Looking at KenPom, the Big10

weatherbird

All-American
Aug 1, 2006
5,930
9,558
113
Dude. You are wrong. Period. The only thing in the model now is what happened this year. There is ZERO EFFECT of preseason assumptions. They do not enter into the calculations whatsoever. The only data entered is what happened this year. There is no such thing as "they played in game 12 and they were assumed to be so strong at that point factoring in preseason data." We beat a team in week 12 and that team's strength is 100% based on what that team did this year.

Yes, and what that team did this year is determine an adjusted efficiency, which is adjusted according to the quality of the opponent. The derived adjusted efficiency is based on preseason assumptions. So even though at January 22 or whatever, his preseason assumptions are completely removed, the efficiency of the team.at that point is tied to the preseason rankings. As he says in his blog, the preseason assumptions are a factor in the main equation. And in january he starts diminishing the said factor, until the factor is completely removed. That removes the factor, as he says. From that respect, it is completely removed. The problem is the adjusted efficiency of that team at that moment in time is still based on the preseason assumption. What he is talking about in his blog is removing the preseason factor in the equation. That makes the rankings more volatile as he suggests. Do you know why he still included the preseason rank when showing the table?? Because you have to start with the teams ranked somewhere in order to calculate the strength of the opponent, and thus adjust the efficiency. If you didnt do that, the equation would be really unstable in trying to factor in the strength of the opponent, or wildly incorrect if you didnt. In order to factor in the strength of the opponent, there has to be a starting point. That starting point influences your adjusted efficiency, which then influences your strength as an opponent moving forward. And even though with more data, there is convergence, it still has influence.

As to why KenPom is the best, because for the most part, his preseason assumptions and the way he derives them are pretty good, and in the ballpark for most teams. My argument was never that it was crap, or that it doesnt work or any of that. My sole argument is that preseason assumptions do still affect the rankings to some degree even after that little 5 game factor is taken completely out of the equation.
 
Mar 13, 2004
14,745
12,925
0
Yes, and what that team did this year is determine an adjusted efficiency, which is adjusted according to the quality of the opponent. The derived adjusted efficiency is based on preseason assumptions. So even though at January 22 or whatever, his preseason assumptions are completely removed, the efficiency of the team.at that point is tied to the preseason rankings. As he says in his blog, the preseason assumptions are a factor in the main equation. And in january he starts diminishing the said factor, until the factor is completely removed. That removes the factor, as he says. From that respect, it is completely removed. The problem is the adjusted efficiency of that team at that moment in time is still based on the preseason assumption. What he is talking about in his blog is removing the preseason factor in the equation. That makes the rankings more volatile as he suggests. Do you know why he still included the preseason rank when showing the table?? Because you have to start with the teams ranked somewhere in order to calculate the strength of the opponent, and thus adjust the efficiency. If you didnt do that, the equation would be really unstable in trying to factor in the strength of the opponent, or wildly incorrect if you didnt. In order to factor in the strength of the opponent, there has to be a starting point. That starting point influences your adjusted efficiency, which then influences your strength as an opponent moving forward. And even though with more data, there is convergence, it still has influence.

As to why KenPom is the best, because for the most part, his preseason assumptions and the way he derives them are pretty good, and in the ballpark for most teams. My argument was never that it was crap, or that it doesnt work or any of that. My sole argument is that preseason assumptions do still affect the rankings to some degree even after that little 5 game factor is taken completely out of the equation.

Theres no other way to say that this is factually incorrect and the models do not work that way. What we did in game 8 is not based on what the model thought team 8 was like at the time. Old calculations dont stay in. The model is run fresh every time and every calculation is redone. And when those calculations are redone, there is no data from last year, no assumptions from preseason, in any form.
 
May 27, 2007
32,244
25,613
113
It does matter because the committee has said they look at it. It also matters because the committee is always looking for ANY reason to underrate UK. If they can offer any plausible answer for why they rated Kentucky so differently from their ranking, they will use it and do just that.

