Average centipawn loss vs Avereage Relative Centipawn loss in Kramnik vs Navara
In a brave public act, David Navara explains in details what he went through after the baseless allegations of Kramnik. Navara describes the emotional torture by Kramnik’s allegations, the silence of FIDE, their posterior inadequate reaction and total disregard on basic FIDE ethical and moral values.
The basis of the story is a post by Kramnik on Twitter/X that is connecting Navara’s name with “stats” on “Cheating Tuesdays”. As Kramnik admitted, he is not making accusations but allegations. These allegations are based on data that Kramnik compiled himself and used an undisclosed method to analyze and tweet about.
In a post titled Checking the cheating allegations made by Kramnik on David Navara on Reddit, the user bhajji22 presents his analysis on Kramnik’s data and specifically the data in the tweet. Bhajji22 says, “I downloaded Navara’s and 8 other players’ all games they played in Titled Tuesdays in 2024 from letsdotheprocedure.com. Then I calculated their average centipawn losses and compared the results. […] Firstly, even the most basic thing – the number of moves Navara and others made when time on their clock < 10 seconds was wrong in Kramnik’s post. For instance, on his post , Kramnik said that 748 moves were played by Navara but I found nearly 1500 moves. Similarly, other values are also wrong.” Bhajji22 continues, “Now the cheating part – It can be seen that average centipawn loss for Navara is infact lower than one might expect but statistics is not that simple. Let us understand through an example. Let us say that a position is +5. Then a move that reduces the evaluation to +4 will not have the same effect as reducing the position from +1 to 0. As we all know, in blitz, the top players have the ability to simplify the game to a winning endgame which is necessary in time pressure and hence the +5 to +4 is a common thing. So instead of the Average Centipawn Loss, a better metric would be the Average Relative CPL. This would scale down the large values. So when this was calculated for each player, Navara’s value came out to be quite normal. Not only this, I also plotted a graph for the critical moves when the evaluation is between +1 and -1. And infact, Navara was the weakest player in that scenario.”
User fuettli comments, “I already did something like this and have done some more in the mean time with a little more info. I limited the data to the same time frame Vladimir Kramnik used (you used the entire year 2024) but of course even if you try to go as close to what VBK did, the results are always way different. According to some tweets he used positions with +2 to -2 and a loss of at least 100 centipawns.
But what about blunders from +8 to -1? Aren’t those also real blunders?”
Bhajji22, unlike Kramnik, published the full method and code behind his findings. While the period of Bhajji22 calculations is off, it is public for the mathematicians, statisticians, and chess experts to analyze the centipawn loss vs the suggested Relative centipawn loss. Check out the full post on Full post on Reddit here
Timeline of events
23.05.2025 Navara publishes an open letter about Kramnik’s actions
23.05.2025 Reactions to Navara’s open letter – Aronian, Polgar, etc
23.05.2025 Navara and Kramnik to meet at FIDE World Rapid and Blitz
24.05.2025 David Navara, “I firmly believe that Mr Dvorkovich should resign”
24.05.2025 Kramnik, “I am deeply disappointed Navara participates in unjust PR campaign against me”
24.05.2025 Kramnik rage-blocks Chessdom over a comment on David Navara’s case
24.05.2025 Live: David Navara at French Top 16
25.05.2025 Kramnik admits to allegations, FIDE has grounds to act
25.05.2025 Open call to action for FIDE
26.05.2025 David Navara now has personal chess calendar
26.05.2025 Czech Chess Federation: We stand behind chess legend GM David Navara
28.05.2025 MrDodgy : “Kramnik bullies Navara, will he be satisfied if he ends in hospital or worse?”
29.05.2025 Emil Sutovsky : “I find Kramnik’s latest address to David Navara inhuman”
12.06.2025 David Navara: “When I heard Kramnik’s reaction, I slept really poorly”
26.06.2025 Official Statement of Hetman GKS Katowice Chess Club + Keti Tsatsalashvili of support of David Navara
28.06.2025 Levon Aronian addresses Kramnik’s lawsuit threats towards David Navara
29.06.2025 FIDE addresses the Kramnik vs Navara case
29.06.2025 The entire Nový Bor Chess Club stands firmly behind David Navara
29.06.2025 Prague International Chess Festival thanks cFIDE for taking a clear stance on Kramnik vs Navara
04.07.2025 David Navara in an interview with Greg Mustreader
08.07.2025 Average centipawn loss vs Avereage Relative Centipawn loss in Kramnik vs Navara
Context (28.05)
Kramnik is known to have published numerous articles and videos regarding cheating in chess. On multiple occasions he uses a small sample size. As a result, his conclusions may be perceived as accusations (this is the case in a multitude of well documented videos, articles, and opinions by experts here, here, here, here are just a few examples), allegations, or suspicions about cheating in chess, and even as insults to players’ professionalism. Despite how each individual perceives his words, on Chessdom, starting 25 May, we have adopted a term that Vladimir Kramnik used himself – allegations.
In accordance with established journalistic practices, we work with information and quotations from sources, including Kramnik himself. Kramnik has never contacted Chessdom via its official communication channel posted in the header of the website. Our attempt to contact Kramnik was futile and met with a swift block. The latest threats by Kramnik on social media leave little room for any good-faith discussions (further details to follow)
Regardless, we stand by accuracy, fairness, and independence of journalism and will continue reporting on the Kramnik – Navara case and its tremendous social importance.