Take a look at the Sagarin Ranking as of 12-4-2011. Enjoy it while it lasts....
http://www.usatoday.com/sports/sagarin/bkt1112.htm (http://www.usatoday.com/sports/sagarin/bkt1112.htm)
Maybe one of you statheads can help me out with the Pomeroy numbers. How is it that Bucky loses to both UNC and MU this week and stays 3rd in his rankings? Is the ranking just an average of adjusted O and adjusted D?
Um, it looks like they're #17. Anyway, seeing Marquette at the top of any comprehensive national ranking is awesome. This team is sick.
Quote from: windyplayer on December 04, 2011, 11:10:41 AM
Um, it looks like they're #17. Anyway, seeing Marquette at the top of any comprehensive national ranking is awesome. This team is sick.
Sagarin they're 17, Pomeroy they're 3. Sorry, I should have been clearer in my question,
Pomeroy is here: http://www.kenpom.com/ (http://www.kenpom.com/)
Washington - #94.
Quote from: marqptm on December 04, 2011, 11:15:12 AM
Washington - #94.
Touche, still not clear enough. My question was in regards to directional school UW hyphen Madison.
Quote from: mugrad2006 on December 04, 2011, 11:16:43 AM
Touche, still not clear enough. My question was in regards to directional school UW hyphen Madison.
It wasn't directed at you, but Wisconsin is #17 is his updated rankings...
Quote from: mugrad2006 on December 04, 2011, 11:01:48 AM
Maybe one of you statheads can help me out with the Pomeroy numbers. How is it that Bucky loses to both UNC and MU this week and stays 3rd in his rankings? Is the ranking just an average of adjusted O and adjusted D?
I think it's adjusted O minus adjusted D, but I could be mistaken.
Quote from: mugrad2006 on December 04, 2011, 11:01:48 AM
Maybe one of you statheads can help me out with the Pomeroy numbers. How is it that Bucky loses to both UNC and MU this week and stays 3rd in his rankings? Is the ranking just an average of adjusted O and adjusted D?
I found this...but basically he weights preseason rankings of returning players and expected impact of newcomers early in the season and eventually these weightings drop out as more sample size (games) occur.
Quote
At the beginning of the year, he ranks the teams based on returning players and expected impact of new players. This is his opinion and it should be looked at in just that way. It is the opinion of one person who has made his reputation on a mathematical model, not scouting players.
At the beginning of the year, he gives a lot of weight to his preseason opinion and not a lot of weight to the games played. Slowly, as the season progresses, the opinion has less and less weight and the outcomes of the games more and more. I think the preseason opinion is totally out of his numbers by the end of January, but any kind of real impact from the projections are gone by the first week of January.
It is interesting that people like his mathematical analysis more than some of the others, but be careful when looking back at his numbers, since his formula is designed to look good post season and not so much as a predictor. I base this on the way he weights the games. I think he weighs later games too heavily.(the 30th game of the season counts more than games 1 to 7...all of them... one game counts more than 7)
Quote from: marqptm on December 04, 2011, 11:19:22 AMIt wasn't directed at you, but Wisconsin is #17 is his updated rankings...
But he was talking about kenpom.com, you're talking about Sagarin. Bucky is still #3 in kenpom as of this morning (though we're up to #6).
Honestly, that confused the hell out of me. I think it may be because the gap between Wisconsin and Syracuse was so great at the time. Wisconsin was right behind #2 Ohio State yesterday, now they are a full 0.0051 back, which seems microscopic, but in kenpom, is a decent amount.
I think it's because of their defensive rating, and I also believe that because it's tempo-free ratings, that tends to help Wisconsin. Think about it like this. The average team has around 68 possessions per game that affect their rating. Wisconsin has 58. So if you hold them to below 0.9 ppp, it isn't as big of a hit to their offensive rating because there are 10 less possessions counted. Defensively, they still held us fairly low yesterday, and they still have the best eFG% in the nation (which as we all know, is the most important defensive stat by about 7,324 times) because we still only shot 38.4% from the field (18.2% from 3).
For kenpom, the tempo-free numbers are going to help a team that is as efficient as Wisconsin is, and their offensive rating isn't going to be damaged by playing the cupcakes they've played before this week. As long as they score a healthy amount per possession, it will take more than 1-2 games to really hurt their offensive rating, and as long as they hold teams around 1.0 ppp even in losses, their defensive rating will still be near the top of the nation.
However, despite them being ranked 3rd in kenpom and us being ranked 6, I take solace knowing that the most important stat between us is still 61-54 :D
Thanks Blackheart. I guess I should be a little more critical of Pomeroy going forward rather than treating his analysis as the end all be all of statistical analysis (not just because of the current rankings, but in general).
Quote from: mugrad2006 on December 04, 2011, 11:50:50 AM
Thanks Blackheart. I guess I should be a little more critical of Pomeroy going forward rather than treating his analysis as the end all be all of statistical analysis (not just because of the current rankings, but in general).
No problem. KenPom does make tweaks to his credit. I think he changed his predictive formula this season based on this criticism already. The point is, at this time in the season, the pre-season weights can lead to anomalies like Bucky ahead of us (and his prediction of a 13% MU win at the Kohl). By January, he will be all 2011-12.
Quote from: mugrad2006 on December 04, 2011, 11:01:48 AM
Maybe one of you statheads can help me out with the Pomeroy numbers. How is it that Bucky loses to both UNC and MU this week and stays 3rd in his rankings? Is the ranking just an average of adjusted O and adjusted D?
there is a preseason starting point each team is given, which Wisconsin is still benefiting from. However, after about 10 games (can't remember the exact number) that preseason bias is dropped and the teams ratings are based PURELY on the games played. i believe Marquette is actually #1 in the country now based on just the results of the season, but because we were plugged in as being 20th to 30th before the season we are held back because that is still 20% or 30% of the rating, just as Wisconsin's higher starting point is.
Just watching the game yesterday, I believe it is pretty evident that the gap between MU and Wisconsin talent is widening to a point that even Bo's excellent scheme is going to be stretched further and further to be able to beat Marquette.