Bias in the ELO-System of Online Chess

There are millions of chess players around the world that play chess on websites like chess.com.1 According to chess.com,1 there are 600 Million chess-players worldwide, and more than 20 Million members on chess.com1 that play up to 1 Million games per day, and there are 360.000 tournament players and 1594 grandmasters of which only 2.2% are female, while it is free to play for everyone.1 Today, bias in research is ubiquitous,2 also in chess research and most likely also in ELO rating, which should be further analyzed in some more details. The strength of a player in the game of chess and additional zero-sum games is usually estimated and assessed in ELO, a rational chess skill rating system that was initially developed by Arpad Elo, a Hungarian physicist who lived, worked and taught in the USA. The ELO-ratingsystem statistically derives numerical outcomes in ELO of the games a player plays against opponents of different strength: it increases the ELO score value of the winner as much as it decreases the ELO score of the loser, roughly speaking, deviations of the rule equal much out as the sum of both directions. A high-rated player can lose more points against a low-rated player, who can win more points against a high-rated player, and vice versa, a high-rated player can win fewer points against a low-rated player, who will lose fewer points against a high-rated one, leading to an ELO conservation effect after deviations and correction factors equal much out, slightly comparable to the conservation of energy that is fixed in the system, like the conservation of energy in the system. This adjustment upwards or downwards goes back to Elo3,4 who suggested it, and USCF, the United States Chess Federation, implemented his suggestion in 1960, subsequently it became an international standard, and USCF, the United States Chess Federation, has implemented his suggestions in 1960. The basis of ELO’s formula3 for the