You Can Batch-Analyze Lichess Games Free — But Free Tools Won't Rank Your Mistakes
Lichess's free engine checks one game at a time, not your whole history
Are you rated 1000 to 1600? Are you stuck? You've probably searched for a way to batch analyze Lichess games free of any paywall. Here's the direct answer. Lichess already gives you a free engine. An engine is a computer program that looks at a chess position and finds the best move. Lichess's engine is called Stockfish. It checks your moves against the best possible move in any spot on the board.
Lichess also gives you a huge database of past games. It gives you opening books and opening trees too. You can use these to study your own games. You can also study games played by grandmasters. All of that costs nothing. It needs no upgrade.
But here's the catch. Lichess's built-in analysis board works on one game at a time. You open a game. You click "Request a computer analysis." Stockfish checks that single game. Then you close it and open the next one. Lichess has no button for your last 50 or 100 games. It can't analyze them all together in one pass.
That matters more than it sounds like it should. A mistake you make once in one game looks like a fluke. Feed that same engine 50 games. If you see the same mistake 15 times, that's not a fluke. That's a pattern costing you rating points every week. Lichess's free tools are real. They work well game by game. They just weren't built to add up your mistakes across your whole history and hand you one answer.
A real batch report grades every move and flags where the game actually swung
Before you go looking for a tool, know what "good" output actually looks like. A real computer game report sorts every move you made into one of six grades. Those grades are Brilliant, Great, Good, Inaccuracy, Mistake, and Blunder. An Inaccuracy is a move that's a little off the best line. A Blunder is a move that throws away a piece or the whole game. The engine decides your grade by comparing your move to its own top choice. It measures how much you gained or lost by not playing that top choice.
That "how much" has a name: centipawn loss. A centipawn is one hundredth of a pawn's value. It's the tiny unit engines use to score a position. Centipawn loss is how many of those tiny units a move costs you compared to the engine's best move. If your move matches the engine's best move exactly, your centipawn loss is zero. That's engine-perfect. If your average centipawn loss across a whole game climbs above 30, that's a red flag. It means you're giving away value. And not just once. Move after move.
A good report also rolls all of that into one accuracy score. That's a single percentage. It weighs every move you made against the engine's read of the whole game. Land above 90% and that's strong play, even at club level. Below that, and there's a gap between what you played and what was actually best.
Last piece: critical moments. These are the exact spots where the evaluation flipped hard. A drawn position turning losing. A small edge turning into a winning attack. Those are the moves worth staring at. They're where the game was actually decided. Everything else was just moves. So when you go hunting for a free batch tool, don't settle for one that just shows you a board. Look for grades, centipawn loss, an accuracy score, and flagged critical moments. Anything less is a partial picture.
Downloading your Lichess archive is easy; turning it into one clear pattern is not
Here's the practical path if you want to do this by hand. Lichess lets you export your whole game history as a PGN file, for free. PGN stands for Portable Game Notation. It's just a text file that records every move of every game. Lichess will hand you one covering your last 10 games or your last 500.
Once you have that file, you need a free engine to actually check it. Sites and apps built on Stockfish will happily open that PGN and grade a game. No account needed. That part works fine.
Here's the honest problem. Each of those tools grades one game at a time. You open game one. You get a report. You write down what went wrong. You close it. You open game two. You repeat. Do that 50 or 100 times and you'll have 50 or 100 separate reports sitting in separate tabs. Nothing adds them up for you. Say you blundered your queen in game 12. Say you blundered your queen again in game 47. Nothing tells you those are the same mistake, showing up again and again.
So the engine access was never the hard part. Stockfish is free everywhere. The missing piece has a name: aggregation. Aggregation just means adding results up into one total. Somebody, or something, has to read all those separate reports. It has to tell you which mistake shows up the most. It has to tell you how much rating that mistake is likely costing you. That step is where free per-game tools stop and batch tools start.
You already know openings aren't the plateau — the same is true across your last 50 games, not just one
If you've switched openings before hoping it would fix your rating, you know it doesn't work. The reason is the same reason single-game review hides your real problem. You're looking at one slice at a time instead of the whole picture.
Here's how that plays out. In game 12, you drop a piece to a fork. A fork is a move that attacks two pieces at once, so you can only save one. You call it a one-off, a bad day. In game 31, you miss the exact same kind of fork. Different opponent, different opening, so it doesn't register as the same mistake. In game 47, it happens again. Reviewed one game at a time, that's three unrelated bad moments spread across a month. Reviewed together, it's one habit that's cost you three games.
The smart move, even inside a single game report, is to start with your biggest mistakes and blunders first. Don't click through from move one. That same logic scales up. Instead of hunting for your worst move in one game, you want your worst-repeated move across 50 or 100 games. That's the pattern that's actually holding your rating down. Not the opening you played on move three.
Your openings were never really the plateau. The plateau is whatever mistake keeps showing up no matter what opening gets you there. You can't see that by reviewing games one at a time. You need every game in front of you at once.
BlackBox Chess runs the batch and names the one mistake costing you the most
This is the gap BlackBox Chess was built to close. Think of two questions. "I have an engine." "I know my one recurring mistake." There's a real gap between those two. BlackBox Chess runs Stockfish across a batch of your games automatically. Then it adds up the results into one answer. You don't end up with a stack of separate reports.
The free version checks your last 10 games. It names your number one recurring mistake. No account, no card, no signup wall before you see anything. You get the pattern, straight up.
If that first look shows you something worth digging into, the paid report goes further. For $10, it runs the same batch process across your last 100 games instead of 10. It ranks every mistake pattern you have by how much it's actually costing you. Not just by how often it happens. Here's why that split matters. A small mistake can happen all the time. A rare mistake can lose the whole game in one shot. The rare one can cost you more, even though it happens less. The report backs each pattern with 3 real examples. They're pulled straight from your own games. You're not staring at a vague label like "tactical weakness." You're looking at the actual position where it happened. A fix course comes with it, built around whatever your top pattern turns out to be.
If you go the free-account route, there's one more piece. BlackBox emails you a daily drill. It's built from your own mistakes. So the fix isn't a one-time report you read once and forget.
You already know clicking through Lichess games one at a time doesn't add up your mistakes for you. If you want that math done for you, the free 10-game check is the place to start. It takes your games in. It gives you one clear mistake to fix. Not fifty separate reports to compare by hand.