There's an interesting paragraph in "Rules of the Trade" by Nassar about gaps. An excellent book, BTW. According to him, Market Makers tend to bid a stock lower pre market if important news are released. They do this because they must quote a price they are willing to buy/sell for, and by bidding it far below/above "fair" value, they effectively soften the impact it has when prices return to fair levels.
Here you can test your theory on one or all symbols as a portfolio. The script test the theory that "all gaps are eventually closed". It looks for substantial down gaps, and enters the next day at market. It attempts to close an open position at the limit price of the pre-gap bar low. The script only take long positions but you can modify it to enter short positions too. The script can manage multiple open positions if there are several successive gap downs. The script reports on the number of gaps, and the percentage of gaps that were closed in the top right corner of the chart. If you go here and you key in any symbol you should get a brief overview of the key statistics of the last 1500 bars. http://www.wealth-lab.com/cgi-bin/WealthLab.DLL/editsystem?id=13631
Volker Knapp.....thank you very much for that website. Gaps are filled, but it doesnt necessarily happen right away either. Also, if you run it for most tech stocks, you will still be in the hole from the downward opening gaps in tech stocks from the crash that began in 2000.
When testing historically you should consider a few things like that the data is dividend and split adjusted. Take a look at MSFT on the link I published. The strategy bough MSFT after it paid out a large dividend in November 2004. BTW, there is still an open position in MSFT from April, 2000 at the 52 level. I don't think all gaps will eventually get filled. Again I am looking at MSFT, this time with dividend and split adjusted data and even then there is still a gap to be closed with MSFT at the 48 level - a long way to go. BTW, when running the strategy on all Dow 30 stocks, you get 88% of all gaps closed profitable.
If your talking about intraday gaps then I would say no. But, you can still make some coin off of just a partial gap move.
Yes, but what happens with the losers? Do the 12% losers outway the gains of the winners. Also, can we put stop losses in place and test that, i.e..... say I put a 1% stop loss on every trade. Sometimes, my stop loss will kick in and sometimes it won't. The question is, does the number of times the gap gets filled before trading down 1% outway the amount of losers that do trade down 1% and kick in the stop loss. Also, dividend paying stocks are really the best to try this on, because realistically speaking, they always gap down.
I know of a gap that never filled, MSTR. Microstrategy was a high filer during the bubble. It gapped down some forty points, and never recovered. I do not believe MSTR exists anymore.
not quite --- i play every gap for a certain position size, and all i go for is 1 point in the ES. my hit rate is very high so i will always play the gap fill {i just do not wait for the gap to actually fill --- i just capture some part of the "attempt"}.
I track a daily raw gap play across the emini futures which looks at what you would make each month if you traded each and every gap and did not use stops. This is a theorectical benchmark which some people use to measure their own gap trades to see if they can out-perform this benchmark. (BTW not using stops is NOT recommended.) If you have the OHLC data for any market it is very easy to test almost any gap trade theory using excel.