No registration required! (Why?)

Opening Gap Trade

The Regular Trading Hours (RTH) opening gap trade is a well known trading setup. This trading setup is certainly something I always consider at the opening of the new RTH trading session.

Guy did an analysis of this setup in his article titled "Fading the Gap"

From his article we learn the following historical statistics from the sample period he used...
  • "On average 76% or three quarters of all gaps close at some point during RTH."

  • Gaps on Monday were the least likely to fill and Thursday gap openings the most likely.

  • Opening gaps up to 7 ES points in size were filled during the session at least 70% of the time. This probability drops off surprisingly quickly above the 7 ES point range.

Given these basic concepts as a foundation, I thought it might be helpful to use this topic for further discussion, analysis and trading ideas.
day trading

thx for getting back to me on this. Sounds like it may be a good idea for me to check out the concept of the Market Profile. I just located a free presentation on the cbot website on this subject (i hope). I will check it out - am sure I will have more questions later.


You said "It would appear logical to me that the gap would have a higher probability of filling if the previous close was very close to the POC. This is something that I don't think that anybody's ever tested and something that I want to test in the future."

Then I re-read the section of your paper (p. 20) where you looked at gaps in relation to support/resistance levels. According to the paper, gaps that are in between 1st support and 1st resistance have ~80% (the best) chance of filling. Is that why you made the above statement?

However, in the same section of the paper, the data show that the avg loss per trade is pretty large compared to the avg gain per trade, at least for these 2 intervals.

Based on that I have a couple more questions:

1) I just want to make sure I understand right - this paper is about trades taken at the open (say as opposed to waiting for a reversal), correct?

2) Just wondering if the heavy losses compared to the gains could indicate that the trade initially continued in the direction of the gap and may have been stopped out before it actually did the reversal and came back to fill the gap? I think you referred to this possibility somewhere in the paper already.

Sorry if these are dumb questions. I read this paper a couple times now and there is a lot in it. thx Mike
Mike, No I didn't make the statement because of the direct relationship to the S&R levels against the study, however the same logic applies. Remember that the pivot point is typically the average of the high, low and close. It is highly likely that the pivot point (which is between S1 and R1) will be in the Value Area. My assumption was made without me remembering that page in the study but your observation is astute inasmuch as you've just pointed out that it has been semi-tested already by using a strongly co-related level - i.e. pivot point vs point-of-control, or PP vs POC.

1) Yes, the trades are taken at the open of trading and do not wait for a reversal. It is assumed that you're entry is the opening price.

2) If I remember correctly, most of the pages assume that you don't use a stop. This means that any reversal back to the previous close will have resulted in a profitable trade no matter how big the draw down.

One the problems with this strategy is that you have an asymmetrical payoff if you don't have stops (and the non-stop strategy appeared to be the most profitable). This is because you have a finite gain (to the previous close) and an infinite loss (no stops). Another study that needs to be done is to look at opening a position in the direction of the previous day's close and then closing the position at the end of the day irrespective of whether or not the previous close was reached or not. This would allow for positions to run to and beyond the previous close. Of course this would need to be tested with stop levels and without stops.

There's no such thing as a dumb question.

Hope that helps.
Isn't there a "rule of thumb" percentage difference between the open and the previous day's close that determines whether to take a "Fill the Gap Trade" ???
Piperian and day trading

Thought you may be interested in this article - discusses relative approx percentages of the time that gaps tend to close, based on whether they are large or small gaps.
Thanks! It's a start.
Thanks for the link mdszj. I have a lot of respect for Brett Steenbarger. Part of what he says in that article is:
When upside gaps exceed 40% of the prior day's range (N = 99), 46 of them fail to close during the day session. When downside gaps exceed 40% of the previous day's range (N = 81), 40 of them go unfilled. Bottom line: approximately half of all large opening gaps don't fill during that coming day's action.

Conversely, when the upside or downside gaps are less than 40% of the last day's range (N = 717), only 144 of them go unfilled. When the gaps are less than 20% of the last day's range (N = 431), only 51 of these fill in. Stated otherwise, 80-90% of relatively small gaps will fill in during the coming day's action.

Which is useful information to some extent but you also have to remember that any price point that's far away from a starting point is less likely to be hit than a closer price point. So as the gap gets larger it's always going to be more unlikely to fill but not because it's a large gap but just because it's a price point that's further away from the starting point.
Day trading

I agree with you. If I understand correctly, I believe that Brett's article and your paper agree on this issue and both have supporting data. I like the data table you have on pg 20 of your paper - shows that 80% of gaps in the S1 to R1 range (to me this means relatively small gaps) close, and gaps outside this range drop off to 60% or lower.

I am hoping that by selectively taking trades based on gap size, combined with entering on a reversal, if it takes place, will help to increase my odds of having a successful trade and reducing stop outs.

later Mike
One of the problems when developing a trading strategy (such as the gap fill) is that you end up with millions of unanswered what-if questions and the more answers that you start to accumulate the more complex the problem becomes and the more difficult it is to remember all of the data.

For example, you could ask "what if we targeted a fixed size profit for the gap trade irrespective of the gap size?" So we could use the gap as the direction and always target 1, 2, 3, or 4 points. Or we could target (say) 3 points or the gap fill. We could also ask which target/gap fill combination yields the greatest profit. Or should we use a target based on a fraction of the previous day's range...

With each one of these target ideas we need to test different stop sizes and similar ideas.

You end up with a Pandora's box of combinations...

I don't want to discourage you but just make you aware that it's a lot of work. Hopefully with a big reward at the end of it.
The opening gap is one of the best setups I have found for day trading. I am not sure why this thread has had no posts recently, but maybe this post will generate some interest and traders interested in gaps will chime in.

One of the best resources I have found for trading the opening gap is Check out some of his videos on the site. I have learned a lot from the GapGuy and continue to work daily to improve my gap trading strategy. If anyone knows of other valuable resources, post them in here or I may even start a new thread so search for "Opening Gap Trade"
great gap play in the ES today. Huge rally sold near the R2 pivot to watch it gap fill and then some.WOOT!