Hmmm....dragontamer5788 wrote: ↑Mon Jul 22, 2019 3:45 pmUnder the original Bradley Terry system that Elo is based on, you assume that the "losing" player has one half-win out of the whole bunch.
So if you played 50 games and the computer won 50 times out of 50, you calculate a score based off of 50/50.5. That is equivalent to 100-to-1, or roughly +800 Elo or so. If it were instead 5000 games and the computer won 5000 games out of 5000, then the score is off of 5000 / 5000.5, or 10,000-to-1 (roughly 1600 Elo).
Under a system I explored (unpublished), where you create acyclic graphs between players and assign the hypothetical computer "infinite" more score than Magnus Carlsen. You can only create a Bradley-Terry (or Elo) score between players who have a loss or draw against each other. You compare players with a Topological sort. All cycles are resolved with Bradley Terry (or Elo), while acyclic / trees are sorted by a Topological sort of some kind.
The fundamental assumption of Bradley Terry (or Elo) is that the win/loss graph is a statistic, and therefore has a chance to be wrong. If the system never observes a loss, then you have a "division by zero" situation. The actual division by zero can be avoided by topological sorts + graph theory. But the original papers talking about the system has the +0.5 methodology to the loser, which seems to work well in practice.
Not sure I understand what you are saying. Say, I have a 10,000 game match between SF (search depth 10 ) and a random player. SF won all games. If I set the random player to ELO = 0 , are you saying that I can calculate SF ELO by assuming it won 10,000 games out of 10,000.5 games?
If this is what you are saying, then what would be the error margins on such a calculation? i.e. How would the error bands be calculated? Using some form of poison distribution?
Regards,
Zenmastur