Hi ProfitSeer,
did the following testing:
1) using SP tick-by-tick data from 01/2000 til 08/2003 (could do back til 1990)
2) defining range: High to Low, including Gaps if happened
3) defining synthetic contract: forward, absolute adjusted (absolute ranges remain same, relative change)
4) defining average range: arithmetic mean over the last 10 ranges (equal our expectancy for today)
Results:
Number of trading days: 899
Number of times a 110% extention of expected range happened: 263 (29.25%)
Number of times a violation of both sides happened (market turned around to make new high/lows after initial breakout of expected range happened): 12 (1.3%)
Average expansion (follow-through), after expected range was exceeded (again measured in multiples of expected range): 29.4%
Chances of sustained follow-through: slim (check the enclosed chart for extensions to the side of the break-out) - only 40% show more than 30% extension after expected 110% range was exceeded.
To finally answer the question: only 45% of all expansions of 110% exected range in the SP showed more than 25% of expected range of follow-through. The good news is that most of the times, the market doesn't turn around for new highs / lows, therefore setting the stop at the other side of the session seems a good idea.
Regards, Oliver