All chess rating systems including the Elo rating system are based on a procedure called "scoring" which assigns a score of 1 for a win, 0 for a loss, and 1/2 for a draw. This procedure reduces the trinomial nature of chess game outcomes to a binomial variable and thereby greatly simplifies the mathematics of comparing the performance of chess players. Of course there is no free lunch in math and so this simplification is achieved at a cost because scoring causes some chess game outcome information to be lost and no amount of mathematical wizardry downstream can recover this information. The extensive effort by many to improve the Elo rating system with mathematical genius is for naught. The only way to improve chess performance measurement is to remain true to the trinomial nature of chess game outcomes which has two degrees of freedom. The way to do that is to use a two-dimensional measure of chess performance. I wrote a paper proposing such a method and posted it online for comments. Here is the link to the download page. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2488369 Your comments appreciated.