One final possibility why the efficiency ratings don't tell the story on some teams, like UK this season, is because it is difficult to measure the quality of a team that just refuses to lose, even in games they could be in danger of losing. Back to my Purdue comparison, and someone mentioned it above as a detractor for this UK team, the Boilermakers have had some close games as well, but they've lost 14 of them. Still, since their efficiency numbers are better, they are better in KenPom. Are they better, though? Is a team that has blown out some good teams but also lost 14 games really better than a team that has only lost 5 games all season, despite playing in close games all season? KenPom can't measure heart and unwillingness to lose. That's the best quality of this Kentucky team.

They get Kenpom numbers but even Ken would tell you he doubts they use his figures.

The use the team sheets. The figures on those team sheets are 95% NET based. The quads they place so much importance on.......all based on NET.

So that's really the important figure........not Kenpom. Never have we ever been dinged a seed and that was their reasoning. I never heard the committee mention Kenpom at all.

As far as the last part about UK, we'll have to agree to disagree. You don't think that maybe UK isn't losing these games not because they have an unwillingness to lose but rather because they are playing in the SEC against weaker competition?

Auburn is 24-4. They are in the same category to me. Is it their unwillingness to lose or is it simply the fact they are in a weaker league where they are better than the vast majority of the teams.

FWIW this notion that the system can't measure heart is bogus. If you show heart to win a ballgame instead of lose one, that's reflected. While it's true in the system it's better to win a game by 25 instead of 5, it's also better to win a game by 5 than lose one by 5. The system DOES reflect winning.
 

Sparkaces

All-American
Sep 19, 2012
6,157
6,289
0
I just don’t understand how you give up 113 and your def efficiency drops 2 spots. I also don’t understand how UK IS behind a 14-14 team!
 

revcort

Heisman
Feb 20, 2003
32,522
30,904
113
They get Kenpom numbers but even Ken would tell you he doubts they use his figures.

The use the team sheets. The figures on those team sheets are 95% NET based. The quads they place so much importance on.......all based on NET.

So that's really the important figure........not Kenpom. Never have we ever been dinged a seed and that was their reasoning. I never heard the committee mention Kenpom at all.

As far as the last part about UK, we'll have to agree to disagree. You don't think that maybe UK isn't losing these games not because they have an unwillingness to lose but rather because they are playing in the SEC against weaker competition?

Auburn is 24-4. They are in the same category to me. Is it their unwillingness to lose or is it simply the fact they are in a weaker league where they are better than the vast majority of the teams.

FWIW this notion that the system can't measure heart is bogus. If you show heart to win a ballgame instead of lose one, that's reflected. While it's true in the system it's better to win a game by 25 instead of 5, it's also better to win a game by 5 than lose one by 5. The system DOES reflect winning.
On that final point, I would agree the system can measure a win versus a loss, but a team like UK, which has been in so many close games and won, gets penalized for not winning by a larger margin. We play to the level of the opponent. We played some of our best games against Louisville, Mich St, Texas Tech, at LSU. But when you have as many close games as this UK team has, even if you win nearly all, the numbers will be bad. And as I said elsewhere, these numbers also don't reflect the superbowl effect UK faces or the bad whistle we always get on the road. That superbowl effect takes out your theory about playing lesser opponents.

Here's what I think my friend- for whatever reason, KenPom has UK rated lower than almost any other system. Look at the Massey Ratings, which takes into account 59 ratings system at this point. UK is now up to #17 in that average. KenPom is one of the outliers. Only 3 other systems have UK rated worse than KenPom.

For reference: https://www.masseyratings.com/cb/compare.htm
 

caneintally

Heisman
Oct 1, 2002
27,455
17,056
0
They get Kenpom numbers but even Ken would tell you he doubts they use his figures.

