How Should We *Actually* Learn From Our Mistakes?
Never waste a good blunder.
Hello from Los Angeles, where everything looks… weirdly normal. It reminds me a little of a “daylight horror” movie like Midsommar — everything SEEMS bright and beautiful, but underneath there’s… evil. Or just ash. Regardless, I’m quite glad to be back home.
Per last week’s newsletter, my OTB chess club is still closed — though we’re back next week! — so I’m going to be focusing on some ideas drawn from my online rapid play for the time being.
The headline news is that I crossed the 1700 rating for the first time… hooray! I’ve always harbored the aspiration — the dream — of being in the top 1% of chess.com’s rapid pool, and at the 98.6th percentile, I am quite close. Of course, as soon as I get there, I’ll then want to be in the 99.9th percentile. And so on. Such is the madness of the human mind, or at least my human mind.
I spoke last week about my reservations with online play. But this week, I want to focus on what I think is the best thing about playing online, and to dive into a chess practice of mine that I believe is applicable to improving at just about every activity you could possibly undertake.
To reiterate something I’ve talked about before: I have a bit of a compulsive relationship to playing online. The constantly changing number is just too irresistible: I WANT TO MAKE THE NUMBER GO UP, and as we all know, “number go up” is a life philosophy that has no negative side-effects whatsoever. This means that my intention to play one game will often turn into two, and then two more later to make up for the ratings losses that I incurred. Some people might call this “tilt” — I call it “just another Tuesday!”
I’ve always assumed that this is how just how chess players are: we’re all junkies looking for the next hit, the next brilliancy or checkmate or beautifully played endgame. But I was listening to Nate Solon’s seminar about creating a study plan for the year — maybe I will write about my own study plan in the future — and Nate talked about how this is far from the case; in fact, many of his students don’t like playing, and they prefer to study instead, safe from the soul-crushing disappointments of one-on-one combat.
This feels like an extreme personality-type thing that isn’t worth getting into here. But it made me appreciate my own tendency toward playing, and it drove home the value of such a habit:
The more reps you get, the more mistakes you make.
“The more mistakes?” you might say. “But aren’t mistakes BAD?”
NO. Mistakes are good. Mistakes are great. Mistakes are the universe saying to you, “Look at this, idiot. Look closely. Rub your face in your failure and understand.”1 This quote from Michael Jordan gets trotted out pretty often, but I still think it’s incredibly telling: “I’ve missed more than 9,000 shots in my career. I’ve lost almost 300 games. Twenty-six times I’ve been trusted to take the game-winning shot and missed. I’ve failed over and over and over again in my life. And that is why I succeed.”
Many of us — certainly me — have the tendency to assume that the best practitioners of any given craft never screw up. That the Coen brothers never hit a wall, that Tom Brady never failed to complete a game-winning drive, that Albert Einstein was never completely and totally wrong.
Of course, they are — that isn’t the news-flash here. It’s also not a revelation that you have to keep on trucking regardless. It’s not even really a huge leap to argue that, contrary to popular belief, accumulating a large amount of mistakes is actually good. After all, LeBron James has missed the most field goals in NBA history. Kobe Bryant is second. The top 25 list of most interceptions thrown in NFL history is mostly made up of current or future Hall of Famers.
The point I want to drive home is this one: if improvement is a combustion engine, then mistakes are the gasoline. You literally can’t improve without them.
Here’s a good basic way to measure how well your improvement machine is working: when you pour mistakes into it, does it produce positive returns? If so, then you’ve got a finely calibrated engine. If not — if you don’t know what to do with your mistakes — then you have a lot of room not just to improve, but to improve your improvement.
Chess provides an excellent example of how to do this — because chess tells you exactly what your mistakes are. Thanks to Stockfish, I can go back through a game I’ve played and know exactly (or exactly enough) where I made a mistake. This is different than other sports.2 An interception or a missed field-goal isn’t necessarily a mistake; the interception could’ve resulted from wide-receiver error, and when even the top NBA players only shoot 40 percent from three-point range, you’re expecting more of your shots to not go in than to go in even when you’re the best of the best. So you end up looking at aggregates and having to learn from a larger sample-size, which takes a bit more sophistication. Same deal with something like Brazilian jiu-jitsu or long-distance running or weightlifting: causation isn’t as easy to separate from correlation.
