# Elo Rating System in Poker?

Red Chipper Posts: 5 ✭✭
https://en.wikipedia.org/wiki/Elo_rating_system

As both a chess player and poker player, I have always wondered what it would be like to implement the Elo rating system (used in chess) in poker. One's rating is determined relative to everyone else in the player pool, so a universal Elo rating (one which you could take from one poker room to the next) would not work. That is, one poker room would have to put on a particular event or series of events in which every player would have their Elo rating tracked. I'm not quite sure how the ratings would increase or decrease based on performance since poker isn't exactly a "win or lose" game like chess is. Assuming that it could be modeled correctly, I think a rating system such as Elo could have some pretty interesting applications in poker. Given an accurate rating:

- Players could be placed at tables where every opponent is at relatively the same skill level. This could attract newer players who aren't yet confident in taking on the pros. Although, this hurts the game for the bum-hunters.

- Players could estimate their likelihood of "winning" based upon the ratings of other opponents at the table. Example: "The difference in the ratings between two players serves as a predictor of the outcome of a match. Two players with equal ratings who play against each other are expected to score an equal number of wins. A player whose rating is 100 points greater than their opponent's is expected to score 64%; if the difference is 200 points, then the expected score for the stronger player is 76%."

There is a glaring issue with modeling one's Elo rating though: variance. The best player in the world could run bad for the first 5 rated events and have his/her rating tank. This is a problem that can't be dealt with easily due to the nature of the game. A solution to this would be to accept the inconsistency, and trust that players will be rated correctly in the long term.

A rated poker system would also introduce a new angle to shoot: the smurf angle. If there are enough rated events being put on by the poker room, it might make sense in the long run to tank your own rating to be placed into softer games later. I'm not sure if this would be profitable--there's a lot of math involved that I can't be asked to do.

I like the idea of having a rating system in poker, given that it's possible to reasonably implement such a thing. What do you guys think about a system like this? Would you play a rated poker series?
Tagged:

• Las VegasPosts: 5,045 -
Hmmm. Pretty difficult to develop an Elo system for a multi-player game. Plus even heads-up games like Hearthstone shy away from a traditional Elo in the algorithm for the MMR. Chess is uniquely suited to it thanks to it being basically 100% skill, three well-defined outcomes (1-0-1/2), exclusively heads up, and with a relatively rapid intrinsic convergence so you don't have problems tuning the K-parameter.
Moderation In Moderation
• Red Chipper Posts: 5 ✭✭
Haha, bad idea--got it. Thanks for the input!
• Red Chipper Posts: 2,495 ✭✭✭✭✭
edited April 2020
jeffnc wrote: »
Chess is also suited to it because it's rarely played for money :) In poker, the point of the game is money. Assuming for sake of argument that such a rating could be achieved in theory, it would be a bad idea for the majority of the player pool. Not knowing how good (bad) you are is a large part of what fuels the game.

And in this setting, who would play - and potentially lose - regularly against players clearly / knowingly better than them ?
(I like it, but I'm a bizarre and rare animal 😆)
• Red Chipper Posts: 5,007 ✭✭✭✭✭
You see it occasionally - guys with money want to play with the best. Just to say they did it - or maybe beat them (and who's to say they got lucky? maybe they'd like to think they were actually better.) But generally, I'd agree with you.