The use the team sheets. The figures on those team sheets are 95% NET based. The quads they place so much importance on.......all based on NET.

So that's really the important figure........not Kenpom. Never have we ever been dinged a seed and that was their reasoning. I never heard the committee mention Kenpom at all.

As far as the last part about UK, we'll have to agree to disagree. You don't think that maybe UK isn't losing these games not because they have an unwillingness to lose but rather because they are playing in the SEC against weaker competition?

Auburn is 24-4. They are in the same category to me. Is it their unwillingness to lose or is it simply the fact they are in a weaker league where they are better than the vast majority of the teams.

FWIW this notion that the system can't measure heart is bogus. If you show heart to win a ballgame instead of lose one, that's reflected. While it's true in the system it's better to win a game by 25 instead of 5, it's also better to win a game by 5 than lose one by 5. The system DOES reflect winning.
Yeah but they care about the NET for the Quad win values and whether you are 1 in NET or 25 in NET doesn't matter still quad 1 if you get what I am saying. . There just is not much need to worry about these dumb computers beyond what talking heads like to discuss. But you are correct NET is the only one that matters at all and that only matters in the number of quad 1 & 2 wins and quad 3& 4 losses .

What KenPom or Saragin ( though they have us about right ) say is not important in the least.
 

*Fox2Monk*

Heisman
Jun 10, 2009
49,283
91,116
113
Sorry but the B1G would have maybe 7 tourney teams in a regular year. No way you should get a bid being .500. Look at the standings. The middle has the same records. Yeah they may be decent teams, but no way 10 teams in that conference are in the top 36 best teams this year.
 
Jan 30, 2018
16,155
24,636
0
One stat I think we should be ranked better is our defense. I mean it's hard as hell to score 70 on us and lately, 60 is tough. I know other things go into it, but dam out of all the BB I have watched this year I would put us as a Top 10 team on defense. I can't remember where we are, but I was surprised when I saw it. Does anyone know where we are ranked?
 
May 27, 2007
32,244
25,613
113
On that final point, I would agree the system can measure a win versus a loss, but a team like UK, which has been in so many close games and won, gets penalized for not winning by a larger margin. We play to the level of the opponent. We played some of our best games against Louisville, Mich St, Texas Tech, at LSU. But when you have as many close games as this UK team has, even if you win nearly all, the numbers will be bad. And as I said elsewhere, these numbers also don't reflect the superbowl effect UK faces or the bad whistle we always get on the road. That superbowl effect takes out your theory about playing lesser opponents.

Here's what I think my friend- for whatever reason, KenPom has UK rated lower than almost any other system. Look at the Massey Ratings, which takes into account 59 ratings system at this point. UK is now up to #17 in that average. KenPom is one of the outliers. Only 3 other systems have UK rated worse than KenPom.

For reference: https://www.masseyratings.com/cb/compare.htm

I don't disagree with any of that. Just as I feel Sagarin is probably an outlier at #7, Kenpom is also an outlier at 26. I've always felt we were in the 15-20 range TBH.

Taking the average of all systems which is what Massey does is always going to take you closer. Wisdom of the crowds working.

As far as the superbowl effect I do believe it's overblown. I'm not saying it doesn't occur but I think we tend to believe we are the only team that has to deal with it and I just don't believe that to be the case. You don't think teams get fired up when they see Kansas or a Duke. It doesn't have to even be a blueblood. What if Baylor comes to town now? You believe teams don't get up for that?

Same goes with bad whistles. UK isn't the only team that deals with bad whistles. So while the system doesn't see "well yeah this was a close game however the refs were awful and that's the reason", that stuff SHOULD balance out. I know a lot of people think refs are out to get us. I just think they are bad in general.

