1/24 Bracketology

Status
Not open for further replies.
#151      

Illini2010-11

Sugar Grove
Wins and losses do not factor into KenPom, so yes they would have more wins in the WCC and be higher in their conference standings but their strength of schedule adjusted net rating would not really change.
That is true, but Minnesota's effiency ratings would likely be better playing in WCC, which would improve their ranking. So yes, wins and losses are not a direct input, but it is also indirect. There are many attributes that dictate KenPom, not simply strength of schedule...so it is not crazy to think MN would perform better in metrics if they played in WCC.
 
#153      
KenPom is based on net rating adjusted for strength of schedule so Minnesota's rating would not get a big bump by playing in the WCC since it already accounts for strength of schedule
This is circular logic. How is strength of schedule determined? Wins and losses still matter. It definitely matters for the Net rankings.
 
#154      
This is circular logic. How is strength of schedule determined? Wins and losses still matter. It definitely matters for the Net rankings.
Strength of schedule is determined by the strength of your opponents, which is determined by the strength of their opponents, which is determined by their opponents, etc... which you are right is circular and the ratings have to be iterated (which is what rating systems like KenPom do). As for the Net it is mainly based on adjusted net ratings because Ohio State is still #29 at 11-10. Point being is that Minnesota's current KenPom rating would not really change if they moved to the WCC tomorrow because yes, while their team efficiency would improve, it would be against worse opponents which would offset the improvement. If that were false then there would be a huge market inefficiency in vegas lines against high major and mid major teams (since Vegas lines closely track Kenpom lines).
 
Last edited:
#155      
Strength of schedule is determined by the strength of your opponents, which is determined by the strength of their opponents, which is determined by their opponents, etc... which you are right is circular and the ratings have to be iterated (which is what rating systems like KenPom do). As for the Net it is mainly based on adjusted net ratings because Ohio State is still #29 at 11-10. Point being is that Minnesota's current KenPom rating would not really change if they moved to the WCC tomorrow because yes, while their team efficiency would improve, it would be against worse opponent which would offset the improvement. If that were false then there would be a huge market inefficiency in vegas lines against high major and mid major teams.
And that is why the Net ranking needs to be adjusted because wins and losses matter. I don’t care if you lose all your games by 1 point you still didn’t win but your Net could be really good
 
#156      

lstewart53x3

Scottsdale, Arizona
And that is why the Net ranking needs to be adjusted because wins and losses matter. I don’t care if you lose all your games by 1 point you still didn’t win but your Net could be really good
By that logic, UNC Wilmington, who’s currently 153 in KenPom with losses to UCONN by 36, Oklahoma by 21, and Hofstra by 24, should be ahead of us in the polls because they have a better W/L record.

Who you play & how well you play plays a big role in projecting how well a team will play in the future.
 
#157      

Illini2010-11

Sugar Grove
Strength of schedule is determined by the strength of your opponents, which is determined by the strength of their opponents, which is determined by their opponents, etc... which you are right is circular and the ratings have to be iterated (which is what rating systems like KenPom do). As for the Net it is mainly based on adjusted net ratings because Ohio State is still #29 at 11-10. Point being is that Minnesota's current KenPom rating would not really change if they moved to the WCC tomorrow because yes, while their team efficiency would improve, it would be against worse opponents which would offset the improvement. If that were false then there would be a huge market inefficiency in vegas lines against high major and mid major teams (since Vegas lines closely track Kenpom lines).
We are not talking about taking the current ratings and then moving to WCC tomorrow. The theoretical assumption was that MN played in WCC the entire year with their current squad. The strength of schedule would have been worse, but I would anticipate that the efficiency metrics would be substantially better for MN as a whole, thus improving their net ratings. MN's terrible efficiency metrics in non-conference could never be overcome in the B1G and got even worse once league play began. However, I can nearly guarantee it looks much different and closer to 100 than 191 if they played entire year in WCC (assuming they were able to compete for top 4 in the league, which I personally still think is a stretch). When you start winning basketball games, your metrics begin to get much better by nature. Ohio State is a curious case of being propped up because, up to this week, they were only losing close single-digit games to better than average teams (not including terrible loss to MN). MN on the other hand, was getting blasted regularly. I do not think that would have happened in the WCC for them (with exception of possibly playing St. Mary's and Gonzaga), especially as they start winning basketball games. Therefore, it still is not a stretch that Minnesota could have been close to top 100 in KenPom if they played in WCC this year.
 
Last edited:
#158      
We are not talking about taking the current ratings and then moving to WCC tomorrow. The theoretical assumption was that MN played in WCC the entire year with their current squad. The strength of schedule would have been worse, but I would anticipate that the efficiency metrics would be substantially better for MN as a whole, thus improving their net ratings. MN's terrible efficiency metrics in non-conference could never be overcome in the B1G and got even worse once league play began. However, I can nearly guarantee it looks much different and closer to 100 than 191 if they played entire year in WCC (assuming they were able to compete for top 4 in the league). When you start winning basketball games, your metrics begin to get much better by nature. Ohio State is a curious case of being propped up because, up to this week, they were only losing close single-digit games to better than average teams (not including terrible loss to MN). MN on the other hand, was getting blasted regularly. I do not think that would have happened in the WCC for them (with exception of possibly playing St. Mary's and Gonzaga), especially as they start winning basketball games. Therefore, it still is not a stretch that Minnesota could have been close to top 100 in KenPom if they played in WCC this year.
A lot of assumption and conjecture here. What is the point of this continued arguing about it? You can say yes it will and more people can say no it wont. I "know" you have no idea how KenPom calculates SOS and adjusted EM. So there.
 
