Standings after 2 rounds:
Today I had the opportunity to continue working on the model's draw rate assumptions. The tentative result that d4 and c4 were more drawish than e4 remained, but there was a big surprise on another front. Last time I reported that the draw rate was not increasing over time, leading to the conclusion that chess was not being "played out." But strangely enough, when the year that a game was played in is multiplied by a player's rating, it is significant in predicting if the game is drawn! So I don't have a clear-cut answer on whether draws are becoming more or less likely over time - it seems to also interact with the players' ratings in ways we don't yet understand.
No comments:
Post a Comment