Abstract
In this work, we establish a ratio between a range of Elo ratings of chess players and their ability to find the best moves according to the chess engine. To do this, we collect over 7,800 games in the classical modality of ten grandmasters, from when these players had Elo rating about 2200 until they reached about 2700. We compare each move of all these games with the best move suggested by the chess engine in order to determine the expected human performance in chess according to the player rating. We found values for two metrics: average centipawn loss (AVCPL) and standard deviation centipawn loss (STDCPL) with more than 0.98% correlation with the Elo rating. Experimental results with other chess players not used to build the model show that the behavior of these metrics follows the expected behavior. The established model can be used as an auxiliary tool to help detect cheating in the game.
Supported by Sidia Instituto de Ciência e Tecnologia.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
These results were first disclosed in reference [7].
References
Doggers, P.: Niemann has likely cheated in more than 100 online chess games. Accessed 6 Oct 2022
Dorn, S.: Chess grandmaster allegedly caught cheating on toilet during tournament, https://nypost.com/2019/07/13/chess-grandmaster-allegedly-caught-cheating-on-toilet-during-tournament/. Accessed 7 Feb 2022
Engine, O.S.C.: https://stockfishchess.org/. Accessed 21 Sep 2022
International, C.: About online chess cheating. https://www.chess.com/article/view/online-chess-cheating. Accessed 7 Dec 2023
Keener, G.: Magnus carlsen accuses hans niemann of cheating. Accessed 3 Oct 2022
Labelle, F.: Chess problems by computer - Statistics on chess positions, http://wismuth.com/chess/statistics-positions.html. Accessed 17 Jan 2023
Leite, R.V.: How I found perfect correlation between chess player rating and ACPL and STDCPL. https://bit.ly/3CWO2j6, Accessed 03 Oct 2022
Levy, D., Newborn, M.: How Computers Play Chess, pp. 24–39. Springer, Heidelberg (1982). https://doi.org/10.1007/978-3-642-85538-2_2
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Leite, R.V., de Oliveira, A.V.C. (2023). Expected Human Performance Behavior in Chess Using Centipawn Loss Analysis. In: Fang, X. (eds) HCI in Games. HCII 2023. Lecture Notes in Computer Science, vol 14047. Springer, Cham. https://doi.org/10.1007/978-3-031-35979-8_19
Download citation
DOI: https://doi.org/10.1007/978-3-031-35979-8_19
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-35978-1
Online ISBN: 978-3-031-35979-8
eBook Packages: Computer ScienceComputer Science (R0)