Chess960 Stats
Look up your percentile
This pulls your most recent live Chess960 rating.
- Live 960 percentile
- Most recent rating date
How we sample the population
There is no chess.com endpoint listing every Chess960 player, so we build the population ourselves. We start with six seed players spread across the rating ladder (roughly 500, 1000, 1500, 2000, 2500, and 3000) and do a breadth-first crawl: collect each seed's 2026 live-960 opponents, then their opponents, and so on. This gives us a rating-balanced sample of the active 2026 community without over-sampling any one neighbourhood.
How large is the community?
As we crawl, new discoveries per visit slow down because we keep bumping into players we already know. Plotting total observed players against API pulls produces a saturation curve; we fit it with a Hill function to estimate the full reachable population.
How the percentile is calculated
We use the mid-rank convention: half the players at exactly your rating count as “below” you. So if 80% of players are below your rating and 4% are exactly at it, your percentile is 82%. Players at rating 100 are excluded from the comparison population for the reason noted above.
Note: this dataset is still growing — percentiles shift as the crawl observes more players. Treat results as a snapshot, not a final number.