Ratings Distribution across the Chess.com community
Sort:
The data are available for chess.com to report the full distribution of rating for each game time, including mean, variance, skew, kurtosis, and so on. It would be interesting to know where one stands in that distribution. Why not provide that in stats?
Sep 24, 2022
0
#2
DeanFurb wrote:
The data are available for chess.com to report the full distribution of rating for each game time, including mean, variance, skew, kurtosis, and so on. It would be interesting to know where one stands in that distribution. Why not provide that in stats?
You can always suggest it. Help > Make a Suggestions
Dec 28, 2022
0
#3
See a similar thread at https://www.chess.com/forum/view/livechess/rating-distribution-graph
There are leaderboards, including the one at https://www.chess.com/leaderboard/live/rapid
which displays the distribution at the percentile (define as: the percentage of players whose rating is worse)
Dec 28, 2022
0
#4
I was wondering which rating at rapid I should improve to, in order to be in the top 10% players.
So, looking at the leaderboard, I copied the values (#players vs binned ratings) and calculated the cumulative sums and percentiles:
| rating | #players | cumsum players | percentile |
| 100 | 371393 | 371393 | 2.43% |
| 200 | 713045 | 1084438 | 7.09% |
| 300 | 1062886 | 2147324 | 14.03% |
| 400 | 1395763 | 3543087 | 23.15% |
| 500 | 1585391 | 5128478 | 33.51% |
| 600 | 1661301 | 6789779 | 44.37% |
| 700 | 1612318 | 8402097 | 54.90% |
| 800 | 1503122 | 9905219 | 64.72% |
| 900 | 1265075 | 11170294 | 72.99% |
| 1000 | 1061899 | 12232193 | 79.93% |
| 1100 | 847289 | 13079482 | 85.46% |
| 1200 | 681612 | 13761094 | 89.92% |
| 1300 | 489680 | 14250774 | 93.12% |
| 1400 | 348934 | 14599708 | 95.40% |
| 1500 | 242416 | 14842124 | 96.98% |
| 1600 | 169290 | 15011414 | 98.09% |
| 1700 | 112122 | 15123536 | 98.82% |
| 1800 | 73300 | 15196836 | 99.30% |
| 1900 | 45790 | 15242626 | 99.60% |
| 2000 | 29797 | 15272423 | 99.79% |
| 2100 | 16357 | 15288780 | 99.90% |
| 2200 | 8582 | 15297362 | 99.96% |
| 2300 | 4130 | 15301492 | 99.98% |
| 2400 | 1677 | 15303169 | 99.99% |
| 2500 | 576 | 15303745 | 100.00% |
| 2600 | 168 | 15303913 | 100.00% |
| 2700 | 61 | 15303974 | 100.00% |
| 2800 | 25 | 15303999 | 100.00% |
| 2900 | 5 | 15304004 | 100.00% |
| 3000 | - | 15304004 | 100.00% |
| 3100 | - | 15304004 | 100.00% |
| 3200 | - | 15304004 | 100.00% |
| 3300 | - | 15304004 | 100.00% |
| 3400 | 1 | 15304005 | 100.00% |
| SUM | 15304005 |
Therefore, the rating for top 10% (alternatively the percentile for which 90% of players are worse) is around 1200.
Notes: the values provided are binned, so I don't know if it's upper bound, lower bound or centered. Besides, access to full data (rather than the histogram provided) would help determining exact rating.
By the way i want to ask if rating actually means anything?
On my 1400s, in like 7-8 of 10 games i meet Q abusers (guys that go Qh4/h5 or fried liver or reversed fried liver involving attempt of early Q attack). If they lose q they resign, they cannot play any other piece and if they do they just go straight up exchange eveything without any tactics asap and pass another Qs.
They all play for the hope of oppenent's blunder and do exactly same blunders on their own
Being 1400, they are naturally like 1100 at max!!
They do not even try to play some various easy gothamchess stuff just rush Q or exchange everything!
I'm sick of this!!
Why there are no players that try to play freaking chess and enjoy game not that stupid Q abuse?
From which rating actual game of chess starts? I want to play interesting combinations and mating nets in the midgame and all but matchmaking system is like in dota where i'm always magnetically matched with a type of players that i hate most!!
Mar 12, 2025
0
#8
MagnutClarsen-X wrote:
Why does it say "cumsum" ?
Cumulative sum
Its just the total amount of people in a section
The table also shows just how different the chess.com players are from the OTB tournament world. In the USCF, the average rating is in the 1500s. A rating of 1200 would be about 1.5 standard deviations below the mean, around the 10th percentile.