- Blind mode tutorial
lichess.org
Donate

The Woodpecker Method, Revisited

"repetitio est mater studiorum"
When I repeteadly solve the same pattern -- I remember the ideas. So, when I see that pattern in another position, I can see it. For example: I started solve puzzles at simplest level and there was a lot of puzzles with Boden checkmate (also known as criss-cross mate). Then I continued on the next level. And the puzzles are more complicated. To see just one move is not enough: I need to sacrifice the Queen. When I met it first time it was a problem, but next puzzles I've done faster. And after repetition I remembered this idea. Same story is with the smothered mate.
Why not solve it once? I don't know, it just doesn't works for me. I need to solve puzzles repeteadly to remember the ideas. Maybe the point is in the fact that I solved more then 10000 puzzles for the last six months (first 1000 puzzles I solved three times and second 1000 - seven times).
And yes, I have great progress in tactics for the last six months.

"repetitio est mater studiorum" When I repeteadly solve the same pattern -- I remember the ideas. So, when I see that pattern in another position, I can see it. For example: I started solve puzzles at simplest level and there was a lot of puzzles with Boden checkmate (also known as criss-cross mate). Then I continued on the next level. And the puzzles are more complicated. To see just one move is not enough: I need to sacrifice the Queen. When I met it first time it was a problem, but next puzzles I've done faster. And after repetition I remembered this idea. Same story is with the smothered mate. Why not solve it once? I don't know, it just doesn't works for me. I need to solve puzzles repeteadly to remember the ideas. Maybe the point is in the fact that I solved more then 10000 puzzles for the last six months (first 1000 puzzles I solved three times and second 1000 - seven times). And yes, I have great progress in tactics for the last six months.

@DailyInsanity said in #11:

I'd think that taking 5-30 seconds on a puzzle is still 'too slow' to have considered the corresponding pattern truly learned, in which case I think repetition would still be beneficial.
Do you think it's possible for club players to learn (instantaneous pattern) positions that are say 2200+ puzzle rating on lichess? At some point everyone has to switch from System 1 to 2. This is noticeable if you check Andrew Tang's puzzle storm runs on stream for example.

@DailyInsanity said in #11: > I'd think that taking 5-30 seconds on a puzzle is still 'too slow' to have considered the corresponding pattern truly learned, in which case I think repetition would still be beneficial. Do you think it's possible for club players to learn (instantaneous pattern) positions that are say 2200+ puzzle rating on lichess? At some point everyone has to switch from System 1 to 2. This is noticeable if you check Andrew Tang's puzzle storm runs on stream for example.

Like another poster said, my main issue with the Woodpecker Method is that the spaced repetition should decrease rather than increase in terms of time. This is the wrong way around for long term memory recall.

It's good to increase the spacings if, for example, you have a tournament coming up and you want to get yourself very sharp, like cramming for an exam, but it makes little sense if you want to convert learning from the short to the long-term memory.

I also disagree with the original blog post (well written, good post) in terms of repetition of the same problems. Like others have said as well, I think it's very beneficial to repeat the same positions in order to really drill that pattern. Mixing up the pattern at the point of training I think just muddys the waters before it has a chance to cement.

In terms of speed, yes you absolutely need to be solving simple tactics within 5-10 seconds - instantly really in order to really have that pattern cold.

Interesting about the second book with the positional patterns. I don't know anything about it but it could be very good. I do think there is a gap with simple positional puzzles.

Like another poster said, my main issue with the Woodpecker Method is that the spaced repetition should decrease rather than increase in terms of time. This is the wrong way around for long term memory recall. It's good to increase the spacings if, for example, you have a tournament coming up and you want to get yourself very sharp, like cramming for an exam, but it makes little sense if you want to convert learning from the short to the long-term memory. I also disagree with the original blog post (well written, good post) in terms of repetition of the same problems. Like others have said as well, I think it's very beneficial to repeat the same positions in order to really drill that pattern. Mixing up the pattern at the point of training I think just muddys the waters before it has a chance to cement. In terms of speed, yes you absolutely need to be solving simple tactics within 5-10 seconds - instantly really in order to really have that pattern cold. Interesting about the second book with the positional patterns. I don't know anything about it but it could be very good. I do think there is a gap with simple positional puzzles.

