dr Bruce Hong said...

I forgot to mention, "reversion to the mean" does not refer to a counter-trend or trend-changing pattern. Merely a normal behavior within the context of the larger trend. Bollinger Bands allow for that kind of behavior without invalidating the initial trend.

pete maurer said...

Hello Dr. Brett,
This last post encouraged me to inquire. I'm just retired and want to follow an inclination to learn futures but equally important is to pre-qualify myself as a worthy candidant for trading so as not to waste time or money.

I'm just learning Market Delta and now have become interested in the Market Profile distribution process found at the Cisco-Futures.com. My question is do you think that using the tools as given on your weblog along with Market Delta is a reasonably adequate beginning for a novice given my prerequisite in the first paragraph or if having the time to commit to more extensive studies, such as given by Cisco-Futures, would offer better promise for a more solid beginning?

Love your committment to learn, grow, and share.

Charles said...

Here is another interesting statistic concerning a nice uptrending ES market of the past 12 months.

I looked at how often the ES futures closed above or below the previous day's High over the past 12 months. Both the Globex 24 hour time-frame and the Open Pit time-frame gave similar results.

The ES futures closed BELOW the previous day's high 70% of the time. On average, the ES market closes above the previous day's high for two days, and then closes below the previous day's high for four consecutive days. We can call this the "Swing Trade Cycle".

This can best be seen on the daily chart by using a 10-day moving average. The ES futures seldom move very far from the 10-day average. Two days up, and four days down.


Brett Steenbarger, Ph.D. said...


John Bollinger describes a variety of ways to utilize bands, one of which is counter-trend. My point in the article is that chasing strength--even in a bull market--can lose you money if you're trading a time frame that doesn't demonstrate continuation. Trading psychology is worthless if you don't understand how markets move on your time frame, as aspTrader indicates.



Brett Steenbarger, Ph.D. said...

Hi Pete,

Thanks; I don't know enough about specific courses to be able to comment on their efficacy. Much depends on the needs of the individual learner.


Brett Steenbarger, Ph.D. said...

Hi Charles,

Thanks for the observations; worth looking into!


steve said...

Hi Brett...Great Blog

Now if your trader had simply backtested his idea, he would have found out in five minutes that it would not work. Instead he burned 2 years and 90 S&P points.

(BTW, this could have been back-tested by hand as well as by software)

I would not call what he did as disciplined. Maybe more like stubborn.

Brett, I would like to hear your views on this question of disciplined vs. stubbornness.

Steve Geringer

S Benard said...

I use Bollinger Bands as the core of my trading. This trader, if he correctly understood them, wouldn't have traded the way you described. Even in a trending market, more than 80% of the time, prices are contained within the bands. I use an EMA and the bollinger bands as a channel to trade between. That's an overly-simplistic description, but that is the essence. There is no way he could have made money taking trades ONLY when the bands are exceeded, while exiting when prices come back within them. He would be buying just when prices are temporarily overbought, and selling just as the heat comes off and prices are ready to surge again.

I was surprised that you seemed to suggest that a trader that takes so few trades is a "short-term" trader. I probably trade more in a few days than this trader did in a year!

TIP: The best book I know of on using Bollinger Bands was written in French by Phillippe Cahen. Even John Bollinger mentioned Mr. Cahen's methods in his own book, which is also very good (but not AS good, INHO). An earlier version was published in English (I have that one, too), but the English method was dated and less effective. The title of the later book is, "Analyse Technique et Volatilite". I translated it into English over 8 weeks and delivered it to Mr. Cahen, but it was never published in English as far as I know. At least Mr. Cahen never told me so -- nor thanked me for the translation.

Steveo KayakDiver said...

So, some basic information on backtesting is requested. I can imagine that backtesting by hand may be so tedious that only 2 out of 100 traders would bother to do it. I wouldnt want to even though numbers are in fact "my thing", super simple strategies, may be doable. So then I would ask...what softwares or trading platforms can accomodate backtesting in which you dont need too much training to particpate and yet have confidence in the results?