The Woodpecker Method: a chess tactics trainer built on repetition
The Woodpecker Method is a tactics-training system built on one counterintuitive principle: you sharpen your tactical eye faster by re-solving a fixed collection of puzzles than by working through an endless stream of new ones. You solve one set of positions, then return to that exact set and solve it again, cycle after cycle, each pass quicker than the last, until a position stops being a problem you calculate and becomes a shape you recognise on sight.
The method was developed by grandmasters Axel Smith and Hans Tikkanen and published by Quality Chess. The name comes from the bird that returns to the same spot until it breaks through, and from Tikkanen, whose Finnish surname means little woodpecker. Lumichess rebuilds the pen-and-paper routine as a free trainer that keeps the set frozen and does the timing and bookkeeping for you.
The principle behind it
Almost every tactic you meet over the board is a variation on one of a few hundred recurring shapes: a knight fork that hits king and queen at once, a deflection that drags away a defender, a back-rank weakness, a pin that wins material a move later. Strong players rarely reason these out from first principles. The pattern registers, the way a fluent reader takes in a whole word without sounding out each letter.
The Woodpecker Method is a deliberate way to build that library. By meeting the same positions repeatedly, at shortening intervals, each shape moves out of slow calculation and into fast recognition. It is spaced repetition applied to chess tactics, with one difference that matters: the act of solving does the encoding, not passive review.
Why re-solving beats chasing new puzzles
Grinding a brand-new puzzle every day feels productive, and most trainers are built to feed you exactly that. The trouble is what an unfamiliar position exercises. When you have never seen it before you have no choice but to calculate from the ground up, and that one-off effort leaves almost nothing behind, because the next puzzle wears a different disguise even when the underlying idea is identical. You end up rehearsing the act of calculating rather than building the memory that lets you skip most of the calculation.
Re-solving a fixed set turns this around. The second and third time you meet a position you are no longer working out the answer, you are practising the recognition, and recognition is what transfers into real games, where you seldom have the clock to rebuild a combination from scratch. This is the opposite of how rated puzzle training works, and the two are meant to be used together rather than instead of each other.
How the method works, step by step
- Choose one fixed set of tactical puzzles pitched at your level. A few hundred positions is plenty. The original book is built on just over 1,100. A Lumichess set holds 50, so a full pass is realistic in a sitting.
- Solve the entire set once, honestly, with no hints, no engine and no skipping the hard ones. This first pass is slow, and it is meant to be.
- Return to the identical set and solve it again, finishing faster while holding your accuracy at least as high.
- Repeat for up to seven cycles. With each pass the set takes less time and fewer positions need real thought, until the last cycles feel less like solving and more like sight-reading.
Two constraints make the method work, and both are easy to break on your own. The set has to stay frozen, with no swapping positions in or out between cycles. And each cycle has to be measured against the one before, or you cannot tell improvement from a good day.
What the trainer records
Every pass is timed. At the end of a cycle Lumichess stores two numbers for it: your clean-solve rate, meaning the share of positions you got first try with no hint, and your average time per position. From the second cycle onward both numbers from your previous pass are shown to you while you solve, as the target to beat.
It is a target, not a barrier. The cycle advances whether or not you beat it, because a bad session is information rather than a reason to be locked out. What the record gives you is the honest version of your progress: a cycle that was slower than the last one shows up as slower.
Finishing early when the times flatten
Seven is a ceiling, not a quota. The published method stops when recognition is automatic and times stop falling, which commonly happens between five and seven cycles. Lumichess checks for that: once your times stop falling after several cycles, the set is marked as flattened and you can finish it there. Repeating a set you already recognise instantly trains nothing.
The mistakes pool
Anything you get wrong during a cycle is collected into a separate pool you can drill on its own. Working through it does not disturb your cycle progress, so you can clear the positions that are still costing you without restarting a pass or corrupting the timing record.
