View

Trying to Time a System Can be Costly

March 21, 2005

 

What to do when a system is doing well, and (perhaps more importantly) what to do when it isn't doing well are two of the more frequent questions put to us at Attain. Many people like to take a contrarian view and not get involved with a system until it has seen some drawdown, and others have wondered aloud whether it's wise to sit on the "sideline" for a few trades when a system hits equity highs so as to miss any upcoming string of losing trades. Once this losing string is over and the system has retreated off its equity highs, these investors would then jump back in.

A similar strategy among some investors is to call up and put their system on hold before a Fed announcement, big earnings news, or my favorite - "until the Middle East calms down". Many successful investors have made money - or saved it — by following such gut instincts, and this begs the question of whether you should ever be "on the sidelines".

Skipping any upcoming drawdowns is obviously a fantastic idea, but the only way to do that is by not trading. If you're not in the game, so to speak, sure you won't lose - but you also can't win. The simple answer is NO! It is statistically improbable to "time"a system in this way.

Trying to time a system by sitting out certain days is a surefire way to negatively affect your performance. But what about earnings releases, Fed announcements, and special events? Trading systems enjoy the volatility these events provide, and often do better in times of higher volatility. The statistics show that trying to time a system, in attempts to sit on the sidelines and miss losing trades, is a losing proposition.

The crux of the problem is that you never know what tomorrow will bring. The next trade, or string of trades could be the best the system has ever seen. They could also be the worst, but the most likely scenario is somewhere between those two extremes.

To test our hypothesis, we gathered data on the performance of the Compass system over the past 5 years. We wanted to see if the savings gained by "missing" some of the worst strings of trades was great enough to overcome the opportunity cost (missed profits) of "missing" some of the best strings of trades.

Our tests showed that the penalty of missing even a few of the most profitable trades over this time period had a severe effect on profits, while the possible savings were not large enough to justify sitting on the "sideline".

We show our results in the table below by looking at the total return of Compass over the past 5 years (+216.93%) as compared to the total return if an investor had missed the best single trade and best 5, 10, and 30 day strings of trades. We then show the total return as compared to the total return that would have been achieved had an investor missed the single worst trade and worst 5, 10, and 30 day strings of trades.

You can see that anyone unlucky enough to time a move to the "sideline" at the worst possible time - thus missing the best 5 day string of trades in the Compass system - would have seen the total return drop from 216.93% to 177.04%, or about $12,000 less in profits.

Conversely, if an investor had timed a "sideline sitting " perfectly and missed the worst string of 5 trades, the total return would have only improved from 216.03% up to 237.50%, or about $6,200. So an investor who went to the sidelines for 5 trades would have had at best saved himself about $6,200 and at worst cost himself about $12,000.

That is not the kind of risk/reward successful investors take on, and neither should you. One look at the near 100% difference between the total return and the total return if missing the best 30 day string of trades, 213% vs 121% should make you a believer.

The bottom of the table below displays the max "penalty" investors would have incurred (in the form of missed profits) on the left and the max "reward" (in terms of missed losses) investors would have saved on the right - for different periods of consecutive trades (strings of trades in 5,10, and 30 day periods) The table shows that the risk is much greater than the reward at every level, as the opportunity cost of missing a good trade far outweighs the potential benefit of missing a bad trade.

This characteristic of trading systems should not be surprising, as the majority of systems are designed this way. In the terminology of America's baseball pastime, trading systems hit "Home Runs", not "Base Hits". So they are constantly trying to survive the less than perfect conditions while waiting for that one big trade.

They are deliberately designed to lose a small amount of money on losing trades, and make a great deal more money on the less frequent winning trades. In fact, the majority of profitable systems lose more often than they win. How can this be? Because they make much more when they have a profitable winning trade than when they have a negative losing trade.

The investor attempting to miss some of the "strikeouts" will undoubtedly miss a couple of "Home Runs" in the process.

- Jeff Malec

IMPORTANT RISK DISCLOSURE
Futures based investments are often complex and can carry the risk of substantial losses. They are intended for sophisticated investors and are not suitable for everyone. The ability to withstand losses and to adhere to a particular trading program in spite of trading losses are material points which can adversely affect investor returns.

Feature   |   Week In Review   |   Chart of the Week   |  

Chart of the Week : Risks/Rewards of Timing a System

Feature   |   Week In Review   |   Chart of the Week   |  

The third week of March was a tough one for US stocks as the S&P futures closed -1.21% lower and NASDAQ futures dropped -1.75%. The swing traders generally enjoyed the sell off as they rode out the storm by trading infrequently last week. The Tzar and Axiom continue to pace this sector of systems.

Trading conditions in the commodity markets were more active than normal last week. Energies continue to be the most volatile market, with crude oil futures rising +3.85% last week remaining the focus, but the usually complacent grain markets have also come alive lately.