I do wonder whether the advent of very fast chess, right up to bullet and ultra-bullet, has skewed the perception and indeed the reality of how fast one needs to be at seeing tactics. Of course, faster tactics are always better if they are accurate and if they do not come at the opportunity cost of neglecting other forms of necessary training. However, training for speed and ultra-speed may indeed come at the opportunity cost of neglecting other aspects of chess in early learning. I would single out deep visualisation and deep calculation (neither of which I possess), positional and strategic understanding (neither of which I possess) openings, middle games and endings theory (bad at all these too).

Down at my beginner/rookie level, all those extra capabilities are moot. My most fundamental tactics and my slowness at seeing them are so egregious that I need not worry for a while (or ever?) about doing a large volume of fast puzzles and so causing a serious opportunity cost on the other training I should be doing. And the faster the time limit I play, the less of a concern this opportunity cost becomes.

However, I would feel it pointless for me play 10,000 chess games, not learn anything deeper than 1 and 2 move tactics and get a bullet rating of 1200 (which is about what I would get on that path. That sounds like purgatory or worse to me. Thus, while doing lots of fast tactical puzzles, I feel I have to learn a bit more about the other aspects of chess and try to apply all my learning in the slowest time limit available on auto-match which is Classical 30 + 20. I am not even playing enough games yet. I am trying to set up my fast tactics puzzle regimen first. I am spending as much time researching chess learning and finding what works for me as I am spending on the actual tasks. I see this overhead as unavoidable at this stage or I will otherwise spend months or years training and playing in the wrong way and/or in a very inefficient way.

I do wonder whether the advent of very fast chess, right up to bullet and ultra-bullet, has skewed the perception and indeed the reality of how fast one needs to be at seeing tactics. Of course, faster tactics are always better if they are accurate and if they do not come at the opportunity cost of neglecting other forms of necessary training. However, training for speed and ultra-speed may indeed come at the opportunity cost of neglecting other aspects of chess in early learning. I would single out deep visualisation and deep calculation (neither of which I possess), positional and strategic understanding (neither of which I possess) openings, middle games and endings theory (bad at all these too). Down at my beginner/rookie level, all those extra capabilities are moot. My most fundamental tactics and my slowness at seeing them are so egregious that I need not worry for a while (or ever?) about doing a large volume of fast puzzles and so causing a serious opportunity cost on the other training I should be doing. And the faster the time limit I play, the less of a concern this opportunity cost becomes. However, I would feel it pointless for me play 10,000 chess games, not learn anything deeper than 1 and 2 move tactics and get a bullet rating of 1200 (which is about what I would get on that path. That sounds like purgatory or worse to me. Thus, while doing lots of fast tactical puzzles, I feel I have to learn a bit more about the other aspects of chess and try to apply all my learning in the slowest time limit available on auto-match which is Classical 30 + 20. I am not even playing enough games yet. I am trying to set up my fast tactics puzzle regimen first. I am spending as much time researching chess learning and finding what works for me as I am spending on the actual tasks. I see this overhead as unavoidable at this stage or I will otherwise spend months or years training and playing in the wrong way and/or in a very inefficient way.

Footnote to my above post. I just played a 10+5 game. I lost badly. That's alright, all part of the learning experience. But what I noticed in the later analysis (after I messed up in the game, or so I thought, and dropped a pawn) was that the opponent's taking of the pawn was a blunder.

However, for me to prove it a blunder I would have had to see a surprising next move (a very subtle quiet move I think I should call it) which would then lead to at least 3 important forcing lines of 6 to 7 moves each. The best the opponent could do out these forcing lines (and I doubt anyone at my level would find it) would have been to return two pawns and be one pawn down and about +2.2 (he was black). Other lines would have returned scores like +5.5 or better for white... you know, with a 2500 player to find them on the attacking end.

The quiet rook move to the king rook file (black was un-castled) would have required seeing that the 4 pieces and pawns (or so) on the file would all be cleared by, among other moves, clearing sacrifices. How am I going to see that??? I wouldn't see it in a 3 hr game. I doubt I would even see it in old fashioned correspondence chess by the post. What use is Lichess's analysis to me? After avoiding hanging pieces, I am next supposed to see 6 move combos starting with an unusual quiet move? This doesn't help me. I am far too basic at chess for 90% of Lichess analysis to be of any use. There are many, many Lichess improvements to my inaccuracies and mistakes where I simply cannot understand why the Lichess move is better. I simply don't haven't the foggiest idea.