But in chess, the computer knows. You can’t argue with it. It would be like arguing with the temperature outside.3
This might sound dispiriting to the romantics among us: alas, now we mourn the death of a once-proud art, etcetera etcetera. But, as the world-championship match this year demonstrated, even with unbelievably good chess computers, there’s still room for innovation and surprise. And, again, at lower levels, it just means absolutely nothing. It would be like saying that, just because cars can drive faster than Usain Bolt can run, we should stop having footraces.
Okay, so: the computer tells you your mistakes. Great, love it, thanks robot. Now here’s the most important part: you need to annotate your game. I’m far from the first person to extol the benefits of this. It’s a cornerstone of the Chess Dojo’s training program, and it’s a big part of Nate Solon’s advice as well. (Nate has a terrific primer on how to annotate your games.)
But I want to focus on how I do it — which I credit with having helped me improve quite a bit over the last few months — and why I think it’s particularly valuable as far as this exact idea of learning from your mistakes. My method has three steps to it:
you need to figure out why you made the mistake — what were you thinking?
you need to describe why it was a mistake. And:
you need to articulate the lesson behind the mistake — what you should do differently in the future.
This is where it gets interesting. It’s easy enough to click through your chess games, see that you made a mistake or a blunder, and be like, okay, cool, I get it. But this won’t do you any good. You won’t remember. You’ll keep making the mistakes over and over and over, and you’ll wonder why you aren’t getting any better, and then you’ll quit chess and start playing Fortnite or whatever it is the kids are playing these days.
Here’s a 10+0 rapid game I annotated just a few days ago.4 The important thing is that I made quite a few mistakes in this game, giving me plenty to work with.
Repasting my comment here so it’s easier to read:
My worst move of the game. I knew it would be tough to get my knight out of there but I didn't see Nd5, which is a nifty double-attack. Strangely, the best move here isn't that hard to find: it's b5, protecting the knight. I literally just wrote a newsletter two weeks ago about protecting your pieces!!! Worst of all, I had considered it but written it off because of how it would open up the h1-a8 diagonal, but even if White went Bf3, I could've responded with Bb7.
What am I doing here? It’s pretty straightforward: I’m telling a story about my mistake. I’m narrativizing it. I’m giving it a beginning, middle, and end, explaining why I made the move, what was wrong with it, and the larger chess concept that it falls under. (In this case, protecting your pieces.) In doing this, I’m turning my mistake into a lesson for myself, a lesson that I might actually remember.
Because that’s the tough part, isn’t it? Not just in chess, but in life: how do you do things differently the next time around? I’m as guilty of this as anyone: there are certain things that I’ve been trying to do for the better part of a decade that I still haven’t managed to accomplish, in large part because, every time I set out to try, I somehow manage to convince myself that this time it will work, this time I won’t run into the same obstacle that I’ve run into every time before. Instead of changing my behavior in a meaningful way in order to overcome that obstacle, I delude myself into believing that the obstacle just won’t be there anymore.
Unlike in this unnamed pursuit, though, chess is a place where I can easily hold myself accountable — because at the end of the day, it’s pretty low-stakes. There’s no part of my ego or identity wrapped up in whether I cross 1800 one day. (Well, it’s a small part of my ego.) So it’s easier for me to be honest about my failings. This is a major reason why I believe that hobbies and amateur pursuits are so essential for us: they give us a playground in which we can try out ideas that we can then apply to our real, high-stakes lives.
By the end of this one game, I managed to generate six different lessons (or, as Nate refers to them, takeaways). Click through the annotated game in the link above if you’re curious to see them in more detail, but to summarize:
When in a cramped position, trade pieces.
Don’t go pawn-hunting when you aren’t fully developed.
Don’t just think about your own tactics — keep an eye on your opponents’ tactics as well, especially against your queen.
Protect your pieces.
Calculate to the end of lines, not just to the part that you want to be true.
Play aggressively when you’re down material.
Obviously, I didn’t invent any of these ideas. But I also haven’t fully absorbed them, as evidenced by my chess. By reviewing my game and spelling out my thoughts, I can teach them to myself and subsequently make them far more likely to sink in.
I then try to play out a line that would’ve incorporated the lesson, to see it in action. Usually, if I’m analyzing my online rapid games, this is just looking at the engine lines, since I don’t want to spend half an hour annotating a 15-20 minute game.5 And finally, I write my takeaways from each game in my chess journal (yes, I have a chess journal), because research shows that that’s an even better way of retaining knowledge.
