I was reading Dr. Brett the other day and he asked the musical question – what happens when the market (by this he usually means S&P 500) goes up 2 days in a row? He found that 3 days later the market is generally weaker than when compared to the average 3-day period. He then tested 2 down days in a row and found the exact opposite. (We could have told him that).
I thought about this for a bit and I said I could test this for individual stocks by using a simple filter. What I did was write a filter that simply looked for stocks selling between 15 and 35 where the close was higher than the day before on the last 2 days.
I then tested this using a 3-day holding period over the dates from 9/01/2006 to 12/15/2006. The return was $11500. (By this I mean starting with 100000 and ending with 111500).
I then switched the filter to two down days in a row and keeping everything else the same the return was 25100 (125100). Obviously you are better off going long off of two down days rather than two up days as a starting position if you are swing trading.
Something to think about - One of the problems with designing filters these days is that the market continues to go up so just about any filter you design and back test in the recent past will return a winning percentage. And that is not reality or at least not normal so be careful with your own screens. Understand that you have to take the results with a grain of salt.
Of course back testing in anything other than the recent past can also be problematic - so watch your step in any event.