But in the absence of understanding the first thing about chess, then I guess a truly stupendous amount of fast puzzles and 1,000s of lost rapid games await me in my allotted year to try to understand the basics of chess.

Footnote to my above post. I just played a 10+5 game. I lost badly. That's alright, all part of the learning experience. But what I noticed in the later analysis (after I messed up in the game, or so I thought, and dropped a pawn) was that the opponent's taking of the pawn was a blunder. However, for me to prove it a blunder I would have had to see a surprising next move (a very subtle quiet move I think I should call it) which would then lead to at least 3 important forcing lines of 6 to 7 moves each. The best the opponent could do out these forcing lines (and I doubt anyone at my level would find it) would have been to return two pawns and be one pawn down and about +2.2 (he was black). Other lines would have returned scores like +5.5 or better for white... you know, with a 2500 player to find them on the attacking end. The quiet rook move to the king rook file (black was un-castled) would have required seeing that the 4 pieces and pawns (or so) on the file would all be cleared by, among other moves, clearing sacrifices. How am I going to see that??? I wouldn't see it in a 3 hr game. I doubt I would even see it in old fashioned correspondence chess by the post. What use is Lichess's analysis to me? After avoiding hanging pieces, I am next supposed to see 6 move combos starting with an unusual quiet move? This doesn't help me. I am far too basic at chess for 90% of Lichess analysis to be of any use. There are many, many Lichess improvements to my inaccuracies and mistakes where I simply cannot understand why the Lichess move is better. I simply don't haven't the foggiest idea. But in the absence of understanding the first thing about chess, then I guess a truly stupendous amount of fast puzzles and 1,000s of lost rapid games await me in my allotted year to try to understand the basics of chess.

@Wodjul said in #15:

However, training for speed and ultra-speed may indeed come at the opportunity cost of neglecting other aspects of chess in early learning. I would single out deep visualisation and deep calculation (neither of which I possess), positional and strategic understanding (neither of which I possess) openings, middle games and endings theory (bad at all these too).

Your strategic and positional understanding won't help, if you blunder in every game. Moreover, you won't get any use from master's games (full of strategic ideas) until you see tactics. Masters see tactics and implement their strategic ideas taking tactics into account. You won't implement what you study, until you see tactics.
So, tactics first. And if you want to be better in strategy, also study endgames. I mean, complex endgames (pawn endings with a lot of pawns, rooks and pawns endings, bishops and knights with pawns endings). Elemental engames are first step to complex endgames. When you understand how to play complex endgames you better understand how to play in middlegame to get better endgame.
I'm also beginner and I'm playing a lot offline. Last year I lost many games in middlegame despite working on my strategy and openings, but this year I have more endgames (in classics time control it's already usual) Tactics training saves from blunders. And woodpecker method helps me much better than solving random puzzles and better then solving puzzles on specific topics.

@Wodjul said in #15: > However, training for speed and ultra-speed may indeed come at the opportunity cost of neglecting other aspects of chess in early learning. I would single out deep visualisation and deep calculation (neither of which I possess), positional and strategic understanding (neither of which I possess) openings, middle games and endings theory (bad at all these too). Your strategic and positional understanding won't help, if you blunder in every game. Moreover, you won't get any use from master's games (full of strategic ideas) until you see tactics. Masters see tactics and implement their strategic ideas taking tactics into account. You won't implement what you study, until you see tactics. So, tactics first. And if you want to be better in strategy, also study endgames. I mean, complex endgames (pawn endings with a lot of pawns, rooks and pawns endings, bishops and knights with pawns endings). Elemental engames are first step to complex endgames. When you understand how to play complex endgames you better understand how to play in middlegame to get better endgame. I'm also beginner and I'm playing a lot offline. Last year I lost many games in middlegame despite working on my strategy and openings, but this year I have more endgames (in classics time control it's already usual) Tactics training saves from blunders. And woodpecker method helps me much better than solving random puzzles and better then solving puzzles on specific topics.

the problem is not the repetition or reexposure to the same pattern. It is about the context being always the same. That might make you learn the action sequence very well on that specific context, but then reduce the context where that automatism of move sequence would apply to the specific of that context too often repeated, it become a hammer.

