<<id like to see what you come up with. the last few days were bad, but all time bad thats a brain scratcher.>>
I developed a 3SMA for my Issues statistic, but results were not "all time bad". The OS was quite deep, but I found 5 or 6 dates since 1996 that were deeper in this 3 day statistic.
All in all, the 3DMA test was not markedly different than the 5DMA test. Would be nice if Peter E would tell us what statistic here was using.
Actually now that I think about it, I was using Nasdaq internals. Perter was probable using either NYSE or a combination, since Nasdaq doesn't go back to 1926.