#159      
You guys are talking in circles.

illiniballknower is correct in that KenPom certainly is attempting to adjust for a team's offensive and defensive efficiency based on their opponent's defensive (when looking at offense) and offensive (when looking at defense) efficiency. This is done to try and normalize the teams to determine how they would perform against the most average offense or defense (theoretically). So even if you have a hyper-efficient offensive outing against the worst rated defensive team in the country (currently Houston Christian), you don't get much of a boost for that because KenPom ratings say even a very average offense should be able to do that. Statistically that's just a fact since they are just methodically calculated numbers that have proven robust.

Illini2010-11 is essentially saying nothwithstanding the normalization, it's possible that a team with Big Ten level talent would have enough of a talent edge, that they'd perform significantly better than they have and thus much closer to a better than average efficiency team against the bad level of competition than they have been able to do as a poor Big Ten team against tougher competition. This can't really be proven one way or the other, but goes more into matchups and stuff that dictate Basketball not being played solely on spreadsheets.
 
#160      

lstewart53x3

Scottsdale, Arizona
We are not talking about taking the current ratings and then moving to WCC tomorrow. The theoretical assumption was that MN played in WCC the entire year with their current squad. The strength of schedule would have been worse, but I would anticipate that the efficiency metrics would be substantially better for MN as a whole, thus improving their net ratings. MN's terrible efficiency metrics in non-conference could never be overcome in the B1G and got even worse once league play began. However, I can nearly guarantee it looks much different and closer to 100 than 191 if they played entire year in WCC (assuming they were able to compete for top 4 in the league, which I personally still think is a stretch). When you start winning basketball games, your metrics begin to get much better by nature. Ohio State is a curious case of being propped up because, up to this week, they were only losing close single-digit games to better than average teams (not including terrible loss to MN). MN on the other hand, was getting blasted regularly. I do not think that would have happened in the WCC for them (with exception of possibly playing St. Mary's and Gonzaga), especially as they start winning basketball games. Therefore, it still is not a stretch that Minnesota could have been close to top 100 in KenPom if they played in WCC this year.
Efficiency ratings like NET take strength of opponent into account. Instead of getting blasted in the B1G, they might have squeaked out wins in the WCC. But because the strength of their schedule in the WCC is weaker, those extra wins would not help their opponent adjusted efficiency ratings. Playing in the WCC would have no impact on Minnesota‘s NET ranking.

If getting wins in a weaker conference helped a teams NET ranking, the top teams in the NET would play in weaker conferences.
 
#161      

lstewart53x3

Scottsdale, Arizona
You guys are talking in circles.

illiniballknower is correct in that KenPom certainly is attempting to adjust for a team's offensive and defensive efficiency based on their opponent's defensive (when looking at offense) and offensive (when looking at defense) efficiency. This is done to try and normalize the teams to determine how they would perform against the most average offense or defense (theoretically). So even if you have a hyper-efficient offensive outing against the worst rated defensive team in the country (currently Houston Christian), you don't get much of a boost for that because KenPom ratings say even a very average offense should be able to do that. Statistically that's just a fact since they are just methodically calculated numbers that have proven robust.

Illini2010-11 is essentially saying nothwithstanding the normalization, it's possible that a team with Big Ten level talent would have enough of a talent edge, that they'd perform significantly better than they have and thus much closer to a better than average efficiency team against the bad level of competition than they have been able to do as a poor Big Ten team against tougher competition. This can't really be proven one way or the other, but goes more into matchups and stuff that dictate Basketball not being played solely on spreadsheets.
It’s pretty easy to prove otherwise.

Depaul is a bad basketball team. They’re 135 in KenPom. They’re 9-13. They’re ranked 9th in the Big East.

They beat Minnesota by 16.

Playing in a weaker conference would have no impact on Minnesota’s NET ranking.
 
#164      

lstewart53x3

Scottsdale, Arizona
Minnesota also beat Ohio State.
🤦‍♂️

My point is:

Minnesota is currently 191 in KenPom which is opponent adjusted efficiency metrics.

If Minnesota played in a weaker conference, where they played more teams like Depaul, their efficiency might improve and they might win more games, but that would not help their opponent adjusted efficiency ratings and they’d likely find them themselves right around where they are now.
 
#165      

Illini2010-11

Sugar Grove
We’re up to a 6-seed on Bracket Matrix. The other 6’s? Auburn, Indiana and Duke 🤮
I believe we have a real shot at getting as high as a 4 or 5 seed. If Illini can win the remaining 5 home games and win 2 out of the 5 road games, then that would get us to 22-9 (13-7). I think we have a really good chance at a 5 seed if we finish with around 22 wins.

