The case for reviewing your games, not just playing more of them.
One of the clearest, most consistent differences between players who steadily improve and players who plateau at the same rating for years isn't how much they play — it's whether they review what they played. A single game reviewed carefully often teaches more than several games played without any reflection afterward.
In the middle of a game, under time pressure and with only one position in front of you, it's hard to spot a recurring habit. Looking back across several reviewed games is different — it's much easier to notice "I keep missing knight forks" or "I keep getting worse positions out of this specific opening" when you're looking at the pattern instead of just the single moment. That pattern is what's actually worth fixing, far more than any one individual mistake.
"I need to get better at chess" isn't something you can actually train. "I lose material to pins because I don't check what's behind my pieces" is. Reviewing real games — your own real mistakes, not generic examples from a book — is what narrows a vague sense of struggling into something concrete and trainable.
It's tempting to run a game straight through an engine and read off the mistakes. That's useful, but skipping straight to the engine misses the most valuable part of review: replaying the game yourself first and honestly reconsidering what you were actually thinking at each critical moment. The engine is best used to verify and sharpen that thinking afterward, not to replace it — an engine can tell you a move was wrong, but only you can figure out why you played it.
Every loss has a specific, identifiable reason, even if it isn't obvious in the moment. Reviewing a loss in a calm, pressure-free setting afterward often surfaces exactly what went wrong — and that honest look tends to be where the most useful lessons live, since a loss usually contains a clearer mistake to learn from than a comfortable win does.
A thorough review doesn't need to take long — twenty to thirty minutes on the one or two moments that actually decided the game is usually enough. The goal isn't to catalogue every small inaccuracy; it's to understand the handful of moments that mattered, and why they went the way they did.
Reviewing games is also how you figure out where to spend your practice time. If several reviews show you missing the same kind of tactic, that's a signal to drill that pattern specifically. If your reviews keep pointing to a particular type of endgame, that's where targeted study will pay off more than general practice.
Ready to review a real game? Analyze runs it through Stockfish 18 for free, and How to Actually Review Your Own Games covers the practical workflow for making the most of it.