Gotcha! It means that these stats are worthless because you don't know in advance if it's gonna be a low-high or high-low day. It's like saying that the Bills have 80% chance of winning when they are leading after 3 quarters.
Well, I have to disagree with that. If you know with very high probability that the day high/low is in by the first 90 minutes surely that's useful information. And who's to say that can't be predicted in advance...? Besides, a good trading model will always be updated intraday as the day progresses and new information is known. In my experience most who say the market can't be predicted base that on trying to make a guess by looking at a chart and never took the trouble to even build a very basic predictive model. I'm only scratching the surface of what I'm looking at personally, but I don't like to say more than I've done already. I only made the initial post in this thread to demonstrate results that doesn't seem consistent with the theory of a randomly moving market to me. If so, you'd have the day high or low coming in at random times every single day. Something that clearly isn't the case.
It is almost 19 years old now (unless there is a later revised edition) but ... How Markets Really Work / A Quantitative Guide to Stock Market Behavior by Larry Connors (2004) ... lays out how non-random markets are. Or continue to believe what you want. Index Market Edges Short-Term Highs and Short-Term Lows Higher Highs and Lower Lows Up Days in a Row vs Down Days in a Row Market Breadth Volume Large Moves New 52-Week highs, New 52-Week Lows Put/Call Ratio Volatility Index Using the Information in this book That is as much as I will copy.
Let's change the perspective a little here. Can a random entry strategy make money in the market long term?
Just for the record, I trade based on price action that's pegged to the exchange open (London/Chicago). That indicates that I don't trade a purely random entry since there is consistent significant volume influx around that time compared to all other times But still curious, can purely random entry work in the long run