SuperLadle Discussion - cont. from time change thread

[Concord]

round wins are accounted for elsewhere in the score, it's redundant to count both total round wins and round winning percentage.
Furthermore, round winning percentage is a weird statistic, because you can only lose one round per ladle.


@sine, not finals match total, but finals won match total.

Code: Select all

Team                fwmt     lw*2     (fwmt / lw )
Revolver            9.63     9.63     (4 / 2)
Crazy Tronners      9.20     8.59     (3 / 1)
Speeders            7.88     7.88     (2 / 1)
MeetYourMaker       7.14     6.43     (3 / 1)
TeamBaylife         7.00     7.00     (2 / 1)
Rogue Tronners      6.77     6.15     (1 / 0)
Team Unknown        5.68     5.11     (1 / 0)
Twi¦×¦ted ¦×¦ats    4.54     4.54     (0 / 0)

some differences. Only change in order is MYM being above Baylife if fwmt are counted. Worth point out that both these teams won a ladle, but Baylife is just rewarded for having a great winning percentage, this wouldn't happen with the mandated 5/9 participation. Otherwise team are just more tightly grouped. But here's why I think it might be worth it. At the margins, for teams 7-15, it's really significant to win a match in the finals. It seems like if, let say, Team Elite had made a finals there would be a significant difference between losing 2-0 and taking a match, because it's possible if unlikely to have an easy draw through the semis, but if you take a match in the finals, you're competing against someone good.

[sinewav]

I fail to see how any of what I said was "overcomplicated."

Sorry owned. That comment wasn't directed at anyone specific. I just wanted to remind everyone that this can get our of control very fast.

sine, not finals match total, but finals won match total.

Ah I see, thanks.

Something about it makes me a little uncomfortable. As unlikely as it is, there is a slight potential for abuse. I can see team in the finals throwing a match in order to push another team lower in the standings. Even worse would be for collusion between two teams. Anytime there is an opportunity where losing has potential gain we have to be careful.

Poke' Master does have a valid point about freezing the changes. If we change something now, it could really piss some people off when they don't make the cutoff. I don't want the Bowl to fall apart because of a management error. Personally, I think the current formula (rw+lw)*(mw/mt) is an acceptable method of determining performance. We can tweak this all month and still disagree because there is some subjectivity here -- what makes a great team? So, maybe we should wait until the May and see how this goes first?

[Concord]

yeah.


current standings based on that formula are and have been at http://playfortress.wordpress.com/ for a couple weeks now. there's some other bowl-related content there too. [/shameless plug]

[0000]

Good analysis Concord, thanks for doing this. I think it is fun to be able to talk about it as a "season" with the new bowl format.

I would suggest for next season to use an elo formula or similar and give minor bonus points to Ladle winners and possibly finalists. This would solve the bye issue and the issue of some teams getting better draws than others. For the bubble teams, a season of bad draws might make the difference since the sample size is so small.