Chess probably doesn’t make you meaningfully better at math. The idea is intuitive and widely repeated, but the strongest scientific evidence paints a disappointing picture: when researchers use rigorous controls, the math benefits of chess instruction shrink to near zero. That said, the story has some nuance worth understanding.
What the Largest Studies Found
The most definitive test came from the Education Endowment Foundation in the UK, which ran a large-scale trial of chess instruction in primary schools. The result was blunt: there was no evidence that the chess program raised children’s math scores on national exams one year after the intervention ended. The difference between chess students and control students was essentially zero. The same held true for science and reading, for boys and girls, and for students from lower-income families.
Two controlled experiments published in a 2017 study tested the idea more precisely. In the first, 233 third and fourth graders received 25 hours of chess lessons during school. In the second, 52 students went through a similar program. Both experiments compared the chess group against an active control group (students learning checkers or the board game Go, rather than just doing nothing). Neither experiment found a statistically significant difference in mathematical problem-solving ability. The effects were described as “minimal.”
These results matter because study design matters enormously here. Earlier, less rigorous research often told a more optimistic story. A meta-analysis pooling 24 studies with over 5,000 young people found a moderate positive effect of chess instruction on math performance. But many of those studies compared chess students against students who received no extra activity at all, which introduces a serious confound: any engaging new activity might produce a bump in test scores, whether it’s chess, music, or art.
Why Chess Feels Like It Should Help
The intuition isn’t crazy. Chess and math share surface-level features: both involve pattern recognition, logical sequences, and spatial reasoning. The geometry of the board, the numerical exchange value of pieces, and the need to think several moves ahead all feel mathematical. Researchers have proposed that chess could strengthen working memory, fluid intelligence, and concentration, all of which predict math achievement independently.
The problem is a concept psychologists call “far transfer.” Near transfer is when skills in one area improve performance in a closely related area, like learning Spanish helping you pick up Italian. Far transfer is when skills generalize across loosely related domains. A theory dating back to 1901 predicts that transfer depends on how many features two activities actually share. Chess and math share some features, but not enough for reliable transfer. The skills you build in chess tend to stay chess-specific. You get better at recognizing board positions, calculating piece exchanges, and planning attacks. You don’t automatically get better at solving algebra problems or understanding fractions.
This same pattern appears across other “brain training” domains. Music training doesn’t reliably improve spatial reasoning. Working memory training games don’t reliably boost general intelligence. The brain is less of a general-purpose muscle than we’d like it to be.
One Approach That Shows More Promise
There is a twist. A study from Aarhus, Denmark, tried something different: instead of teaching chess as a standalone activity, teachers replaced one weekly math lesson with a lesson that taught math concepts through chess. The curriculum used the board and pieces as tools for teaching arithmetic, geometry, and problem-solving directly. This approach showed positive effects on math scores.
The distinction is important. Pure chess instruction, where kids learn openings, tactics, and endgames, doesn’t transfer to math. But using chess as a vehicle to teach actual math concepts can work, because at that point you’re still teaching math. The chess board becomes a manipulative, like blocks or fraction tiles, not a substitute for math instruction. Researchers who’ve reviewed the evidence suggest that chess-based math programs should explicitly incorporate the features chess shares with mathematics: the geometry of tactical patterns, numerical reasoning through piece values, and structured problem-solving.
What This Means for You
If you’re a parent considering chess lessons hoping they’ll boost your child’s math grades, the honest answer is that chess alone is unlikely to do that. Twenty-five to thirty hours of chess instruction, roughly what a school-year program provides, doesn’t produce measurable gains in math problem-solving when tested against proper controls.
That doesn’t mean chess is worthless. Chess builds concentration, teaches patience, and develops strategic thinking within its own domain. Kids who enjoy it are engaged in a complex intellectual activity, which is valuable on its own terms. But if math improvement is the specific goal, time spent on math instruction or math-enrichment activities will be more effective than time spent learning the Sicilian Defense.
If you’re an adult wondering whether your own chess hobby is sharpening your quantitative thinking, the same logic applies. You’re getting better at chess. That’s a real cognitive achievement. It just doesn’t reliably spill over into balancing your budget or interpreting statistics. The skills are more domain-specific than the popular narrative suggests.