Suppose I told you that I had studied the market's present situation going back a large number of years. I examined the historical record and found that, two months later, the market was higher 45 of the 50 occasions studied for an average gain of 5%. Would you consider buying that market?
Now suppose I added one piece of information to the analysis: the average adverse excursion for these trades was -12%. In other words, before the market registered its 5% average gain over the two month period, a trader would have had to endure a drawdown that was more than twice as large.
Now would you consider buying that market?
The point here is that, for traders, the journey from point A to point B is just as important as the end point. If risk management guidelines and a trader's own psyche won't countenance a 12% drawdown, then the seemingly great edge is not attractive at all.
Wouldn't it be nice to have an indicator that tells you the ratio of the average expectation to the average adverse excursion for a particular time frame?
It turns out that Henry Carstens has developed precisely such a tool and has posted it to his website, along with the basic logic. When the indicator shows positive expectation, but less expectation than adverse excursion, Henry color-codes the indicator yellow to indicate caution. When we have a positive expectation and more expectation than adverse excursion, we get the green light on his indicator. A red light suggests a negative expectation--no edge or positive returns in the trader's favor.
Interestingly, Henry showed us as having a yellow light for today's trade: positive expectation, but even larger adverse excursion over a five-day period. So far that looks like a good analysis.
Over dinner in Portland just yesterday, Henry shared with me that taking adverse excursion into account had saved his capital during the recent market decline, as he decided to not take trades that otherwise looked good from an expectation perspective (e.g., buying a weak, high volatility market). Calculating adverse excursion thus becomes a risk management aid.
It will be interesting to follow Henry's tool as we attempt to put a bottom in this market. I will also make an effort to add an analysis of average adverse excursions to my historical pattern analyses to see if they help to define edges that traders can actually participate in.
There's a saying among physicians: pills don't work in the bottle. If the side effects are too onerous, the patient won't take the medication and derive any benefit. Similarly, a statistical edge is worthless if it entails so much drawdown that traders won't stick the trade out. Even the best vacation destinations are unappealing if the journey seems unsafe. Perhaps that's why so many traders end up cutting their winners short, sensing adverse excursions to come.
Self-Inflicted Problems of Trading
Risk Management and Trader Biology