What I was saying was that without the review of each specific trade at the time to establish why it had X drawdown and why it left Y pips/points/cents on the table, those data aren’t going to change too much.
Well, not if they are normalised by volatility....
This review malarly is a bit of a catch 22 situation, because you can always find a technical reason as to why price did something. If price isnt bouncing off a 61.8% retracement on a 4 hour chart then its bouncing off a 30 period bollinger band on the 1 minute chart, or a 34 period EMA on the 15 minute chart, or a pivot point, or there's oscilator divergence on the 30 minute etc etc etc.
You can always find a technical reason to support a move retrospectively. Mr Socco talks about being trapped in a hall of mirrors, ( and Talab talks about being fooled by randomness) but its this kind of trying to rationalise a problem using a limited toolset with incomplete understanding that causes these problems.
A pragmatist would say I have 1000 winning trades and the winners didnt exceed an MAE of X. The technical analyst might say, I have 1000 winning trades, 90% didnt exceed an MAE of X ATR. You can add in as many layers of complexity as you like over and above that with the danger of curve fitting.
The idea is to make some money, and you do that by having a knowledge of the game that you are playing. These simple metrics just provide some useful boundaries, and after that its nothing more than a case of recognising when price is moving appropriately for a given set up.
I'm not saying that you shouldnt review trades. I certainly do, and I wasted thousands of hours developing the technical capabilities to replay them on various timeframes etc, overlaying every technical study known to mankind, but if I had my time over again, I'm not sure I'd focus quite so much on that particular aspect.