Gaps

I was pondering over the subject matter of how one would go about trading a gapping market. Or, rather, what criteria would warrant trading such an action? If a market opens higher than the previous day's close it either keeps going, consolidates, or fills the gap it created on the open. So, I found an interesting post written by Brett Steenbarger on his blog, TraderFeed.
In this posting Mr. Steenbarger takes a sample size of 897 trading days in the S&P500 index. He compares the day's gap to the previous day's range (high-low) in the form of a percentage, thereby measuring "the gap relative to the prior day's volatility." The example given is as follows: If the market opened up 2 points higher than the previous day's close, and the previous day had a 6 point range, the result is a 33% gap (2/6 x 100 = 33). By doing this he found a 27% average on opening gaps that provides "a benchmark for defining relatively large and relatively small gaps." His research found that about half of all large gaps (over 40%) don't end up filling and 80-90% of "relatively small gaps" (less than 40%) will fill in. Another conclusion he drew from his study was "large gaps in either direction tend to be bullish for the next day's price change."
So, what I've gained from his posting falls to having a keen eye on your technical analysis. Especially, when it comes to the large gaps. If one has a 50% chance of a large gap filling then you should be looking out for those reversal signals. At the same time, with the smaller gaps, the odds of a gap-fill are in your favor and anticipation is prevalent.