I can't write anything on the TrueSkill blog. What I tried writing:

SebCorps, I knew you were on my payroll for a reason. A fat red envelope is on the way. 新年快樂恭喜發財! Happy Chinese New Year! (Too bad you assume one's ability to memorize patterns and learn how to pick/play on the 1v1 ladder is the sole determining factor as to who is best, as if team games, FFAs and ability to play/adjust to other/new maps/settings is irrelevant. There are many guys who excel at the 1v1 ladder settings but are only decent or above average players on other maps/settings or formats.)

Fizzer, why not create your own ranking system, something that is a tweaked basket of ranking systems that best suits WL? Or, maybe just average the two systems mentioned? An average (or basket of ranking systems) would be best I think: get the best of both, reduce the problems either might have.

Also:

1 zaeban 2421.536 41 6
6 Heyheuhei 2264.112 62 23
11 chas 2203.465 36 15
13 20AquaHolic 2158.945 61 31
15 13CHRIS37 2143.782 44 16
18 TheEmperorCornInMyTight 2079.722 61 29
24 Xyphistor 2018.273 71 39
26 REGLMentysh 2005.368 28 17

TrueSkill awards consistency over a larger number of games. This is a good thing.

Hah, Gui, I could understand your Chinese! Can't read it, but obviously I know what it says. I saw the post about the true skill method, and as a new player, can't really comment. It seems a bit more fair, as well as rewarding consistency. Also, if it drops Chaos' rankings, maybe it will stop him from bragging as much. He's becoming insufferable!

yeah, common expressions posted everywhere in the chinese-speaking world around this time of the year.

just ask chaos to surrender in his 2v2 game against unknown and me so the top spot returns to its rightful owner!

As is typical with my very very limited Mandarin, I can say what you wrote, but couldn't possibly hope to put it in Pinyin. I will, however, try. XinNienKuaiLe, GongXiFaCai. And as for asking Chaos to surrender, he doesn't really listen to what I have to say. Ever. So rude. As a side note, referring to him as "Chaos" is really really strange.

I think that Trueskill is great if we could ascertain that nobody's skill is changing. Since that's almost certainly not true (people get better as time goes on), then what we might be measuring, in addition to consistency, is the person's ability to learn the format. That's even assuming that they join at the same skill level. Two recent additions are zibik and lobstrosity. They both were very skilled before becoming members, so they will have unusually high rankings compared to someone who buys membership the first day on the site, and tried out the ladder.

If most of us had to earn our ladder rankings while learning how to play the game at all, our rankings would be pretty bad, compared to what they might be now. Some new players might go weeks or months before winning a single game, at which point they would start winning more often.

Unless they played an unbelievable number of games, that would continue to tarnish their record, even if they became really good players later on.

i thought it only counted the last 3 months.

That would be nice. I don't recall seeing a time limit in the blog post, but maybe I missed it. Only counting a limited timeframe, then it wouldn't unduly burden rookies joining the ladder.

It's still gameable, but I agree with Gui that playing and winning more games should be rewarded with higher rating, so I think it would be an overall positive change.

persnally, I don't think the person that has time to sit there playing 100 games a month should show a greater skill then someone playing 10 games a month *purely because he has more games*.. skill is a measure of quality rather than quantity

While I understand that point, it really is more impressive to win 60 to 10 losses than it is to win 6 to 1 loss. The ratio is the same, but if you can win that many and maintain that consistency, it shows that it wasn't a fluke or dependent on the specific opponents.