To Scale Out or Not?
Wednesday, December 10, 2008 at 11:18PMMost folks that know or follow me know that I have absolutely no problem showing off my losing trades since I find it therapeutic and helpful for the process of continuous improvement. Today, I was stopped out for my first losing gap trade in over a month (6 consecutive winners during that time). Here's the 5 minute chart for today's play:
My target for today's 8 pt gap was just in front of gap fill (the blue line). Even though prices sold off fairly rapidly the open, they stalled around the 75% gap fill area - about 2 pts above my target. They then reversed and stopped me out near the highs of the day before ultimately filling the gap.
One of our new members at MasterTheGap.com questioned my process of not taking partial profits at the half gap fill and stated, "there is something to be said for the psychological satisfaction of not watching a nice profit turn into a significant losing trade."
He is absolutely right about the psychological aspect. It is not easy on the psyche - even when trading a reduced position size like I did today. On the other hand, I have learned that doing what feels good when trading is rarely the right thing for maximizing profits. My gap trading plan is based upon historical probabilities that have been built using "gap zones" (i.e. where the gap opens relative to the prior days' OHLC), market conditions, seasonality, and multi-day price patterns. I have adjusted my stops and targets for each signal to provide a compelling blend of profits, win rates, and acceptable drawdowns.
My historical (actual, not back-tested) win rate averages 65 - 70% trading gaps. I could easily increase this to 80% by simply scaling out some or all at partial gap fill areas (i.e. 50-75%) versus the full gap fill or extended prices that I often use. BUT AT A SIGNIFICANT COST TO MY OVERALL PROFITABILITY.
Why? Because scaling out early on every gap play will dramatically reduce the size of my average winner while only reducing the size of my loss on those infrequent occasions when a partial gap fill target would be hit, before prices reversed and stopped me out. Check out the chart below that shows the frequency of the various partial gap fill areas being hit:
As you can see the difference in probabilities for a half gap fill versus a full gap fill are fairly small - a lot smaller than many think, but the difference in the average winning trade is HUGE. Food for thought the next time you are thinking of taking some profits prior to the gap fill.
It might feel good.... but over the long run, how much is it costing you?
View Printer Friendly Version
Email Article to Friend
Reader Comments (1)
Wow - this post has generated quite a bit of email traffic for me from folks that subscribe to the email delivery option. Here's one:
"Your discussion is indeed thought provoking. However, the "Scale Out or Not?" chart only tells half of the story. The new fellow's comment about letting winners turn into losers prompted this discussion in the first place, and the chart only describes the size and frequency of winning trades according to when profit was taken. Data on the size of the losers is missing. I'm certain that you have a positive expectancy system, but for the sake of completeness in this discussion, my opinion is that light must be shed on the losing side of the equation as well. Thanks for putting your gap research out there. -Don"
RESPONSE: Don, I agree. I told the member that made the comment that I would run the reports to show profit expectancy, win rate and drawdowns so that he could see the trade-offs himself for the various targets. But you bring up an interesting point regarding losses. I probably need to test and show multiple stop sizes too since this would impact the system's performance and possibly the decision to scale out or not. Though I am fairly confident that my stop size still favors holding for gap fill, perhaps a bigger stop would not. Thanks for the thought! fyi: it will likely be the New Year before I get to this, but stay tuned.
-GG