ELO & Poker

Well, here is my scoring applied to a few random shuffles of your field. Every first line is the scores before the match, every other the points granted for the placement:

Is that what you were asking for?

Just to make that explicit: I would strongly recommend against the ESL style of simply seeding at 1000 points. That's a terrible idea for a number of reasons, and I'm really surprised they haven't fixed it yet.

If you don't like my proposal, consider the original Elo idea of having a "startup" phase where nobody can lose points against the player, and afterwards gets assigned a score depending on his performance in that period. That's what Chess uses, but it's obviously quite a bit harder to implement.

By taking the given scores from this table you can calculate the 1vs1s an accumulate the final given scores for each round.

I'm not sure I understand what you mean, sorry. Are you telling me your new (old?) system is also based on breaking everything down into 1vs1s?

I find Maytsh's model actually quite interesting.

Any further thoughts on this from the PokerTH crew?

LuisCypher wrote: During the next days we are busy with the cup and the reset, but we will be on this task asap. And maybe a living example invites also the others into the discussion. Thankyou in advance!

Just checking - any news on this?

Sorry, no news at the moment. We have other priorities: performance & system stability. We will be back on this asap.

Just to let know, that to me you get the ranking on PokerTH more or less right the first time.


With 30 games ranking is not as much like playing a low stakes game that is meaningless to you as far as your bankroll:

1. Each individual game has a significant impact on your ranking (is not like a drop in the ocean).
2. Lots of fun to see what your 30th result is and try to improve on it in the next game (with your 100th result is not going to matter as much how that just one result changes).
3. If play a high-ranked player in a game and give a bad result to, then you "got him" a little bit more (which makes more fun from the possibility).
4. With only 30 games to play giving full attention to them to get a good ranking, can play games as a side competition to reading a book or something more, or just try out new ideas for how to play poker, and so on, all knowing that just 30 games w/ full attention to get your normal ranking back (which is what most of games have played on PokerTH were, incidentally, i.e., side competitions, to something else).
5. Also as players may constantly be changing their poker strategies and playing with more or less discipline, last 30 games may be a more accurate reflection of who is the best player on PokerTH (at the moment).

With 30-100 games ranking, there is incentive to create/delete accounts, and try to just get 30 good games (i.e., to get the best possible ranking), and also, as want a combination of more short-term luck to go along w/ skill.

I also not like the idea of an ELO-based ranking system (for poker).
1. Is not a 100% skill game.
2. Even if were a 100% skill game the skill is not linear (with players who tend to call too much tending to outplay players who tend to raise too much tending to outplay players who tend to fold too much tending to outplay players who tend to call too much)...
3.

Well, are other reasons as well, but have to cut it short, as have to go somewhere, i.e., something else to do.

I completely agree with all your points, gfgh