The grain markets rallied significantly last Tuesday, only to see sharp reversals on Thursday and Friday. For the week soybean futures were hit the hardest falling -2.04%, while wheat and corn remained almost unchanged.

**Day Trading**

As a general rule, systems that were able to capitalize on Tuesday’s sharp decline were profitable for the week, while those that didn't capture Tuesday's move saw losses. Leading the pack was Magnitude SP, which profited $2425 for the week, the bulk of which was made on Tuesday. R-Mesa tacked on profits for the second week in a row, making $2350 on a single short trade from Tuesday, while Helix SP strung together several winning trades for the week to tack on a total of +$2525 per contract.

BWT Rock N Russell had quite an impressive week, making $1201.60 per money management unit using the position manager, while AG Xtreme continues to perform well in 2005, returning $667.50 for the week. Compass didn't do much after a big winning trade on Tuesday, making $331.75 for the week.

Elsewhere, the founder suite of systems: Magnitude ES, Helix ES, and Cipher ES were all profitable for the week, making $460, $392.50 and $195 respectively. RC Miracles ES was very active last week and posted gains of $625 per contract, but related RC Success ES found the going a little tougher - losing -$800 for the week. Rounding out the profitable systems was the Electric Day Breaker portfolio, which made $195 for the week.

An exception to the aforementioned rule regarding Tuesday was BWT Zones SP and BWT Zones Russell. Both systems took profits on Tuesday but finished the week in the red, giving back $1025 and $902.20 per contract. Finally, Clipper eRL struggled with the choppy conditions of the emini Russell market last week, giving back $434 per contract.

**Swing Trading**

The short term down trend in US stock indices ha been a welcome environment for most swing trading index programs.

Last week's trading included long reversals in Tzar ES and NQ and short entries in Axiom and Eclipse. Tzar ES lost -$210 and is holding long with open equity losses of -$1240 while the NQ locked in $50 and is holding for a loss of -$410 per contract. Tzar eRL is also holding long from the prior week with a loss of -$440.00.

On the positive side, Axiom NQ locked in $50 and is holding short in all 3 other markets for open trade profits of $3,120. At this point the system has begun to trail its profitable positions and in the event of a market rally we can expect to see it attempt to lock in profits and even reverse. In a similar strategy, Eclipse eRL had been holding short from the prior week only to see its trailing stop elected on Thursday for a small gain of $5.50 after accounting for $30 in commission.

Finally, last week saw the culmination of three month trade in Mesa Notes, as it was stopped out of the long position it has been holding since late December. At one point the system had been ahead approximately +$1,359; but a 20 day span where 10 yr Notes lost -3.02% caused ten system to lose -$2,545.40 per contract The system went long again later in the week as it attempts to capitalize on the counter trend movements of the bond markets.

And finally, Attain officially pulled the plug on the I-Master system last week, recommending anyone still trading the system to shut it down in light of its eclipsing its pre-tested 1 in 100 yr drawdown levels, pushing us over the line in the sand many investors had set as system stop loss.

**Long Term**

In addition to the large moves that traders saw in the energy and grain markets, the softs markets were on the move as well. These markets comprised of commodities like cotton, coffee, sugar, and cocoa are also known as tropicals. Cotton and coffee took center stage in this sector last week as cotton futures dropped off -3.25%, while coffee futures moved -2.52% lower.

Most systems have avoided coffee trades altogether as the market has been much to volatile and risky. However several systems have cotton positions on, including Andromeda which is long for open trade profits of +$1640.00 per contract, Brix which is long for open trade profits of +$875.00 per contract, Synergy which is long for open trade profits of +$100.00 per contract, SEMA4 Symmetry which has open trade losses of -$480.00 per contract, and Checkmate which is long for a open trade loss of -$1075.00 per contract.

There's an old theory that once a trend hits the cover of "Newsweek", its time to get out - and that appears to be what's going on with the dollar as we continue to bounce off the lows amid daily reports on why the dollar could go lower. Last week both Eurocurrency and the Swiss Franc moved -1.05% lower as the dollar began a slight comeback. Other currency markets including the Australian Dollar, Canadian Dollar, and Japanese Yen markets moved slightly lower.

Systems with currency trades include Aberration which is long the Mexican Peso for an open trade loss of -$625.00 per contract, Brix which is long the Swiss Franc for open trade profits of +$262.50, and Checkmate which is long in the British Pound for open trade losses of -$250.00 per contract.

Please Login to: http://www.attainaccess.com for the latest updated statistics.

IMPORTANT RISK DISCLOSURE
Futures based investments are often complex and can carry the risk of substantial losses. They are intended for sophisticated investors and are not suitable for everyone. The ability to withstand losses and to adhere to a particular trading program in spite of trading losses are material points which can adversely affect investor returns.

Feature   |   Week In Review   |   Chart of the Week   |