Almost every chess player does puzzles, and almost none of them improve at tactics. The reason is usually the same. They solve a fresh puzzle, get it right or wrong, and move on. They never see that exact pattern again until it turns up in a real game, by which point it is too late. Free chess puzzles in Lumichess fix the first half of that problem. The Woodpecker Method tactics trainer fixes the second.
The principle behind Woodpecker is repetition under rising speed. You solve one fixed set of puzzles cleanly, then you repeat the same set again, and again, each time faster and at least as accurate. The motifs stop being puzzles you have to solve and start being patterns you simply recognise. That shift, from calculating a fork to seeing it instantly, is most of what separating a 1200 from a 1600 actually looks like at the board.
Why most chess tactics training fails
A random puzzle stream is good at one thing: variety. Every position is new, so over months you brush past a wide range of motifs. The trouble is that breadth without depth does not build instinct. You can solve ten thousand puzzles and still hang the same back-rank idea, because you saw that specific shape once, three weeks ago, and never again. Pattern recognition is built by spaced repetition, the same way you learn a language or a phone number, by meeting the same item enough times that recall becomes automatic. A puzzle you solve exactly once is a puzzle you are likely to miss when it matters.
This is the gap I kept falling into as I climbed. Most of the rating I gained over the years did not come from new ideas. It came from missing the same tactics less often. The Woodpecker trainer in Lumichess exists to make that closing-the-gap work deliberate instead of accidental. If that framing sounds familiar, it is the same idea behind the whole product, which you can read about on the about page.
Free chess puzzles in Lumichess
Puzzles in Lumichess are free, including the Woodpecker trainer. You solve on the same warm board used everywhere in the app, with the feedback you expect. A clear correct or incorrect signal on every move, the solution line when you genuinely need it, and a calm interface with none of the streak-celebration noise that gets in the way of actually thinking. The point of a puzzle session is to think hard about chess, not to chase a badge, so the trainer stays out of your way.
What the Woodpecker Method is
The method is simple and well known among coaches. Instead of an endless stream of new puzzles, you train on a fixed set. You work through it once, carefully, until you have solved every puzzle in it. Then you do the entire set again, but the bar moves. Each cycle has to be faster than the last and at least as clean. Over a handful of cycles, recognition that used to take you thirty seconds of calculation starts happening in a glance. The patterns are not new the second or third time through, which is exactly the point. Repetition is what burns them in.
- One frozen set. The puzzles do not change between cycles. That sameness is the entire mechanism. It is what turns a position you can solve into a pattern you own.
- Clean means clean. A puzzle counts as cleanly solved only if you got it with no wrong move, no hint, and no peek at the solution. Anything less is honest data that you do not yet have that pattern cold.
- Faster each time. Lumichess times every cycle, so you can watch your time drop as the motifs become automatic. The clock is the visible proof that the repetition is working.
Lumichess ships the trainer as a catalog of frozen sets. There are mixed sets banded by difficulty, from beginner up to master, and themed sets that drill a single motif at a time, such as forks, pins, mates, and endgames. You pick the band that matches your level so the puzzles are hard enough to make you think and not so hard that the first cycle never finishes.
The do-better gate
This is the part that makes Woodpecker training honest. After you finish a cycle, Lumichess checks it against your last passing cycle. To advance, the new cycle has to clear a do-better gate. It must be at least as clean and no slower than the cycle before it. If you raced through and made mistakes, the gate holds you back and you redo the set. You cannot rush your way to the next cycle, which means the speed you build is real speed and not a number you gamed.
A set runs across several cycles before it graduates. The first cycle is about getting everything right. Every cycle after that is about doing the same work cleaner and quicker, until the patterns no longer cost you any thinking at all.
The solving screen
Solving a Woodpecker set looks like normal puzzle solving with a cycle wrapped around it. You get the board, a live timer counting your time on the current puzzle, and a progress indicator showing your place in the set and which cycle you are on. As you solve, each puzzle is marked clean or not, so by the end of a cycle you already know how you did before the gate even runs. There is no mystery and no scorekeeping you cannot see.
The Mistakes pool
Getting a puzzle wrong should not quietly poison your whole cycle. In Lumichess, the puzzles you miss collect in a separate Mistakes pool that you can drill on its own. Solve one cleanly there and it clears out of the pool, without touching your cycle progress. It is a focused way to clean up exactly the patterns that tripped you, rather than grinding the whole set again to chase a few stragglers. Those misses are the most valuable puzzles you own, because they are precisely the shapes you do not yet recognise, and a few minutes spent on them pays off more than another full lap of the set.
Breadth versus depth
Everyday puzzles train breadth, because you see a new position every time. Woodpecker trains depth and speed, because the same motifs come back until you no longer have to calculate them. You want both, and Lumichess gives you both, free. Use random puzzles to widen your range and Woodpecker to make a chosen range automatic.
How to actually use it
- Pick one set and commit to it. The whole method depends on repeating the same puzzles, so resist the urge to start ten sets at once. One set, finished properly, beats five sets abandoned halfway.
- Solve the first cycle slowly and cleanly. The first pass is about getting every puzzle right, not getting through it fast. Speed arrives later, on its own, once the patterns are familiar.
- Repeat the set in faster cycles. Let the do-better gate be the judge. If it holds you back, redo the cycle rather than arguing with it. The gate is doing exactly what a good coach would do.
- Clear your Mistakes pool between cycles. Those are the exact patterns you do not yet own. A few minutes there is worth more than another full lap of the set.
Where this fits in the rest of Lumichess
Tactics training is one piece of getting better, not the whole of it. The natural companion to it is reviewing your own games, so you know which motifs you are actually missing in real play. A Game Report breaks a single game down move by move and names the tactic you walked past. A Profile Report steps back across your recent games and tells you whether tactics are even your biggest leak. Read those, find the pattern that keeps costing you points, then drill it here in a themed Woodpecker set until it is automatic. That loop, diagnose then train, is how a rating actually moves.
Frequently asked questions
Are the puzzles free?
Yes. Puzzles in Lumichess are free, and so is the Woodpecker Method trainer. There is no charge to solve puzzles or to run a Woodpecker cycle.
What is the Woodpecker Method, exactly?
It is a tactics-training approach where you solve one fixed set of puzzles cleanly, then repeat the same set in progressively faster cycles. Drilling the identical patterns over and over is what turns slow calculation into instant recognition. It is spaced repetition applied to chess tactics.
Why repeat the same puzzles instead of doing new ones?
New puzzles train breadth, but they rarely fix a specific weakness, because you almost never see the same motif twice. Repetition trains depth and speed. Lumichess scores each cycle with a do-better gate, so a cycle counts only if it was at least as clean and no slower than your last passing one. That is what makes the repetition compound rather than just pass the time.
What happens to puzzles I get wrong?
They go into a separate Mistakes pool you can drill on its own. Solving one cleanly clears it from the pool without affecting your cycle progress.
Does Lumichess celebrate streaks or pop up modals?
No. The trainer is deliberately calm. No confetti, no celebration modals, no streak alerts. Just the board, the timer, the gate, and your progress.