Still a long way to go, but it is refreshing to be trending upward after it looked like the wheels came off in Evanston. Still plenty of opportunities to improve upon the 6/7 seed line that we are currently projected.
 
#167      
By that logic, UNC Wilmington, who’s currently 153 in KenPom with losses to UCONN by 36, Oklahoma by 21, and Hofstra by 24, should be ahead of us in the polls because they have a better W/L record.

Who you play & how well you play plays a big role in projecting how well a team will play in the future.
no I am saying W/L should be a factor. An ugly win should be rewarded more than a pretty loss and a team should be punished for a bad loss. I think Net overall is good but there should be a way to modify it slightly

UCONN, O$U, even WSU at 68 (10-13) record are examples why metrics alone don’t paint the whole picture. I have no problems with Saint Mary’s at 6 but Creighton at 18?
 
#168      
no I am saying W/L should be a factor. An ugly win should be rewarded more than a pretty loss and a team should be punished for a bad loss. I think Net overall is good but there should be a way to modify it slightly

UCONN, O$U, even WSU at 68 (10-13) record are examples why metrics alone don’t paint the whole picture. I have no problems with Saint Mary’s at 6 but Creighton at 18?
Wins and Losses are the primary metric. NET is a tool more so to judge the quality of such wins and losses (and thus strength of schedule) more so than a means of ranking the teams. There's a reason they don't just seed teams in accordance with NET rating and call it a day.
 
#169      
Confused Always Sunny GIF by It's Always Sunny in Philadelphia
 
#170      
As important as seeding is, I really, really, really hope we break our unlucky streak of the last two years where our R32 opponent was vastly under-seeded compared to their actual quality.

2021 - 8 seed Loyola-Chicago was a "true" 3-seed (KP #9)
2022 - 5 seed Houston was a "true" 1-seed (KP #4)

So, basically, we played 2nd-round games as if we were a 6 seed and an 8 seed. That's gotta end eventually, right?
We're going to be the vastly under seeded team this year.
 
#172      

redwingillini11

White and Sixth
North Aurora
If we do well down the stretch and end up with the 3 seed in the conference tournament:

I'm not saying it would be a cakewalk for us to the B1G title game, but I'm not not saying that...
 
#173      
If Minnesota played in a weaker conference, where they played more teams like Depaul, their efficiency might improve and they might win more games, but that would not help their opponent adjusted efficiency ratings and they’d likely find them themselves right around where they are now.

Just to echo this, KenPom has spent years researching this, tweaking his metrics as more seasons are played, in an attempt to normalize teams regardless of the competition they play. There's a very large, multi-season body of evidence to say that on balance, a team's schedule can be normalized.

No one can prove a what-if about changing conferences one way or the other, but if you put some stock in Kenpom's data, it's been accounted for. It wouldn't surprise me if he could study this given teams do move conferences, but the sample size might not be big enough to draw much of a conclusion. I think one of the validation's of Kenpom is how closely his metric works for predicting Vegas lines, regardless of the fact that team's often play much different competition.
 
#174      

Illini2010-11

Sugar Grove
Just to echo this, KenPom has spent years researching this, tweaking his metrics as more seasons are played, in an attempt to normalize teams regardless of the competition they play. There's a very large, multi-season body of evidence to say that on balance, a team's schedule can be normalized.

No one can prove a what-if about changing conferences one way or the other, but if you put some stock in Kenpom's data, it's been accounted for. It wouldn't surprise me if he could study this given teams do move conferences, but the sample size might not be big enough to draw much of a conclusion. I think one of the validation's of Kenpom is how closely his metric works for predicting Vegas lines, regardless of the fact that team's often play much different competition.
This is 100% accurate, and I apologize to the board, because I believe that my statements were causing contention last night, when in fact it was not intended.

KenPom has done an excellent job, as a whole, in determining strength of a team, and I believe it is one of the best metrics around. Minnesota would be around 190 regardless of what conference they played in, because they are truly a poor team by whatever metric you want to use, and they may squeak out a couple more games in a lesser conference but would not move the needle in KenPom. KemPom is actually one of the tools that I take a look at before making sport bets.

When I was speaking last night, rather ineloquently, was that I believe State of Chicago (the initial post that started this discussion) thinks that Minnesota would be competing for a top 4 position in WCC (likely based on perceived talent of the worst B1G roster). In order to get to top 4 in that league, unless you have a ton of "lucky" type factor wins to get there, they would be somewhere in the 100 range of KenPom (as reference San Francisco is 6th in WCC and ranked 97 KenPom). Could Minnesota get there? Maybe in 1-2% of models at best. The extreme likelihood is that Minnesota hovers very close to their current ranking if they played in the WCC (or any conference for that matter). Unless defying general statistical odds, it would be very difficult to obtain a top 4 position in WCC this year and not be around 100 KenPom at worst. In no way do I think MN would be anywhere close to that number...they are really bad (and Illini need to blow them out next Tuesday for good measure).
 
Status
Not open for further replies.