There was a bug in the seasonal ladder that caused it to create more games than it should have been.

At this point, everyone should only have had 9 games (4 when the ladder started, and one more every 4 days after that.) However, due to the bug, most players have more than that.

I have fixed the bug. This just means that if you're over 9, you won't see any new seasonal ladder games created for a while.

The good news is that this has not impacted the integrity of the season. You just got games earlier than you would have - the opponents are the same as what you would have gotten. The goal is for everyone to have 15 in the end, anyway, so the players who got extra games will just have a break in new games getting created.

Four players got more than 15 games due to this bug. Thankfully, this doesn't really matter, as the ELO rating system doesn't reward nor punish having more games. These players just won't get any more games for the rest of the season, and all of their games will count.

Sorry for the extra games around holiday time. I hope nobody gets booted.

aww.. I only got 14.. thought I was one of the lucky ones ;)

I just noticed my rating is 2101... in 48th!
omg..

Well that explains what was going on. I thought I missed something. Less time to wait!

Hmm, why not let all remaining games start as soon as possible then? Maybe the season can be over sooner? It eludes my why I now have over a month time left to play out 3 of my 4 remaining games, while there will be only a few days for the 4th?

Suddenly I have 6 seasonal ladder games going at once...

"The good news is that this has not impacted the integrity of the season. You just got games earlier than you would have - the opponents are the same as what you would have gotten."

I don't think that is the case. Ranked 3rd, I just got a game against number 95 or something. For the rating system to make sense, every player needs to lose at least one game, and I don't think that will happen. There are too few matchups between the best players for that. Possibly, 5 or more players will end the season with just wins. ELO ratings have never, and will never, provide a meaningful rating for a player with a score of 0% or 100%. It is made to estimate players' ratings between something like 8% and 92% score I think (take the numbers with a bowl of salt - trying to remember from back when I was active in organizing official chess tournaments).

Then again, it's just a game, and we all get nice ratings (everyone are rated above 1600 at this point) to enjoy... :)

Yeon, if a player beats all other players.. *not all games played, but all other players* then the rating system should be optimal, Imho..

|> *Possibly, 5 or more players will end the season with just wins.*

The most wins an undefeated player has right now is 9, and most of them have tough matchups that are ongoing. It's too early to make this prediction. However, if we do end up with multiple undefeated players, I may add a tiebreaker game.

Perrin: The rating number of someone with just wins will always be completely arbitrary since his/her skill could be any number of points better than the other players.

Fizzer: I did say "possibly", several players could end up undefeated, I did not predict that they will. It could also be that none does so (though I have great faith in players like zaeban and Heyheuhei to pull it off). My primary point was that we were not getting the opponents we should.

the first season is an experiment. only 15 games. hard to prove much more than who roughly belongs in the top 20. elo rating systems are inherently more volatile the less games you play. i doubt 15 games (including 4-8 against lesser talent if you are a player who naturally belongs in the top 10, making the games that would best determine one's rating less) is enough to make the eventual rating substantially less volatile.

for RRs that hhh or i make: we invite 20 of the best 50 1v1 players. each player has 19 games against truly tough opponents. the winner tends to lose 2-5 games. people ranked 7th to 12th tend to win only about half their games. if everybody does their best (i tend to use these intense periods against top talent as a way to experiment and try new things to try to improve, since i know i am not good enough to win such a tournament yet, while others sometimes get autobooted, which also skews the results a bit), the results of such a tournament are a better determinant of actual ability. however, even such RRs are not perfect: some guys finish 1st in one RR and 7th in the next against similar players, which proves the 1v1 settings/map's luck/chance factor is a bit too high to reliably predict outcomes.

in any case, i've been trying to keep my game count low and am glad i quit the seasonal ladder 1-4 days before the glitch happened. if i had to play 4-8 more 1v1 games all at once i'd be very unhappy.

Gui, you have an absolute list of the top 50 1v1 players? or are you just using arbitrary standards to determine a makeshift list?

"however, even such RRs are not perfect: some guys finish 1st in one RR and 7th in the next against similar players,"
you can have the same result playing chess.. the human ability to learn and counter what you encounter will give different results game to game..

Imho, the 1v1 strat gives a greater ability over individual skills then 3v3 europe, because of wastelands.. 3v3 europe no cards is so basic.. no wastelands means that the same tactics will work relatively the same, time in and time out.. wastelands mean that you have to alter your tactics on a game to game basis, showing more true skill over regurgitation.

Good point about the 1v1 RRs, Gui. I don't have much experience with them yet (yours is my second, and I played my first when I was very new to this game, so I didn't play that well). Look forward to joining more in the future :)

|> "some guys finish 1st in one RR and 7th in the next against similar players, which proves the 1v1 settings/map's luck/chance factor is a bit too high to reliably predict outcomes"

Great! Otherwise why bother playing the games? This is sport, not math.

perrin, what do you think? by the way, what is imho? i'm a ho? please use english.

az, my point, in chinglish, to challenge the use of msn-english: setting no stable, champ no always champ, other competition champ always champ or almost champ (chinese like commas instead of periods)

Heyheuhei might disagree, if he reads Chinglish...
(imho = in my humble opinion)