lichess.org
Donate

A Data-Driven Approach to Chess Improvement [Part 2 - Chessable Effectiveness]

Chess
By combining science-backed learning techniques, most notably SRS (spaced repetition), with interactive tools, Chessable has gained widespread popularity and is used to help players retain and apply chess knowledge effectively. But how effective is it in practice? To answer this, we examine data from users' experiences, rating improvements, and other metrics that provide insight into the platform’s impact on chess performance.

Link to survey where this data was gathered: Chess Studying Survey

Previous Statistical Article About Training Methods: A Data-Driven Approach to Chess Improvement [Part 1 - SRS & Woodpecker-Method]


Chessable’s effectiveness in tactics

An analysis of the survey data reveals that players who use Chessable to train tactics report a self-assessed tactical ability that is slightly higher than that of non-users.

While these differences highlight a positive association between Chessable usage and tactical confidence, it is important to consider the role of increased training frequency. Chessable users train tactics 27% more frequently on average than their non-chessable user counterparts, which likely contributes to their higher self-assessed tactical ability.

However, one might argue one of the main reasons why chessable users study more is that chessable contains a gamified environment where you're awarded for studying more frequently, which could be considered a good thing about chessable as a website.

Chessable’s effectiveness in openings

Chessable is often more correlated with opening theory than tactics or strategy. This begs the question, how effective is chessable for learning openings?

The data shows that Chessable users study openings 33% more frequently and review their variations 49% more often than players who do not use the platform. This greater focus on opening study appears to correlate with a stronger sense of confidence, as Chessable users rate their opening skill relative to their rating strength as 22% higher than non-users. Additionally, Chessable users tend to maintain slightly narrower repertoires, indicating a more focused and specialized approach to opening preparation compared to non-users.

These results suggest that Chessable's features not only encourage more consistent opening study but also help players develop a deeper understanding of their chosen repertoires. The combination of frequent study, systematic review, and a tailored approach to repertoire building likely contributes to the enhanced confidence Chessable users report in their opening play.

Chessable's Impact on General Chess Improvement

A comparison of Lichess blitz ratings over one year among 450 players provides insight into how Chessable users progress relative to non-users.

On average, non-Chessable users gained approximately 11% more rating points (100 Elo) over the year than Chessable users, who gained 90 Elo. However, it’s important to consider the context: Chessable users had an average rating of 1893, compared to 1661 for non-users. Since rating improvements become increasingly challenging at higher levels, this discrepancy suggests that Chessable users are performing well despite the diminished returns typically associated with higher ratings.


My Opinion and personal experience

I’ve found chessable to be very useful. The only issue I’ve found with the opening repertoires is that they are very easy to prepare against because anyone can see what moves are recommended by either owning the course or using the repertoire tree. There are a few things I would like to improve, and who knows, I might create an alternative website in the future with interactive exercises just like it.

Your Thoughts

What are your thoughts on this? What metrics would you have suggested and should I have brought more into consideration? Feel free to leave a comment down below in the discussion forum!