(2023) Smithy's Review of "The Woodpecker Method"

From approximately December 2018 into March 2019, I went through the infamous "Woodpecker Method" on Chessable.  I'm sure by now the method is well known, but for completeness, the book gives you a large set of exercises and asks you to repeat them over and over.  If you run through the 700+ problems seven times, you are supposed to do it faster, do it better, have more confidence, become a GM and nab the girl of your dreams... or at least that's what the marketing spin tells you.

I recorded the entire journey on my old blog ... which has since disappeared when I cancelled my hosting plan.  Chessable doesn't record user stats, and my old notes have vanished when my ancient computer finally gave up the ghost.  I won't be able to point to any hard stats as I would like to, but my conclusion will still ring loud and clear.

I really didn't like the Woodpecker Method.

To start, I should quickly distinguish three different factors: there's the book itself, there's the recommended method and then there's the Chessable conversion.  Let's start there.

The Chessable Version

I have long-since fallen out of love with Chessable, but this experience first showed the limitations to the website.  Most relevantly, you cannot train any of the sidelines.  Given that this is a tactics book and most puzzles have multiple lines, this is a huge limitation.  Worse, the mainline was sometimes arbitrarily chosen, so that you did not get to play through other critical variations.

This is really bad from a training perspective.  The whole gimmick is to go faster and faster through these problems.  If you aren't being tested on sidelines, you are just going through on autopilot.  You start memorizing moves.  I distinctly remember wondering what happens if Black plays X, but I knew it didn't matter and I wanted to beat my old time, so I just ploughed ahead with the move I "knew" was right.

There are many minor annoyances as well:

• Considering the book is all about tracking time and improvement, it is amazing that Chessable does not track any of these stats.  Unreal, really.  
• For a book all about speed, Chessable server lag added 1-5s between problems; that doesn't like much, but across 700 problems, that adds nearly an hour to your time.
• Chessable's normal spaced repetition is ill-suited, and their attempt at "cyclical" repetitions is clumsy.  There's a tension with how users wants to use the book and how Chessable is designed, in other words.
• Relatedly, Chessable uses its normal points system, rather than the intended "checkmark" scoring system. I don't think either makes a different, but there is a sense that Chessable is not presenting the book the way it is meant to be trained.
  
• Finally, many problems have poor "soft fail" functionality, where equally good moves are marked wrong. Arguably, this is a fault of the book, because puzzles should ideally have zero equally winning moves, but I'll get to that soon enough.

In short, for a $40 chess product, the Chessable version does not give you much, to the point I legitimately wonder if I would have had a better experience with a non-digital version.

The Book and the Puzzles

I have high standards for tactics books.  After all, the Internet has an infinite amount of puzzles available for free.  A book needs to do more than just have a bunch of puzzles. At the very least, a book needs to do one of the following:

• curated, with interesting or instructive positions
• organized by theme or difficulty
• include tips, tricks or other instruction to assist the solver
• explain why tempting moves are wrong

Charitably, the book does one of these: problems are grouped by difficulty... but not very well.  There is huge variance within the "Intermediate" section, where a 7-move combo is sandwiched between two simple forks.  The final 10% of each section is consistently harder than the first 10% of the next section.

This is exasperated by the heavily materialistic focus of many problems, which involve winning a sole pawn.  I have no problem with this in isolation.  In many instances, though, we trade our good Bishop and other prominent attackers to get that pawn.  Other times we grab a pawn that opens lines for our opponent.  Is that worth it?  Unfortunately, there is zero discussion of these downsides, and it got to the point where if I saw a way to win a pawn in a messy position, I did it.  Invariably it was the right answer, but I did not get there with the right reasoning.

Equally exasperating were the "red herring" problems, ones where there are no tactics and we are just supposed to play a normal move. Again, I have nothing against this in theory: we shouldn't assume every Queen sacrifice works just because it is a tactics book.  However, there were only a handful of these, approximately 7 out of the 700+ intermediate problems.  That's 1%.  If we are training to spot whether a tactic exists or not, we need at least 20% of the problems to be red herrings.  I found absolutely no benefit in wracking my brain every hundred problems trying to find a solution that didn't exist.

In other words, a book filled with 25-50% red herrings would be interesting, because it forces you to evaluate every problem like a regular game position.  That would be useful training.  When the number is 1%, it just becomes an annoyance.  Worse, these red herrings go against the entire idea of Woodpecking.  You are supposed to ingrain these tactical patterns and improve your vision.  How does having a non-tactic help?  It doesn't.  I got to the point where I remembered, "Ahh, this is the stupid position where the solution is 1.h3 because nothing else works."  That's not helping me improve. Red herrings are fundamentally at odds with the Woodpecker Method.

