Interesting. I could see my rating going down a bit, but what I've noticed is that winning 6+ player games is a huge boon to ratings. In games of 5 players and under, if you're good you can certainly win a bunch, but to me, in games of 7, 8, 9 players, if you are good at those, you can win a really disproportionate number of games. When some people on this thread talk about not wanting to play games with low ranked players, I feel the complete opposite way. If I'm in a 10 player game and there are two 875 rated players, I'm thinking my odds of winning are already better than 10%, and that's just considering how unlikely they are to win. When you also favor in how likely those players are to do something wacky and can sort of predict that behavior and use it to an advantage, you can win quite a bit. Also, I've played many, many games with 6+ players in other places, so there's that.

I think the big difference SW that as the number of players increase, the effect of luck decreases for Wargear Warfare.  For example, when playing 1 v 1, bad luck on the first round can easily lose the game for you.  So, I agree that you can increase your H rating by playing more games with multiple players. 

I have played over 600 1 v 1 games for Wargear Warfare.  I think that for 1 v 1 is it difficult to get over 65% H rating even for very good players because luck is such a strong factor.  I may be wrong, but I think most of the highest rated players for this board play mostly multi-player games.

soft wizard wrote: Interesting. I could see my rating going down a bit, but what I've noticed is that winning 6+ player games is a huge boon to ratings. In games of 5 players and under, if you're good you can certainly win a bunch, but to me, in games of 7, 8, 9 players, if you are good at those, you can win a really disproportionate number of games. When some people on this thread talk about not wanting to play games with low ranked players, I feel the complete opposite way. If I'm in a 10 player game and there are two 875 rated players, I'm thinking my odds of winning are already better than 10%, and that's just considering how unlikely they are to win. When you also favor in how likely those players are to do something wacky and can sort of predict that behavior and use it to an advantage, you can win quite a bit. Also, I've played many, many games with 6+ players in other places, so there's that.

I have to second what soft wizard is saying.  After about 30 or so games played I came to the same conclusion, it is better for me to play against many players because I feel I am better than average, so I should win more than 1/8 of my 8 player games, more than 1/9 of my 9 player games, etc.  I was able to increase my ratings to over 2,000 by about my 75th game in this fashion.  I actually think my ratings are artificially inflated due to this.

As for the lower rated players doing something wacky: yes, but sometimes the wacky thing is to attack me when they have many better options.  :)

-Trabi

Right, but there is a translation. If you are a little over 1/9 on 9-player and way over 1/5 on 5-player games, you should play 5-player games. Winning 50% of your 5-player games should get you similar results as winning 33% of your 9-player games.

That said, I think you're right about how it plays out in reality on Wargear Warfare and similar maps. (Higher H-Ratings for a good player in games with more players.)

Nice start to both of your stats, by the way!

I don't think a hard cap is the solution to the perceived problem or lack of fairness of a 2000 player losing to a 500 player.

Generally, I think after you play enough games, depending on who plays, how many play, what the luck is, how attentive to that game each player is, each players experience level etc etc then things even out.  However, that said I think it is obvious from the scores on some games some people are just better than everyone else at that game at a specific time. 

Regardless of how good you are your score prior to the start  of the game is only an indicator of how you will perform against the other players.  Therefore it is inevitable that a player with a 2500 score will lose to a player with 500 if they play enough games.  Sorta like the old saying, "There are only 2 types of pilots-those who have landed with their wheels up and those who will."

Therefore I think it is better to go with the current system than to try to find a numerical/mathematical solution that will not be able to factor in all the relevant reasons for why you won or lost.

I've heard a few things in this thread which all make sence. First the problem of losing against an opponent which has an extreme low rating. A rating system without a logarytmic or exponential scale will only be unfair if someone has a rating which does not even close reflect someone's skills. In theory everybody would lose all of it's rating if he lost against an rating below 20. While this wil never happen IRL, it shows it becomes unbalanced when differences get too high. My personal opinion is a logarythmic or exponential scale is better, because it lowers volatility, but this does not mean the current system is unfair.

But, not losing to an extremly low rated player should also be part of the strategie. Often in the game you can choose who to eliminate. In these scenario I will most likely choose the lowest rated player, because this decreased my maximum loss. So while high rated players have a huge drawback of losing big to low rated players this drawback will be minimalized (if not totally mitigated) by the fact people tend to let high rated players win more often then lower rated players because they will lose less themselves.

Last but not least I think it's true good players benefit from 10+ games to boost their H-rating. this is because in 10+ games luck is less important and strategie is rewarded instead of  tactics. And since H-rating is the decisive factor for your equilibrium rating, high rated players are eventually forced to go up in games with multiple players because they already have reached their equilibrium rating at the lower-men games. 

Your true H-rating is a weighted average of your 2P-H-rating, 3P-H-rating ...16P-H-rating. But your H-rating is not only affected by the number of opponents you play, but also the type of board and the gameplay settings your favour most.