try [Removed- DB]
How can I memorize chess openings?

I always analyse every game I play to see what the best response should have been, how the opening should have been played, and then I promptly forget it and play the un-optimal responses over and over. I think a big problem is the analysis on here and lichess doesn't actually explain to you why the best move is the best move. I'm trying to learn it by rote. We really need a proper chess AI that you can talk to and say "why is this move better" in game review.
Wait so what should you do if you are below 1800. I'm talking specifically about ranges between 1350-1450.

And thinking about it some more, the analysis feature here doesn't actually analyse all of your games each and every time, if it did and had an AI analaysing it, then it would tell me "hey, this specific blunder has cost you dozens of games. Let's really explore why you are intuiting that blunder and how we can retrain you to spot it". That would help people in any skill ranges, and is probably how Magnus already trains with custom prompts.

And thinking about it some more, the analysis feature here doesn't actually analyse all of your games each and every time, if it did and had an AI analaysing it, then
...then it would take a lot of time.
You played 2757 games. If we say that the average length is 30 moves (reasonable) that gives 165 thousand positions to analyze. If the engine spends only 2 seconds on each position (which is probably not enough) the analysis would take about 4 days.

You played 2757 games. If we say that the average length is 30 moves (reasonable) that gives 165 thousand positions to analyze. If the engine spends only 2 seconds on each position (which is probably not enough) the analysis would take about 4 days.
Well no, the AI would have access to the already pre analysed previous games data. It could see sudden drops on win odds based on positions in the analysis. When it analysis future games it could compare the sudden drop there to all the sudden drops in the middle of previous games, and look for patterns, even start intuiting itself how you are unknowingly intuiting those positions.

You played 2757 games. If we say that the average length is 30 moves (reasonable) that gives 165 thousand positions to analyze. If the engine spends only 2 seconds on each position (which is probably not enough) the analysis would take about 4 days.
Well no, the AI would have access to the already pre analysed previous games data. It could see sudden drops on win odds based on positions in the analysis. When it analysis future games it could compare the sudden drop there to all the sudden drops in the middle of previous games, and look for patterns, even start intuiting itself how you are unknowingly intuiting those positions.
This could work, but I have doubts. A chess engine could do the game analysis part, but I think the second part would require some new AI model. Existing AI (large language models) aren't designed to do statistical analysis.
H8