In the summer of 2010 a Swedish international master named Hans Tikkanen scored three grandmaster norms in seven weeks. Not three in a year. Seven weeks. When people asked what he had changed, the answer was unglamorous to the point of insulting: he had been solving the same pile of tactical puzzles over and over, faster every time, until he could rip through the whole set in a single day.
That is the Woodpecker Method, and it has been quietly annoying chess improvers ever since. It is boring. It is repetitive. It works better than it has any right to.
What the Woodpecker Method actually is
The Woodpecker Method is a tactics training system in which you solve a fixed set of puzzles, then solve that exact same set again, and again, with each cycle taking roughly half the time of the one before it. Six or seven cycles is typical. You never add new positions. The whole point is to meet the old ones faster, until recognition takes over from calculation.
Tikkanen’s countryman Axel Smith saw the results, tried it himself, went from 2100 as an adult to grandmaster, and the two of them wrote the book. The Woodpecker Method came out from Quality Chess in 2018 with just over 1,100 exercises sorted into three difficulty tiers. A positional sequel followed with a further thousand or so. The name is Smith’s: a woodpecker does not peck a hundred different trees, it hammers the same spot until it gets through.
Why repetition beats “just do more puzzles”
Here is the uncomfortable arithmetic of ordinary puzzle grinding. You solve 40 puzzles today, 40 tomorrow, 40 the day after. Every one of those positions you will see exactly once in your life. You get exposure. You do not get learning.
Strong players are not calculating faster than you in most positions. They are calculating less. They look at a board and the candidate move is simply there, because they have met that shape a few hundred times and their brain filed it. That filing happens through repetition, not through volume of novelty.
2 
7 
g 
e 
d Four seconds versus four minutes is not a small difference. In a 15+10 game it is the difference between having time for the rest of your plan and not. That gap is exactly what the repetition closes.
The Woodpecker Method schedule
The engine of the method is the halving. Whatever cycle one took you, cycle two gets half as long, and so on down. Here is what that looks like for a 300-puzzle set with a four-week first pass.
| Cycle | Days | Puzzles per day | What you are actually training |
|---|---|---|---|
| 1 | 28 | 11 | Calculation. Slow and honest, to the end of the line |
| 2 | 14 | 22 | Recall. Half of them tickle something, half feel brand new |
| 3 | 7 | 43 | Recognition starts doing the work |
| 4 | 4 | 75 | Speed. Answers arrive before you finish reading the board |
| 5 | 2 | 150 | Stamina. This is the cycle that hurts |
| 6 | 1 | 300 | The victory lap, and it feels ridiculous |
Two things people get wrong about that table. The halving is a target, not a law; if cycle four takes five days instead of four, nothing is ruined. And cycle one should be slower than feels comfortable, because everything downstream is built on having genuinely solved those positions once.
Take two or three days off between cycles. Not more. The forgetting is part of the mechanism, but a two-week gap turns cycle two back into cycle one.
How to run it without buying the book
You do not need 1,128 exercises. Honestly, most club players should not use 1,128 exercises, because cycle one becomes a two-month slog and the whole thing collapses somewhere in week five.
Build a set of 200 to 400. Pick positions you can crack with real effort in two or three minutes, which usually means around your puzzle rating or up to 100 points below it. Anything you cannot solve at all in cycle one will still be unsolved in cycle six, and it will wreck your timings on the way.
Then you need somewhere to keep it. An adaptive trainer is the wrong tool for this particular job, because it hands you fresh positions by design, which is exactly what you do not want. A study you assemble yourself, a saved collection in a puzzle app, a course built around a fixed list, even a text file of FENs with solutions: any of these work as long as the set stays identical from cycle to cycle. That constraint is the method.
There is a variant I like better than the classic one for anyone under 1800. Instead of one giant mixed set, build six themed sets of 50, chosen from the patterns you keep missing. A month of deflection alone does more for your deflection than a year of mixed puzzles, because mixed sets let you dodge your weakness by accident.
8 6 
c d e 
f h 1 In a mixed set you meet a queen check like that once. In a fork set you meet forty of its cousins inside a month, and after that you stop missing them, including the ones with the bishop on a different square and the king already castled.
Does the Woodpecker Method work?
Let me separate what is proven from what is marketing.
Proven: Tikkanen’s norm run happened, and Smith’s climb from 2100 to GM as an adult happened. Both are real, documented, and unusual.
Less proven: that you will get the same thing. There is a confound in the origin story that the book is fairly honest about. Tikkanen did not just solve the set, he built it. Finding, vetting and annotating a thousand tactical positions is hundreds of hours of extremely deliberate chess work that a reader of the finished book never does. Some of his jump came from the woodpecking. Some came from the labour of assembling it.
What club players actually report, once you filter out the enthusiasm, is 50 to 80 rating points over a couple of months, with the most consistent feedback being fewer one-move oversights and more comfort in sharp positions. That is a real result. It is not a transformation, and anyone selling it as one is selling something.
The standard objection is memorisation: by cycle five you are not solving, you are remembering. That is half true and mostly fine. Remembering the shape is the goal. Remembering the coordinate is the failure mode, and the difference shows up in whether you can still say why the natural alternative loses. So on later cycles, keep checking the refutation, not just the answer.
Who it suits: roughly 1400 to 2200, players who already know the basic motifs and want them instant. Below about 1200 you are better off learning the shapes first. Above 2200 the set has to be genuinely hard or you are just doing calisthenics.
Who should skip it: anyone who knows, honestly, that they will quit in cycle three. A 120-puzzle set finished beats a 500-puzzle set abandoned, every single time.
Five ways people ruin it
The set is too big, so cycle one takes seven weeks and motivation dies before the payoff.
The set is too hard, so a third of the positions never get solved and every cycle carries the same dead weight.
Cycle one is rushed. People guess, click the solution, feel like they learned something, and build the entire structure on sand.
The set gets edited halfway through. Adding twenty puzzles in cycle four means those twenty are on cycle one while everything else is on four; the repetition schedule quietly stops meaning anything.
It gets run during a tournament stretch. Cycle five of a 300-puzzle set is a genuinely draining day, and you do not want it sitting between two classical games.
The lighter version most players should do
If the full thing sounds like a lot, it is, and there is a smaller version that captures most of the benefit.
Take 50 puzzles on one theme you are weak at. Removing the defender, say, or back-rank mates. Four cycles: two weeks, one week, three days, one day. Then switch themes and do it again. Six themes gets you through most of a year and covers the tactics that decide club games.
And keep a second, permanent set: every puzzle you fail, anywhere, goes in it. Woodpeck that set once a month. It is the cheapest training in chess, because a puzzle you missed is diagnostic information about your blind spots, and most players throw it straight in the bin. Pair it with the right daily volume and a trainer whose themed drilling you actually use, and you have a training plan that survives contact with real life.
Repetition is not a fashionable answer. It does not sound like a secret. But if you have been stuck at the same rating doing fresh puzzles every day and wondering why nothing sticks, the woodpecker has been pecking the same spot this whole time for a reason.