Finally, the book does not meet any of my other requirements listed above.  The positions are indeed curated from World Champion games, but aside from historical curiosity, this does not help with training or improvement.  There was no discussion on how to find the right moves, and very little text on why tempting moves, or different move orders, were inferior.  Ironically, all of this combined likely makes the actual "Woodpecker Method" book a poor choice for using the actual Method.

The Method

Why repeat tactical problems?  The idea is to ingrain the patterns deeper into our brains, to enhance our pattern recognition.  The more we do the problems, the more these patterns become automatic.  That makes sense to a certain degree, but I wonder, is it better to do 50 knight fork puzzles 4 times, like the Woodpecker Method, or to just do 200 knight fork puzzles total?  Both have the same number of reps, and both likely take the same amount of time.  Will one really work better than the other?

I don't know.  Certainly for the fundamental tactical ideas, so forks and pins and such, you will absorb the patterns either way, and arguably the greater variety from 200 unique positions will help more.  That said, if you have an excellent curated sample of thematic tactics (for example, the checkmate motifs from CraftyRaf's Checkmate Patterns), then I can see the value of repeating them.  Either approach works.

When you move beyond the basics and into harder calculations, the benefits of the method wither, in my view.  If you have the iron discipline to go through these harder puzzles with proper calculation every time (eg, calculating all the lines in the proper order, at the proper depth, with no shortcuts), then I think Woodpecking could help train proper calculation technique.  I certainly didn't have the discipline for it, and neither do most people based on what I've read.  Because the central theme is increasing speed, the method naturally encourages shortcuts, of relying on memory rather than calculation.  You would have to drop the speed component and couple it with amazing discipline to make it work.  At that point, why not just do unique puzzles and avoid the needed disciple?

Now, again, I've argued on various forums that memorizing tactics isn't necessarily bad.  If you can memorize 10, 20, 50, 100+ puzzles, that has to have an effect on your chess.  I don't know how big, but it has to be more than zero.  The question becomes, especially with these harder problems, whether the time investment is better spent on repeating (and thus memorizing) known problems or solving new ones.  Unquestionably solving new problems is good; the other way, I don't know.

My Experience

All told, I invested 70 hours in this book, and I largely regret it.  With the exception of bullet, none of my online ratings improved much.  When I consider my time investment, that seems a searing indictment.  Surely those 70 hours could have been spent better, with greater return.

It also completely burned me out.  I suffered a type of chess exhaustion I had never felt.  I didn't even want to look at the board, let alone play a game.  Indeed, I stopped playing for about three months.  Doing the complete Method, where you go through 700+ problems in less and less time, became a second job.  I would not recommend that at all.  Repeat problems, sure, but don't put a time limit on it.

I experimented with a much lighter version afterwards: repeating the Checkmate Patterns course, basically doing a chapter day with no pressure to go faster or to do X positions a day.  I do think this helped solidify the patterns in my brain, but ... it soon felt almost lazy, like I was just going through the motions. It may have been a useful warm-up, and it kept the patterns fresh in my mind, but at best I was maintaining my skill level, not improving.

Now, I do think repetition has a role in improving.  I had very few chess books growing up, so I was forced to read the same material over and over.  I got to a respectful 1800 online in just a few years, so the repetition certainly didn't hurt.  It may even have helped in that initial beginner phase.  It remains an open question how useful it is afterwards.

Conclusion

Should you buy the book?  No.  There are better sets of puzzles out there, not to mention infinite free puzzles online.  The book is not curated enough, and not filled with enough instructional value, to justify the large cost, in my opinion.

Should you try the method, of repeating puzzles multiple times?  Maybe.  I don't think it hurts, but I don't know if it helps much outside the fundamentals.  The focus on speed can create bad habits, especially with calculation-focused puzzles.

My original review from 2019 was far less damning.  Believe me, at the time, I did not enjoy it, but I'm less sure why I hedged my criticism so much.  I wonder if it was simply a sunk cost: I had invested 70 hours and didn't want to admit it was a mistake.  With the benefit of time, I can safely say it was.

I would not do it again.  I would rather spend those 70hrs doing all unique puzzles, or, perhaps, running through a very large set of difficult puzzles twice, to see if I improve my accuracy (not speed) and to ensure it is all calculation, not memory.  That may be an experiment for a different day.