Author Topic: Advanced stats vs Mizzou  (Read 4271 times)

0 Members and 1 Guest are viewing this topic.

Offline mocat

  • Pak'r Élitaire
  • ****
  • Posts: 40561
    • View Profile
Re: Advanced stats vs Mizzou
« Reply #25 on: January 09, 2012, 01:07:17 PM »
Sat Feb 4      9   Missouri   L, 74-73   71   47%   Away
Do you have any idea how angered up I would be if we lose to MU by 1 point?

That's the prediction for KU @ MU.

Speaking of advanced stats, kenpom's explanation of Wisconsin at #2 in his system is entertaining.

I enjoyed most of it, but he never addressed a possible issue that somehow Big 10 teams are more heavily favored than they should be in his system (4 teams in the top 8)

Offline kougar24

  • Pak'r Élitaire
  • ****
  • Posts: 5380
    • View Profile
Re: Advanced stats vs Mizzou
« Reply #26 on: January 09, 2012, 01:15:32 PM »
Speaking of advanced stats, kenpom's explanation of Wisconsin at #2 in his system is entertaining.

Quote
If you are going to trash the entire system based on the biggest outlier over the last six years, I suspect you had no desire to use the system in the first place. Either that, or your team is ranked lower than you think it should be.

get'em kenneth

Quote
Q. When are you going to fix this?

First off, I’m not sure it can be fixed. I mean, sure I can fix it so Wisconsin is ranked more reasonably. I could add some code to my algorithm like this…

if Wisconsin:
   do something to fix Wisconsin
else:
   do normal calculations

heh

That sounds like something college football fans wanted the BCS to do to K-State since we "shouldn't have won" most of our games. Granted, Wisky is a different case since they aren't winning.

Offline MakeItRain

  • Pak'r Élitaire
  • ****
  • Posts: 45942
  • big roas man
    • View Profile
Re: Advanced stats vs Mizzou
« Reply #27 on: January 09, 2012, 04:27:07 PM »
Speaking of advanced stats, kenpom's explanation of Wisconsin at #2 in his system is entertaining.

Quote
If you are going to trash the entire system based on the biggest outlier over the last six years, I suspect you had no desire to use the system in the first place. Either that, or your team is ranked lower than you think it should be.

get'em kenneth

Quote
Q. When are you going to fix this?

First off, I’m not sure it can be fixed. I mean, sure I can fix it so Wisconsin is ranked more reasonably. I could add some code to my algorithm like this…

if Wisconsin:
   do something to fix Wisconsin
else:
   do normal calculations

heh

Ken Pomeroy :flush:
This is a huge win for me.  His pompous ass response is even more win for me.  His outlier isn't like a team with 4 early losses in the top 20, Wisconsin being where they are is a giant flaw in his rankings.  I dont have a problem with him using statistical formulas to determine how good a team is and he wouldnt have an earthly clue as to what to do with game tape.  My problem is his attitude.  I'll continue to get my advanced stats from the free and infinitely superior statsheet, they have individual websites and extensive historical info for every team, while simps continue to eat the BS he feeds.

Offline sys

  • Contributor
  • Pak'r Élitaire
  • *****
  • Posts: 40815
  • your reputation will never recover, nor should it.
    • View Profile
Re: Advanced stats vs Mizzou
« Reply #28 on: January 09, 2012, 05:09:20 PM »
I enjoyed most of it, but he never addressed a possible issue that somehow Big 10 teams are more heavily favored than they should be in his system (4 teams in the top 8)

nor should he.  they aren't related.  wisconsin is highly rated because they beat the fuckshit out of bad teams,  osu is highly rated because they beat the fuckshit out of bad teams, and also out of nw and duke and only lost to good teams on the road, msu and indiana are highly rated because they've beaten a lot of good teams and only lost a small number of games to other good teams.  there is little reason to think that any of the teams other than wisconsin are misrated.


the flaw in pomeroy's ratings is that mov* between two teams of very dissimilar quality is not as useful as a predictor as mov between teams of similar quality.


* - doesn't actually use mov, i know.  don't care.
"a garden city man wondered in april if the theologians had not made a mistake in locating the garden of eden in asia rather than in the arkansas river valley."

Offline mcmwcat

  • Pak'r Élitaire
  • ****
  • Posts: 5313
  • trips: "MCMW"
    • View Profile
Re: Advanced stats vs Mizzou
« Reply #29 on: January 09, 2012, 06:23:03 PM »
i like looking at kenpoms conference only #s

Offline Fake Sugar Dick (WARNING, NOT THE REAL SUGAR DICK!)

  • Racist Piece of Shit
  • Pak'r Élitaire
  • ****
  • Posts: 18431
  • Kiss my ass and suck my dick
    • View Profile
    • I am the one and only Sugar Dick
Re: Advanced stats vs Mizzou
« Reply #30 on: January 09, 2012, 06:39:07 PM »
i like looking at kenpoms conference only #s

This.  It's Fake Sugar Dick (WARNING, NOT THE REAL SUGAR DICK!) to look at KenPom early on in conf play for obvious reasons.

goEMAW Karmic BBS Shepherd

Offline mocat

  • Pak'r Élitaire
  • ****
  • Posts: 40561
    • View Profile
Re: Advanced stats vs Mizzou
« Reply #31 on: January 09, 2012, 08:47:28 PM »
I enjoyed most of it, but he never addressed a possible issue that somehow Big 10 teams are more heavily favored than they should be in his system (4 teams in the top 8)

nor should he.  they aren't related.  wisconsin is highly rated because they beat the fuckshit out of bad teams,  osu is highly rated because they beat the fuckshit out of bad teams, and also out of nw and duke and only lost to good teams on the road, msu and indiana are highly rated because they've beaten a lot of good teams and only lost a small number of games to other good teams.  there is little reason to think that any of the teams other than wisconsin are misrated.


OK, i agree with what you're saying, but you know who else beat the fuckshit out of a combination of bad teams and some good teams, with no bad losses? Missouri, Baylor, Georgetown, Alabama, or even Marquette, etc. And they are all rated below those 4 Big 10 teams. I'm just curious IF there is some reason hidden somewhere in his formulas for the excessively heavy Big 10 presence at the top.  :dunno:

Offline BEFR

  • Fan
  • *
  • Posts: 170
    • View Profile
Re: Advanced stats vs Mizzou
« Reply #32 on: January 09, 2012, 09:34:54 PM »
robert scored some great seats... must be in charge of the stationary bicycles  :cheers: