Monday Rejection

In my boredom yesterday, I started to take a look at 1st trading days of the week (usually a Monday) that open inside Friday's range then push to a H > previous week's H, but ultimately get rejected and close the day BELOW the PP for use during that day. (PP based on RTH prices only)

It doesn't happen that often. Only 16 times since April of 1999, about 75% of the time this (in the past) has marked the beginning of "at the least" a correction.

7/19/1999 began trend down 16 trade days , loss of close to 130 pts

9/7/1999 stuck around Monday range for the rest of the week then started down H 1574 down to 1463 L 15 days later

1/7/2002 downtrend started H of day was 1262, 22 trade days later, L was 1164.

12/27/2004 & 1/3/2005 consecutive MOndays (or first trading says of the week) The 12/27 signal was followed by a week of sideways in which prices were virtually held captive by Monday's RTH range, managing to move out of it only a pt or two before the second signal on the 1/3 that started a decline from H of 1305.50 down to 1249.25 14 trade days later

7/25/2005, price managed to produce the signal and then over the course of the next 7 trade days only managed to print a H about 7 points ABOVE the H of the signal bar. then down first leg down was L 1277 from H after sigbar of 1324, leg down was 17 days after failure H of 8/4/2005

10/3/2005 day after sigbar was a big black candle and price moved down without a HH on RTH daily until 9 trade days after sigbar. MOve was from H of 1308 down to L 1241 on 8th trade day after signal

2/21/2006, kind of a clunker, mostly sideways

5/1/2006, sideways and a little higher for 7 trade days after signal H of sigbar was 1373, H of subsequent 7 trade days was 1382, then DOWN, in 2 legs, 1st leg about 9 days L print was 1297.50, a bounce and then another leg down with L print 1269.50

1/3/2007, failed signal. It was a big losing day, but prices stabilized over the following 5 trade days and then headed Higher.

4/23/2007, Failed signal, just sideways

7/16/2007 H of sigbar was 1554.75, sideways for 4 trade days managing a print H of 1554 and 1553.50, then down Low print 1363 about 23 days after sig

4/7/2008, NOT great for a protracted move, 6 trade days of net lower prices H on Sigbar was 1352.00, L print 6th trade day later was 1288.00

9/2/2008, look at it, H of sigbar was 1264, 28 trade days later Low print 796

4/26/2010, H of sigbar was 1197.50, Flash crash was 8th trade day after sig, on 21st trade day after sigbar, L print was 1019.50

3/28/2011 Yet to be determined.

If anyone can see reasons (technical reasons) for the apparent failure of the signals dated

I would really appreciate a response.

you can look at these dates yourself and see if there is anything you run that could improve "predictability" or eliminate, or render extinguished the 3 signals above
thanks for all your observations (none, nada, zip)
Looks like the signal needs an action bar, a "trigger" PA of a daily close below the low of the prior week. in effect for 15 trade days.
Perhaps some kind of trigger would help. However, the fact this setup has worked 75% of the time is pretty good evidence its worth trading. You just need to define your risk. There are no guarantees in this business or life for that matter.

The only downside is the small sample size.
Thanks for your input Lorn.
Agree, sample size too limited, but, rejection is rejection and I was curious.