Student: “I thought I played well, but I only got 60% accuracy”.
Me: “But that was still an awesome game!”
This conversation happens often in my lessons.
Accuracy scores can be fun, but they can also give us a misleading sense of how well we’ve played. Often, students play good practical chess but get a low score – or they blunder early on, then fight in a hopeless position and get a high score.
In both cases, they get the wrong takeaway from the game.
To better understand why accuracy scores don’t fully reflect your progress, let’s take a look at a real-world analogy.
Real World Example
Who would you prefer to be, Investor A or B?
- Investor A makes 99 perfect decisions out of 100. They build up a lot of wealth. They just have one problem: every 100th decision is a massive mistake. They fall into a crypto scam or gamble all their money away in a casino on a single day.
- Investor B gets a lot of stuff wrong. From 100 investments, 30 turn out failures. Knowing that investing is tough, they never invest all their money and so can afford to lose 30% of the time. The 70% they win clearly outperforms the losses.
Investor A would be the clear winner if you calculated their accuracy score. But after 100 decisions, they are bankrupt. One big mistake weighs more than 99 great decisions.
This is why Warren Buffet’s most famous investment rules apply perfectly to chess:
Rule 1: Never Lose Money. Rule 2: Never forget Rule No. 1.
The same happens in a chess game. Not every decision is equal. Getting most moves right doesn’t matter if you royally screw up on 1-2 moves per game. And thus the accuracy score doesn’t necessarily indicate good play.
What Is A Good Accuracy Score?
I see this question a lot online. Sadly, there is no answer to it. You can get a 90% accuracy score, but I’d rate your play as bad, or you can get 60%, but I, as a Coach, was super happy with how you played.
Relying too much on this number to judge how well you played can be misleading, as it often leads to the wrong takeaways. Remember, your goal is to improve, not to chase a perfect score. Don’t take it from me, Lichess writes on their page explaining accuracy scores:
“A very high accuracy percentage isn’t necessarily indicative of superhuman, “GM-level” play.”
And:
While there is some correlation between the players’ ratings and their accuracy, it is not straightforward. A more skilled player tends to play more principled, theory-heavy openings and put more tactical pressure on the opponent. This can create more complicated positions and provoke inaccurate play on both sides. Moreover, lower-rated players are often more reluctant to resign. As we discussed above, protracted lopsided endgames can increase the accuracy score.
In other words, playing for 100 moves a Queen down, you can play “perfect” chess for a long period and thus get a high accuracy, but the practical value of that is… 0.
You can also get better at chess, and your accuracy drops.
The accuracy score also depends on the opening you get on the board, how well your opponent plays, how long the game went… Most of those things are out of your control.
As there is no definite, real correlation between your accuracy score and whether you can be happy with your game, I recommend simply ignoring this score altogether.
So, what should you focus on instead?
What Really Matters
Three main things matter in every single game you play. Of course, there are more once you go more in-depth, but on a surface level, think about these 3:
- Your process – How well did you focus during the game? Did you apply the blunder check on every move?
- Blunders (??) – Did you miss any big, obvious things? Anything that shifted the Engine evaluation drastically – when it still mattered (nobody should care about going from -5 to -15 or going from +15 to +10).
- Missed Chances – Were there any big tactical oversights by your opponent that you didn’t punish?
Why Do Accuracy Scores Exist Then?
Gamification is a great tool for making chess engaging and accessible, but it’s important to remember that not all of it serves the purpose of improving your game.
When commenting on a game, it is great to say:
“Wow, the players are playing basically perfect chess, 99.8% accuracy!!”
Then, when you play on your own, you get a game with 95% accuracy – which makes you feel on top of the World. You nearly played like a top GM!
So, you continue playing and get more drawn into chess. That, in turn, is good for chess sites – and chess in general. The incentive on chess sites is to make you a repeat user. The more you play, the better for them.
If your goal is to improve your game, then your goal doesn’t fully align with their incentive. That doesn’t mean they try to trick you.
It just means that what is best for the site – getting you to play again – isn’t necessarily great for your chess improvement. So what has worked for them to get more consistent players – gamifying your game analysis with an accuracy score – is not best for your chess improvement.
Keep improving,
GM Noël
PS: This article was initially sent out to my Newsletter list. If you want to get chess improvement advice for free in your inbox, join 17,000+ chess improvers by signing up for Friday Grandmaster Insights here.
Whenever you’re ready, here is how I can help you:
- Want to know How to train chess well? Check out Next Level Training – The Chess Training Blueprint for Adult Improvers. This course taught 800+ students the How of Chess Training. Create your high-quality chess plan and learn how to study each part of Chess, from tactics to openings & endgames. Click here to learn more.
- Rated below 1200 Chess.com? Need to refresh your fundamentals? Check out my course, Beginner Chess Mastery. You’ll learn all the fundamentals, from strategy to how to get the most out of your pieces, tactics, and endgames. You even get a full opening repertoire for free. Click here to learn more.
