If a stock is gapped up and the futures go up from the open, it is unlikely that the gap will fill. Conversely, if a stock is gapped down and the futures go down, the gap will usually not fill either. Apparently this is a more of a 90s phenomenon whereas the specialists could manipulate the price by opening up/down at much more extreme levels. Now, the specialists gap stocks much more on customer orders and not artificially for the benefit of their own accounts. One exception with my thinking is that it applies to the more liquid and heavier traded stocks..like S&P 500 stocks. I don't track small Russell stocks, so I'm not sure about their case. But artificial manipulation is obviously much easier on thinly traded companies on scare orders.
I see this all the time - a 100-share lot prints at some crazy price that leaves a wick on the bar, but doesn't affect the bid/ask at all. I've asked around how that happens, but haven't yet gotten a firm answer. I've never had this trigger a hard stop, and I've been told by my broker that stops are triggered by the best bid or ask price, not by a print at a different price. The other thing I see all the time is after-hours prints at prices having nothing to do with the bid or ask. I had an after hours bid to cover a short position at 28.02, while the best ask was 28.24, and I got a 100-share fill at my price. Then suddenly a large lot was sold @ 28.13, then some smaller lots at 24.18. The whole time the ask remained at 28.24. So I imagine after hours trading does not involve a lot of transparency.
Often those prints you see are prints that went off during the day but haven't yet hit the tape, so in the AH they report them. "Late prints" is the term.
I noticed this also, and this drives me insane.. I developed a statistical strategy based off of gap behavior, and these erroneous prints basically destroyed its results.. I love how the SEC and the NASD are all about "fair and accurate" reporting of trades, yet base the open off of 141 shares trading .50 from where the stock naturally opens...Its a load of crap. And unfortuantely Tradestation only works on a minute time fame, so you cant exclude these initial seconds from its analysis.
I basically took this idea, and attempted to break it down into statistics based off of the first 5-20 minutes of trading.. To me, knowing AAPL fills the gap 90% of the time didnt do me much good, if the stock trades in $3 ranges and no time study was done.. In other words, where do you enter and how long to you hold to position yourself to profit?
I looked at the same in an attempt to automate. Impossible in my opinion. This is a discretionary trade. I trade 3 intra-day strategies...opening gap fills are 1 of them. Some things to consider in determing entry/exit: previous hi-lo previous close strong s/r levels on the dailes. I do not look for the entire gap range, only a percentage or a discretionary level based on the considerations above. r/r must be determined which allows for an easy placement of mental stop. Also stack the odds in your favor....way overbought and gaps down?...market turning up....good probabilitiy it will fill.
That is actually and interesting point.. I guess my idea of playing the opening gap always involved a large amount of symbols (spreadsheet) and automation, but using a more stock specific discretionary approach could be a good idea.
in "trade like a hedge fund" altucher states that gap fill is bread and butter for traders and hedge funds. you can read some of it online: http://books.google.com/books?lr=&q=fade+gap+hedge&btnG=Search+Books the main points from his book: gaps fill 80% of the time gaps should be faded (but not all of them) the bigger the gap the better for fading shorting the gap up is much trickier than going long the gap down.
I currently am studying gaps, and found that of the gaps i studied so far only 20% of the gaps will fill in a 30 day period. anyone thats intereste hit me up on skype and i'll send you the files of all the gaps. I also figured that if you had 100k, and bought every gap up at the open, and held for a 12% gain or for a 5 day period, you'd be up 67% (67,000 profit!) well here look for yourself