From the September 01, 2007 issue of Futures Magazine • Subscribe!

Using one time-based trading

Strict mechanical traders don’t allow emotions to contaminate their order execution, but that doesn’t necessarily mean they feel comfortable about what they’re doing. In some ways, systematic traders are more psychologically vulnerable than their discretionary counterparts.

When reactive traders don’t like what’s going on, they merely alter their strategies, often in the heat of battle. Properly disciplined mechanical traders don’t have that option. They must stay committed to rigid plans, which in bad times can be a stressful and financially draining challenge.

Many, if not most, developers are on a constant trek to invent new systems or improve existing ones. Much of the latter endeavors are borne out of in-progress punishment. What could be done to keep a given losing trade from happening? Or, at least, how could its impact be lessened?

An example would be finding yourself short through a series of up-days in the stock indexes. The experience would be unnerving in many ways, one of which would likely be that you start questioning whether you could improve the situation by making spontaneous adjustments. (Hopefully, it’s a short-lived inkling.) But, still, it’s natural to wonder if you should keep punishing your trading account with such an obviously wrong position.

That begs the question, however: What exactly is “wrong” about it? Such are the type of irritating questions mechanical traders are forced to ask. Of course, it’s easy to see that you’re losing a substantial amount of money. The hard part is alleviating that.

Any system trader knows losses are inevitable. The whole game is finding some way to come out ahead after all the gains and losses have been thrown into the hopper. The long run is all that matters while any lone trade in progress is pretty much meaningless.

Not everyone agrees with that, of course. One of the most memorable razzle-dazzle charlatan trading commentaries you’re likely to come across goes something like this: “How do you know you’re wrong? Look at the last tick. If you’re long, is it lower than where you bought it? If it is, you’re wrong.”

The implication of this statement is to liquidate all losers instantly, as if you were scalping ticks in the trading pit. Of course, we’re not pit traders and we’re probably a tick wrong the second our trade is executed because, as upstairs traders, we’re knowingly accommodating someone else’s bid or offer. In other words, if we were to cry uncle at the first sign of loss, we’d have to lose most trades the second we open them.


But our concern now is with the other extreme, not a mere tick of slippage. That being, having lost a significant chunk of trading capital on basically one market move. That conjures up another question once posed to this systematic trader by a prominent floor trader: “It really takes you that many points to figure out you’re wrong?”

Well, yes. Most of us are not blessed with a natural floor trader’s innate radar.

Everything a good system trader does must be tested and proven. After each close, we might stare at the charts and columns of numbers and wonder how we could be applying more effective mechanical tourniquets to the constant flow of price action.

Here’s one idea that came out of just one such session of market study and self-contemplation. There are days where you’re wrong in a position from minute one and, as the day progresses, the situation keeps getting worse. Yes, this is a feeling, but hunches and feelings are fine at the brainstorming system-planning stage, however, they are not fine in actual trading. Quantifying such notions, however, is not as easily done as the “throw-your-one-tick-loss-away” guy would have you believe, but sometimes you do get lucky.


Testing has confirmed this reliable system for filtering out many big losing days that also tends to correspond to those trades that just plain feel wrong. As with any legitimate filter, the system does well standing on its own. The logic is pretty simple.

If, in the first half-hour of a regular pit trading session, the open is within the bottom 10% of the range, and if the close is within the top 50% of the range, buy at the market. Exit one tick past the existing daily low on a stop or at five minutes before the day session close. You have a corresponding short sale if the open was within the top 10% and the close was in the lower 50% of the range.

“Testing the range” (below) shows the results of setting the filter for 9:00 in the S&P 500, Nasdaq and Russell E-Mini contracts. These are based on five-minute bars, trading day sessions only. No slippage or commission is included.

The results suggest the kind of robustness that you want in a trading methodology. The profits are good considering that it doesn’t trade most days, and when it does, never more than once. Still, with several hundred trials per market, our sample size is statistically valid. The drawdowns are small and therefore we’re seeing some large “percent profit as a percentage of drawdown” figures, as seen on the bottom right of the tables. This is a significant measure because it describes gains in relative terms.

That is, it doesn’t just explain how much the system hypothetically would have made, but how much pain you would have had to endure before achieving those gains. If you had invested $2,537.50 in the S&P E-mini — the amount of the worst drawdown — the ultimate $21,525 account size would have represented an 848.28% increase (net profit as percent of drawdown, also referred to as return on account).

That also demonstrates its value as a filter. You would use this to get out of a position just prior to a painful drawdown.

Note the similar large percentages in the other two markets. Also note the relatively high profit-per-trade figures. Keep in mind, also, that this shows results for one E-Mini contract traded in each market. Across the board, the average trade figures blow significantly through the $20-per-trade amount you figure to give up in actual slippage and commission.


The other encouraging thing about the results is how well they conform to widely accepted facts about markets. Markets tend to open near highs or lows of the day more frequently than mere chance would suggest. This is one market tendency that is often cited to counter the argument that markets are random (see “On the right side,” below).

Trading ideas specific to the indexes are worth considering if they hold up well enough throughout the field. If they offer further bias confirmation in other financial sectors, so much the better. In “Another time,” (below) the results for the non-equity markets are nominal at best, but uphold the logic and confirm the potential for the half-hour methodology as an exit strategy for existing systems.

If in the first half-hour, the daily open is within the lowest 10% of the range and the close is in the top 50%, get out of your longs. Give your system (and yourself) a potential day off from punishment. If the move is false and the high is taken out, you generally have not given up much by getting back in at that point.

On the other hand, you may ride your flat position all the way to the close. It’s an idea that could help smooth out your equity curve — one that could make your drawdowns more bearable. It’s a way of eliminating some of your most financially and psychologically punishing losing days.

Even if it didn’t significantly increase your overall bottom line, wouldn’t the relief be worth it?

Art Collins is the author of Beating the Financial Futures Market: Combining Small Biases Into Powerful Money Making Strategies. E-mail him at

comments powered by Disqus
Check out Futures Magazine - Polls on LockerDome on LockerDome