I does depend on the context content is irrelevant information. But as the blog author mentions (if I remember right or read ok) there is not method being shared about the position set choice.

Also, I wonder what a pattern is now, in most people "definition". Is it only the action sequence?

The problem with the hammer. is that you have to take a swing. and one in motion.. well not exactly.

I think people may have different takes, with a hammer..

it might be that they get stuck with the exact context or not having many clues about the trigger patterns (the context part, the underlying static pattern of it), they might just do the hammer swing because it can legally work there, or can be attempted, and yet might not end up what it constitutes the gain as in the training.

others might wait for the exact context of poisitino information, to perform the hammer, while it might rarely be exactly the same thing on the static triggering that they are exposed to.

Also not speaking from chess expertise here, just from the problem of training, extrapolated to chess. And also those are not certitudes but suggestions to ponder oneself, about this. It takes time to see oneself own abliity or not to misgeneralize. If though everyone was playing the same exact repertoire, and the position set was from there, maybe then there would not be a problem. But we do not play like that, do we? There are novelties, etc...

keyword: generalizagtion from training to the reality

the problem is not the repetition or reexposure to the same pattern. It is about the context being always the same. That might make you learn the action sequence very well on that specific context, but then reduce the context where that automatism of move sequence would apply to the specific of that context too often repeated, it become a hammer. I does depend on the context content is irrelevant information. But as the blog author mentions (if I remember right or read ok) there is not method being shared about the position set choice. Also, I wonder what a pattern is now, in most people "definition". Is it only the action sequence? The problem with the hammer. is that you have to take a swing. and one in motion.. well not exactly. I think people may have different takes, with a hammer.. it might be that they get stuck with the exact context or not having many clues about the trigger patterns (the context part, the underlying static pattern of it), they might just do the hammer swing because it can legally work there, or can be attempted, and yet might not end up what it constitutes the gain as in the training. others might wait for the exact context of poisitino information, to perform the hammer, while it might rarely be exactly the same thing on the static triggering that they are exposed to. Also not speaking from chess expertise here, just from the problem of training, extrapolated to chess. And also those are not certitudes but suggestions to ponder oneself, about this. It takes time to see oneself own abliity or not to misgeneralize. If though everyone was playing the same exact repertoire, and the position set was from there, maybe then there would not be a problem. But we do not play like that, do we? There are novelties, etc... keyword: generalizagtion from training to the reality

The method actoully work me choach is GM and he recomend this methos now i playng so good.

Do not trust that guy.

The method actoully work me choach is GM and he recomend this methos now i playng so good. Do not trust that guy.

@Wodjul said in #15:

I do wonder whether the advent of very fast chess, right up to bullet and ultra-bullet, has skewed the perception and indeed the reality of how fast one needs to be at seeing tactics. Of course, faster tactics are always better if they are accurate and if they do not come at the opportunity cost of neglecting other forms of necessary training. However, training for speed and ultra-speed may indeed come at the opportunity cost of neglecting other aspects of chess in early learning. I would single out deep visualisation and deep calculation (neither of which I possess), positional and strategic understanding (neither of which I possess) openings, middle games and endings theory (bad at all these too).

The point of the woodpecker method is that most of our decisions in chess come from our intuition: in any given position there are far too many possible moves for us to consider purely analytically. That's why even when engaging in calculation, we still need to use our intuition, to separate the wheat from the chaff.

@Wodjul said in #15: > I do wonder whether the advent of very fast chess, right up to bullet and ultra-bullet, has skewed the perception and indeed the reality of how fast one needs to be at seeing tactics. Of course, faster tactics are always better if they are accurate and if they do not come at the opportunity cost of neglecting other forms of necessary training. However, training for speed and ultra-speed may indeed come at the opportunity cost of neglecting other aspects of chess in early learning. I would single out deep visualisation and deep calculation (neither of which I possess), positional and strategic understanding (neither of which I possess) openings, middle games and endings theory (bad at all these too). The point of the woodpecker method is that most of our decisions in chess come from our intuition: in any given position there are far too many possible moves for us to consider purely analytically. That's why even when engaging in calculation, we still need to use our intuition, to separate the wheat from the chaff.