I think it splits the players into two groups, "players who have just lost a game" and "other players", and whenever a member of one group plays a member of the other, the player who just lost a game goes first.Yeah. More generally, when the game starts, everyone who has just lost a game goes before everyone who has not. This means that if you just won, you can't log out and log back in to try to go first.
This explains why i keep getting stuck with 3rd position in 3p games! I think i have a new algorithm for deciding whether or not to play a 2er or 3erHa! Play 2p til I lose, Play 3p til I win? Love it.
Captain_Frisk takes out his losses on level 20+s.
If he just lost, he won't play anyone rated lower than level 20.
I automatch against any reg. But I generally only accept challenges from 25+. Of course, I'm not 40+, though I was a couple of days last week. The only way I do anything to "game" the system at all really though is to almost always make sure I "win-quit".
I actually haven't looked at the math that close - is there a way to see what % you need to score to maintain a certain level difference? Obviously +/-0 is 50%, but is +1 like 51%, 60%, 90%, where? and +2, +3, so on. Is this out there somewhere?
This is exactly what I was asking about. And I meant to say level difference in the mean, not the levels as posted on the leaderboard. But you did what I was looking for anyway, but these numbers don't make sense with each other quite - easiest way to see that is by looking at difference ranges 13-16, but if you just look at the differences between consecutive numbers, you get a sequence that doesn't make a lot of sense:I automatch against any reg. But I generally only accept challenges from 25+. Of course, I'm not 40+, though I was a couple of days last week. The only way I do anything to "game" the system at all really though is to almost always make sure I "win-quit".
I actually haven't looked at the math that close - is there a way to see what % you need to score to maintain a certain level difference? Obviously +/-0 is 50%, but is +1 like 51%, 60%, 90%, where? and +2, +3, so on. Is this out there somewhere?
Well, it's not just about how many games you win, it's who you beat. Just like in chess or other games with a global ranking system.
From TrueSkill's point of view, it's more meaningful to compare the players' mean skills, not their levels. And these probabilities are interesting enough that I'll make a table of them.
diff win prob.
0 50.0%
1 54.2%
2 57.6%
3 60.0%
4 63.5%
5 68.6%
6 71.5%
7 75.5%
8 78.7%
9 82.2%
10 82.6%
11 87.7%
12 90.8%
13 93.9%
14 97.6%
15 97.3%
16 98.6%
>=17 100%-ish
This is exactly what I was asking about. And I meant to say level difference in the mean, not the levels as posted on the leaderboard. But you did what I was looking for anyway, but these numbers don't make sense with each other quite - easiest way to see that is by looking at difference ranges 13-16, but if you just look at the differences between consecutive numbers, you get a sequence that doesn't make a lot of sense:I automatch against any reg. But I generally only accept challenges from 25+. Of course, I'm not 40+, though I was a couple of days last week. The only way I do anything to "game" the system at all really though is to almost always make sure I "win-quit".
I actually haven't looked at the math that close - is there a way to see what % you need to score to maintain a certain level difference? Obviously +/-0 is 50%, but is +1 like 51%, 60%, 90%, where? and +2, +3, so on. Is this out there somewhere?
Well, it's not just about how many games you win, it's who you beat. Just like in chess or other games with a global ranking system.
1-0 4.2
2-1 3.4
3-2 2.4
4-3 3.5
5-4 5.1
6-5 2.9
7-6 4.0
8-7 3.2
9-8 3.5
10-9 0.4
11-10 5.1
12-11 3.1
13-12 3.1
14-13 3.7
15-14 -0.3
16-15 1.3
17-16 1.4ish
What did you do here? The player variance matters a lot in terms of pushing probabilities toward 50% regardless of mean difference. I don't see how you can reasonably ignore variance and still get output from trueskill.I think if the database is high enough, the variance should just integrate out and you get the avarage win probability given the difference of the mean (integrated over the "variance distribution of isotropic"). That is some kind of output, question is how much this matters for one individual player.
Don't know what exactly you mean that doesn't make sense, but if is the oszilating and even negative values in the table, I think that's mostly noise.Yeah, but if the noise is so big, the data is next to worthless.
I'd like to see it switched to just a random player goes first. It discourages re-matches since the winner knows they're definitely going second vs a... dunno 1/4 to 1/3 chance of going first if they play somebody else. And other weird matching effects.From what I understand, dougz goes to great lengths to make sure Isotropic plays by the Dominion rules as written. And them's the rules: player who just won goes last.
I'd like to see it switched to just a random player goes first. It discourages re-matches since the winner knows they're definitely going second vs a... dunno 1/4 to 1/3 chance of going first if they play somebody else. And other weird matching effects.
I'd like to see it switched to just a random player goes first. It discourages re-matches since the winner knows they're definitely going second vs a... dunno 1/4 to 1/3 chance of going first if they play somebody else. And other weird matching effects.
From TrueSkill's point of view, it's more meaningful to compare the players' mean skills, not their levels. And these probabilities are interesting enough that I'll make a table of them.
diff win prob.
0 50.0%
1 54.2%
2 57.6%
3 60.0%
4 63.5%
5 68.6%
6 71.5%
7 75.5%
8 78.7%
9 82.2%
10 82.6%
11 87.7%
12 90.8%
13 93.9%
14 97.6%
15 97.3%
16 98.6%
>=17 100%-ish
What did you do here? The player variance matters a lot in terms of pushing probabilities toward 50% regardless of mean difference. I don't see how you can reasonably ignore variance and still get output from trueskill.
Run trueskill for each game g between p1 and p2
let d = abs(diff(mean(p1), mean(p2))
associate win or loss with d
So the table you posted is this?
Run trueskill for each game g between p1 and p2
let d = abs(diff(mean(p1), mean(p2))
associate win or loss with d
Then print win/total for each d?
I agree that makes sense, it basically the average win probability given level difference over all games on isotropic.
But the official rules state "When playing multiple games, the starting player is the player to the left of the winner of the last game". If you're having a rematch, you're playing multiple games. Thus, the winner of the previous game should go last.
If two players match up, and they both either won or lost their previous games, how is turn order decided?
Is that then a coin flip?