From Sentimentrader: (Nice market research stats--have a look-- I have no affiliation)
Small-caps Take Another Beating on Reversal Day
05/15/07 5:00 PM EST
Earlier this afternoon, I went over some stats regarding large intraday reversals in the S&P 500, after it had scored a new yearly high intraday.
The reversal looks nasty on a chart, but the historical precedents didn't back up that reaction - the S&P was higher a week later 14 out of 17 times.
I've been trying to tie in the relative weakness in higher-beta stocks, since that has been a theme for the past month and has been gaining steam. So I checked the reversal stats again, this time looking for any time we got the following:
* A new yearly high in the S&P 500 intraday
* A gain of at least 0.5% from the previous day's close
* A reversal to close the day below the prior day's close
* A loss in the Russell 2000 of at least .75% from the prior day's close
That study pulled up six precedents from the past 20 years (1/4/96, 4/6/98, 2/24/99, 1/3/00, 12/3/03 and 12/15/03). While short-term weakness was seen in 4 of the 6 cases, none of them preceded a lengthy decline, and in fact the S&P went on to a new high usually within a couple of weeks.
I've looked at today's reversal a number of ways, expecting to see some kind of bearish bias afterwards. But based on the reversal alone, I can't justify a bearish stance, and that's why we do the testing - it has saved us more than once.
