Monday, September 18, 2006

Stop Loss Exits: Managing Risk vs. Managing Anxiety


In a recent post, I suggested that stop-loss points have a way of limiting opportunity as well as risk. My observation was the result of testing hundreds of trade setups with the Odds Maker program and noticing how such staples of the market literature, such as fixed price-based stops and trailing stops, severely degrade the performance of those setups.

Should one abandon stops altogether? I think not. If we consider every trade to be a hypothesis that is based upon our understanding of the marketplace, we owe it to ourselves to abandon that trade once the hypothesis is disconfirmed. The problem with most price-based stops is that they are not formulated with these hypotheses in mind. They are employed, I suspect, as much to manage the trader's anxiety as to manage objective risk.

Above is the Market Delta chart from the Friday AM ES market. The Odds Maker trade setup suggested that fading breakouts of the morning's 30 minute opening range and holding that position for 30 minutes produced, in the past three weeks, eight winners in fourteen trades and an average win size considerably larger than the average loss size. (See my Trader Performance page for details).

Notice how buying dried up after the attempted upside breakout. (My Trading Markets article scheduled for later today explains this in detail). Order flow was telling us that it was reasonable to hypothesize that we had made a short-term price high and would move lower. My own research suggested we would test the average trading price for the morning, as I posted to the update for that day. Suppose, however, that a news item hit the wires and new volume came to the market, lifted offers, and gave us expanded volume at 1336 and above. Clearly that would have invalidated my idea of buying drying up. Stopping out the trade even before it became a loser would make sense in such a circumstance.

A trailing 3 tick stop, however, in the name of "locking in" profits, would have led to the premature abandonment of a profitable trade. Three ticks of heat is normal for such a trade. We know this from studying the individual trade setups and their maximum adverse excursions. Simply because a market moves against a position does not necessarily mean that the underlying trade idea is invalid. The important thing is to know your trade idea and what would truly invalidate it.

Not knowing what would invalidate a trade can lead to much frustration. A good example of this is the common practice for traders to place their stops just beyond obvious price points of support or resistance. Large locals, knowing this, will work orders in the book beyond those points, push the market past support/resistance, trigger the stops, and then let the puking traders fill their resting orders. If you look at any daily ES chart, you'll see plenty of such "false breakouts."

A good trade idea already has risk management built in. If your trade idea is a tested one, it has to have both an exit and an entry. It cannot be tested otherwise. The time-based exit of my trade accomplishes a great deal of risk management. Beyond that, judiciously allocating limited capital to each trade and diversifying trades among time frames and asset classes accomplishes the bulk of risk management. Think of managing portfolio risk, not (micro) managing each trade. The stop-loss on my 30 minute trade--or any single trade--is a last resort. It has a value and purpose, but it cannot shoulder the entire risk management burden.

Is your stop loss criterion truly protecting you from the risk of future adverse price movements, or is it stopping your temporary level of anxiety about a drawdown during a trade? Do you really know how your stops impact your trading performance? What makes us most comfortable in the markets is rarely what enables us to pursue opportunity. The best stops protect us from our fallibility, not from our frailty.

4 comments:

yinTrader said...

Hi Brett

Your analysis of putting Stop loss exits is eye opening.

I have two strategies on where to put my stops for intra day trades on emini SP and they reflect more or less what you have pointed out.

You have reinforced my beliefs on why, how and where to put stops.

One has to discern what is managing risk and what is managing anxiety as you put it, to trade well.

Brett Steenbarger, Ph.D. said...

Thanks, Yin. I think a worthwhile exercise is to review each trade that you stopped out and actually see what the market did subsequently. That will give you a feel for whether your stops are protecting you from loss or from opportunity!

Brett

TI Maven said...

Hi Brett,

Very good observation about the stop loss issue. One thing that we have been playing around with is bracketing an order from a strategy with a stop loss and profict target at the same time. Those numbers are derived from the Average Winner and Average Loser in the Odds Maker. The idea is simple in each strategy try not to lose too much more than the average.

Brett Steenbarger, Ph.D. said...

Hi Dan,

That's a very interesting idea. You could develop stop loss and target zones based on the distribution of maximum positive and adverse excursions during the trades. I suspect those excursions would provide better guidance for targets and stops than the avg win/loss data alone.

Brett

Brett