The sets
The catalog holds 580 fixed sets of 50 positions each, drawn from real games. They are organised three ways. Mixed sets give you a balanced spread of motifs at one rating band, from 600 up to 2599. Theme sets drill a single motif: forks, pins, skewers, deflection, interference, back-rank mates, promotion, quiet moves, endgame technique and fifteen more. Opening sets collect the tactics that arise out of a specific opening family.
There is also a personal set, forged from the tactics you have actually missed in your own games rather than from the catalog. It is the sharpest version of the method, because the positions are already the ones you do not see.
One mixed set at each of the four levels is free for everyone, with no account needed to start. The rest of the catalog is part of a Lumichess membership.
Who it is for, and what to expect
The method works best once you already know the basic motifs and simply need to spot them faster. If forks and pins are still new to you, learn what they are first with rated training, then make them automatic here.
Treat a full set as a multi-week project rather than a weekend. The early passes take the most time and the last ones go quickly. The authors documented strong results, and Tikkanen earned grandmaster norms within weeks of training this way, but the method is not a shortcut: it works because the repetitions are honest, and that part is yours.
What usually goes wrong
Most people who abandon the Woodpecker Method do not fail at the chess. They fail at the discipline around it. The usual three: swapping in fresh puzzles because re-solving familiar ones feels unproductive, when that flat, too-easy feeling is the pattern being burned in. Losing track of times and accuracy across seven passes until there is no honest way to know whether a cycle was better. And quietly counting a position as solved when you used a hint, needed three tries, or guessed. The trainer removes the first two: the set is locked, every solve and time is recorded, and each solve is marked clean or not. The third is still down to you.
Where it fits with everything else
Woodpecker builds recognition. Rated training builds calculation depth on positions you have never seen, and Puzzle Storm builds speed under a clock. If you want to know which motif to drill first, run a Profile Report, which reads your last 1,000 games and names the weakness costing you the most rating, or review a single game with Game Report and start from the tactic you missed in it.
For how to choose a set, how many cycles to run and the ways the method goes wrong, read the Woodpecker Method guide.
Frequently asked questions
What is the Woodpecker Method in chess?
A tactics-training system from grandmasters Axel Smith and Hans Tikkanen in which you solve one fixed set of puzzles, then re-solve that same set in repeated, progressively faster cycles. The goal is to convert slow calculation into instant pattern recognition.
How many puzzles should a Woodpecker set contain?
A few hundred is enough to cover the common patterns, and the original book uses just over 1,100. Lumichess sets hold 50 so that a single cycle stays realistic and you actually reach the later, high-payoff repetitions.
How long does the Woodpecker Method take?
Plan for a multi-week project. Up to seven cycles of one set, each faster than the last, so the early passes take the most time and the final ones go quickly.
Do I have to beat my last cycle to move on?
No. Your previous cycle's clean-solve rate and average time are shown as a target while you solve, but the cycle advances either way. The record stays honest, so a slower pass reads as a slower pass.
Do I have to finish all seven cycles?
No. Seven is a ceiling rather than a quota. Once your times stop falling after several cycles, the set is marked as flattened and you can finish there.
Is the Lumichess Woodpecker trainer free?
One mixed set at each of the four levels is free for everyone, and you can start without an account. The rest of the 580-set catalog is part of a Lumichess membership.
Is the Woodpecker Method good for beginners?
It works best once you already know the basic motifs and need to spot them faster. True beginners usually do better learning what forks, pins and skewers are first. The Woodpecker Method then makes those familiar patterns automatic.
How is the Woodpecker trainer different from ordinary puzzle training?
Ordinary puzzle modes feed you new positions endlessly, which trains calculation but builds little lasting pattern memory. The Woodpecker trainer locks a fixed set and repeats it in timed cycles, so repetition turns into measurable recognition.
Based on The Woodpecker Method by Axel Smith and Hans Tikkanen, Quality Chess.