E4stat is a blog about chess and statistics. Here you will find forecasts for upcoming and ongoing chess tournaments as well as analysis of chess statistics.
E4stat is run by Matthew Wilson. I'm currently a PhD student in Economics at the University of Oregon. Economic data is often messy; we seldom have the luxury of using controlled laboratory experiments to obtain data. Due to this, all economists are trained quite heavily in statistics. As an amateur chess player, I'm delighted to apply this knowledge to our great game. Here are some projects I want to explore in the future:
-Is 1.e4 really more aggressive than 1.d4 ? More precisely, if all else is equal, will playing 1.d4 instead of 1.e4 significantly increase the probability of a draw?
-Improve the draw rate estimate in tournament forecasts
-Write a program that produces forecasts for Swiss System tournaments. Right now I'm only doing matches and round robins.
-What is the best defense to 1.e4? 1.d4? As explained in my essay "1.e4 - 'Best by Test'?", there is more to be done than simply comparing the percentage scores for each move.
-Move-by-move forecast updating: with enough data, I could use the computer evaluation of each position to update the win/draw/loss probabilities for the game and also the probabilities of each player winning the tournament. You could watch a player's chances plummet immediately after he makes a blunder. Currently I can revise forecasts after each game is finished, but this also could be refined (the estimates for the 2014 Candidates Tournament did not account for whether a player had white or black).
-What would be a reliable way to determine the best challenger to the world championship? Was the old method with the zonals, interzonals, and candidates matches a good one? The best format for the world championship was already discussed in my article "Are the Chess World Champions Just Lucky?" Part 2.
If you have an interesting chess question that can be studied statistically, feel free to let me know.