I guess the last point to make about close wins is this. There's two ways to look at it. One would be look at all these games we were able to close out at the end. We showed a will to win and determination (tho I believe being a good FT shooting team certainly gives us advantage in these games as well). Or you can look at it and wonder why are we in all these close games to begin with. It's one thing when it's Texas Tech or LSU on the road. But games vs Vandy, Ole Miss even Texas A&M on the road. And I feel if we are being honest it's a bit of both...........we do win close games, that will help us going forward. But the lack of blowouts should also be a cause of concern as well.
 
  • Like
Reactions: revcort
May 27, 2007
32,244
25,613
113
I just don’t understand how you give up 113 and your def efficiency drops 2 spots. I also don’t understand how UK IS behind a 14-14 team!

I'm not clear on what team you're talking about but in general think of it like this.

Say based on a prediction Team A is supposed to score 80 points. And they are supposed to do it in 60 possessions. So 1.33 points per poss. Which is absolutely a monster offensive game right.

Now say you only put up 1.25.

Your offensive rating despite having an outstanding game still on offense, will likely drop.

So it all boils down to expected vs actual with this stuff. It's all in the context of what you were expected to do based on the talent level of each team, vs what actually happened.
 
May 27, 2007
32,244
25,613
113
Yeah but they care about the NET for the Quad win values and whether you are 1 in NET or 25 in NET doesn't matter still quad 1 if you get what I am saying. . There just is not much need to worry about these dumb computers beyond what talking heads like to discuss. But you are correct NET is the only one that matters at all and that only matters in the number of quad 1 & 2 wins and quad 3& 4 losses .

What KenPom or Saragin ( though they have us about right ) say is not important in the least.

In reality the committee only cares about NET. Because they are looking at the NET number and the Quads are made from the NET numbers.

I look at NET and team sheets for seeding purposes.
I look at other systems if I really want to have a good estimation on team strength tho.
 
May 27, 2007
32,244
25,613
113
One stat I think we should be ranked better is our defense. I mean it's hard as hell to score 70 on us and lately, 60 is tough. I know other things go into it, but dam out of all the BB I have watched this year I would put us as a Top 10 team on defense. I can't remember where we are, but I was surprised when I saw it. Does anyone know where we are ranked?

So a couple of things.

We are currently 45th in adjusted defensive efficiency.

And you are correct. But we need to get away from looking at total score as a barometer for this stuff. The reason teams score a low amount on us is because we are only 213th in adjusted tempo.

It's basically the opposite of Virginia last year. That team was a very good offensive team, but no one thought they were because the games were in the 50s/60s. The same thing here is occurring with UK to a certain extent.

Our defensive rating is 93.3. It means that an average NCAA team would score about .933 points per poss on us.

Why are we only 45th. The thing is this:
Our effective FG% defense (same as reg FG% just weighing the 3 more because more points obviously) is outstanding. We rank 15th. We always rank very high under Cal in this regard. We are damn tough to make shots against.

But to be an elite defensive team, you need more than this.

We are 276th in forcing turnovers.
We are 187th in rebounding % on defense.
And we are 163rd in Free Throw Rate.

These numbers aren't just mediocre, they are BAD. Combining the turnover number and rebounding, despite being tough to shoot against, teams are putting up more shots against us.

And this is the reason our defense is just 45th and not better.


And i think this is an important note. When people look at Kenpom, most just see the front page. Most look at the rank or they look at the offense/def eff numbers. But looking at the underlying figures it's not that hard to see why a team is rated where they are.
 
  • Like
Reactions: Rideordiewildcat
KSR

Latest

  1. 01 KSR Football Kentucky vs. Missouri Availability Report: Starting WR Could Return
  2. 02 KSR Basketball Jerone Morton nears Kentucky debut, Trent Noah officially week-to-week
  3. 03 KSR Football Will Stein Show Recap: Who's Improved the Most for Kentucky?
  4. 04 KSR Basketball Kam Williams is showing rust, but Mark Pope is confident he'll 'get his vibe back'
  5. 05 KSR Football Will Stein knows Missouri is better than its recent losses suggest