Thursday, October 20, 2016

Predicting the opens, analysis

In the past 30 days, which is as far back as I can go on the chart, it was 20-10.  You make money long term if you are above 50%.  Not as good as I originally thought though.

Distance between the two lines seems irrelevant.

Now, one thing that was abundantly clear, if you are wrong, sell at the very opening minute if possible.  This would only have been bad 2 time out of the 10 fails. 

If the signal is correct, the best time to close is usually about 30 minutes into the day.

Applying those two rules would have made wins bigger than losses.  I should mention, it might be difficult to close a position in the opening minute.

If Evolution agrees (doesn't require an entire signal, but should be obviously bearish or bullish) then it is 10-1, which is very good.

When Evolution disagrees, evolution was right 3/4 times.  (this is not statistically significant, I'm sure), at least not enough to say evo is better, but maybe enough to say, don't play.

Evolution by itself was 13-2, so maybe that is significant enough to say that Evo is better. 

Note that evolution only gives a signal like half the time.

So I'll try the following system
Signal on my system and nothing on Evolution - single bet
Signal on both (same signal) then double bet
disagreement - do nothing

When correct, sell 30 minutes after the open
When incorrect, sell immediately.

There, no room for human misinterpretations.  

Fun fact, the ONE day that went against both signals, was today :D  Watch out for days when the market breaks.

And for some reason, I just thought of considering the signals on the daily charts :P  I'll go back and look into it.


  1. This system of combining Evolution with the Overnight can be used on all kinds of things. I am going to experiment with the following:
    Evo + O/N both strongly agree = 2.0 points. Go For It!
    One strong one neutral = 1.5 points. Go if you feel lucky.
    Disagree = 1.0 points. Stay away.
    Very similar to your combined strategy.
    Check out NKE. Clearly a 2.0 Call. Good luck!

  2. Are you using a 30 minute frequency on both charts?