New Study Reveals How To Best Improve On Chess.com

What’s the best way to improve at chess? A new study from Stanford University sheds light on the most effective practices for players across various skill levels.

The study (PDF version here) Deliberate Practice and Chess Mastery: Analyzing Activity Impact Across Skill Levels on Chess.com, conducted by Stanford University students Divya Bade, Shivam Patel, Chanwool Leem, and Yedu Pushpendran, analyzed the activities of 96,000 established Chess.com users between March 1 and October 1, 2024. 

The findings indicate that persistent practice on Chess.com leads to significant improvements in rating. However, users who engaged in a mix of various activities, beyond just playing, had significantly higher rating gains than those who only played games.

According to the authors, the data shows that for Beginners (0-500 rating) structured lessons were the most effective feature for rapid improvement. Beginners saw gains of up to 53 points in the first four months after reaching a stable rating by focusing on lessons. 

For intermediate players (500-1000 rating), Game Reviews provided strong initial benefits, leading to an average gain of 31 points. As you progress, Lessons also become increasingly valuable, contributing to a 41-point improvement.

Advanced players (1000-1600) had the most benefit from Game Reviews, with gains of 20 rating points. The results also shows that lessons have a notable impact in later periods (28 points).

Expert players (1600+) saw the smallest improvements, and the study suggested these players need more advanced learning materials to improve. Lessons appeared to be the most effective tool,  although the study noted that the results were statistically insignificant, likely due to a low sample size.

The study concluded:

Beginners find lessons useful as they get familiar with the game. Intermediate and advanced players benefit from analyzing their in-game play and conducting lessons on strategies they may not be familiar with.

Another important finding, was that Chess.com's Premium subscribers consistently improved faster than non-subscribers:

• Beginners with Premium gained an average of 49 rating points more than free users.
• Intermediate players gained 45 points more.
• Even the more Advanced users saw an effect of 33 rating points.

The impact of Premium features diminish over time as a user becomes more advanced, but are still materially significant. "This makes sense intuitively as it is harder to see outsized rating changes as an advanced player vs. beginner," the study noted.

"We can confidently conclude that being a subscriber results in significant rating improvement compared to not being a subscriber."

We can confidently conclude that being a subscriber results in significant rating improvement compared to not being a subscriber.

—Authors of Stanford study on chess improvement

The study attributes this to early access to tools such as unlimited puzzles, complete lesson libraries, and game reviews, which helps players adopt good study habits during their most formative months.

"This analysis validates Chess.com's core value proposition that deliberate practice leads to faster chess skill development, with clear evidence that premium features deliver measurable rating improvements across player segments," the study noted.

Finally, the authors made several suggestions to Chess.com, such as targeted feature recommendations based on skill levels in the user interface and focusing retention efforts on the critical early months when players see the most significant gains.

It also recommended designing advanced learning materials that continue to challenge expert players which would address the diminishing returns observed in this segment.