3 ms·
By the way, I would be interested to see a graph of the average rating of the players by year, as well as what the other graphs look like when controlled for ra
by dfan 12y ago
By the way, I would be interested to see a graph of the average rating of the players by year, as well as what the other graphs look like when controlled for rating. I wouldn't be surprised if the explosion of the data set in recent times is partially due to many more games between lesser players being recorded, and this could certainly have an impact on the stats.
- rhiever 12y agoHere's a distribution of Elo scores in the data set: http://www.randalolson.com/wp-content/uploads/chess-elo-score-dist-plot.png http://www.randalolson.com/wp-content/uploads/chess-elo-scor... This data set is from chess tournaments, so it's predominantly games with skilled players.
- dfan 12y agoWhat I was wondering specifically was how the average rating varies by year. My hypothesis was that it has gone down over time as it has become easier for games between less skilled players to get into databases.
- rhiever 12y agoHere's a quickly made chart to show that: http://i.imgur.com/lHbVzXM.png http://i.imgur.com/lHbVzXM.png (I don't have enough games w/ Elos pre-1960 to show a reliable mean.)
- dfan 12y agoThanks. FIDE didn't implement ratings until 1970 anyway. (I'm not sure what the ratings you have for the 1960s actually are - perhaps USCF ratings.)