tt like I said ..... after 10 years of active trading I have yet to find a consistent method of trading gaps. With do respect to the previous poster stating all those statistics I have calculated similar statistics MANY moons ago, they are widely known. Building a black box to trade around them in my opinion is a very silly thing to try and do. I know I've tried too many times to count and like I said I consider myself lucky to be trading even on them now.
I doubt anyone has too much long term success with a black box. Its better to have the stats in mind when you enter a trade IMHO. If your just even on them though and consider yourself lucky to be that I would think you should focus on something other than gaps and just forget about them. Brandon
That's why I asked. I haven't found the basic concept to hold true at all. To the contrary, the gaps seem to be (as I said, this is empirical) a measure of intent, and size really doesn't have much to do with it, particularly since the gap may be an illusion in the first place, a function of a pause in trading. Finding support and resistance from the previous day(s) seems to be more pertinent than the size of the gap itself.
db I concur with your findings, the previous days levels do play far greater significance in trading them. I also like to measure the psychology of the gap by understanding the current market conditions. This has proven very valuable. I have successfully traded large gaps with the same success percentages as small and medium. One very useful tool that I have also found is to measure the globex volume before the open, especially after a big 8:30 news release. this combined with the bond action (Open @ 8:20) can also tell you allot about the potential of the gap closing or running.
Although, if you can, it's sometimes better just to trade whatever caused the ruckus in the first place: http://www.elitetrader.com/vb/attachment.php?s=&postid=444676 If I'm lucky, the above will take you to a chart posted in my journal.
This article from the cbot website was posted on ET once before...I found it to be a more comprehensive look at how one person plays gaps than many...if anyone finds it of use, great - http://www.cbot.com/cbot/pub/cont_detail/0,3206,928+14719,00.html
from my persepcive there are two points to add: 1. within the sp future significant gaps down increase the likelihood of seeing a positive day. gaps up do not show significant changes in foreacsting the day. 2. whether a gap closes or not is not so much relevant as whether that day ends positively or not. a 1% gap has to make twice the distance to "close" than a 0.5%. it is obvious that - no matter which environment - a 1% move is less likely than a 0.5% one. that means that aggregating them under the one criteria "gap closed or not", is not useful. apples and peaches (or wonderbras). peace
Unless the futures are actually closed, tho, they technically aren't gaps. I don't collect data all night, but the action usually takes place - when it does, which lately hasn't been often - as a result of the 0830 reports, if any.
An additional bit of refinement on the concept of overnight session volume being predictive of future moves, take the overnight session's range and divide it by the previous session's day only range. If it is greater than 50% then odds are very high the following RTH session will be a big range day. I define big range days as 120% or more than the average daily range.