In doing so, I transform my online chess, which might otherwise be a mindless time-waster, into a formula: make mistake, identify mistake, learn from mistake.
Like I said: annotating games is a commonly advocated technique in chess improvement — though I suspect it’s often ignored — even if the approach I’m pushing here is heavier on explanation than it is coming up with a million alternate lines. While this definitely comes from my background as a writer — it really is a literary format; I’m kind of surprised that Nabokov never wrote a novel in the form of an annotated chess game — it would be beneficial for anyone.
But because this is a familiar technique in chess, it had escaped me until recently how you can annotate pretty much anything in a similar fashion, attempting to turn your mistakes into lessons. Try it: if you’re struggling at or trying to improve in something, whether it’s a hobby, a sport, or an art, take a notebook and write down what you perceive to be the errors you’ve made in the work you’ve done so far and why you’re making them.
Even better, do this with someone who you trust and respect, who can help you make sure that you’re seeing what are really mistakes and not just features of the practice. (For example: not writing a perfect piece on your first draft isn’t a mistake — that’s just the nature of writing.) Then write out the takeaways you’ve identified from those mistakes and how you can attempt to incorporate them on your next go-around.
I’m going to attempt this with that unnamed part of my life I keep running aground at: we’ll see if it works. In the meantime, let me know if you have any practice like this, either in your chess or in your other pursuits, and if you’ve seen benefits from it.
READING/LISTENING/WATCHING:
LISTENING: At the end of the year, I always try to catch up with the albums I miss, which tends to be a futile effort: suddenly trying to listen to like, 20 albums at once doesn’t exactly promote deep connection. But I’ve been really digging Death of a Momma’s Boy by Growing Stone, which I found via my favorite music critic Ian Cohen’s year-end indie list. Ian describes it as “A Sun Kil Moon album you don’t have to listen to in a private session,” which is pretty hard to improve on. Check out “The Gym” and you’ll quickly know if it’s for you or not.
READING: After many years of intending to, I finally got started on The Book of the New Sun, a sci-fi tetralogy by the American writer Gene Wolfe that is often heralded well outside its genre — the critic Larry McCaffery ranked it as one of the 100 best books of the 20th century, in between The French Lieutenant’s Woman and A Clockwork Orange. I’ve made it so far through The Shadow of the Torturer, the first volume, and it’s pretty mesmerizing: a dizzyingly beautiful and deeply imagined portrait of a world. Now, if you’re looking for plot, you might be better served elsewhere, but I do think the story is compelling — it just comes in a slow burn, and with many digressions about blind master-librarians and the customs of a professional guild of torturer-executioners.
WATCHING: I finally watched Barbarian. It ruled.
Until next week!
Ahem: I would imagine it goes without saying that there are some realms in which mistakes are not good — drafting governmental legislation, for example, or designing oxygen tanks for a crewed mission to the moon. Endeavors in which said mistakes might impact the lives of others. But I’d argue that just about anything else — even an Olympic performance or a championship game — can ultimately produce productive mistakes, even if they cost you victory in the moment.
Yes, I am referring to chess as a sport. It’s a hell of a lot more like mixed martial-arts or soccer than it is, I don’t know… Risk. Or Clue. Or Chutes and Ladders.
This statement becomes less true and more nuanced the better you get, but I think it’s pretty hard to challenge below, at minimum, master level.
You’ll see that I won even though I was getting destroyed at the end. My opponent disconnected when I had 20 seconds left and I won on time. Never resign!
My approach to annotating my online rapid games is different than how I do my OTB games, which I go through first without the engine and try to identify my own lines and mistakes. This has a lot to do with trying not to spend more time analyzing a game than I did playing it: if you’re *actually* going to analyze a game you spent 15 or 20 minutes playing, you’re only going to want to spend ten minutes doing so. On the other hand, a three-hour game during which you thought deeply about every move can easily support an hour of analysis.
That was really well written about why one should annotate one’s games. I’ve played chess since before engines and the advice that you should analyze your own games have been there in all these years. But strangely, with engines, it seems like less people analyze their games properly. Or is it because there are a lot more adult improvers now? Or perhaps my generation never was as good at annotations as I thought ;-)
I agree words can be more important than lines. Especially computer lines. But I believe it’s a good habit to add at least one line with at least five ply. And sometimes a move isn’t bad per se but it simply is one much better move in the position.
Nice article. You might want to consider making that library public so that readers can check your annotation.