Aike’s Model Update - 02/27/25

[Aike]

1. Auburn 4.35
2. Duke 3.74
3. Tennessee 3.17
4. Houston 2.95
5. Gonzaga 2.87
6. Florida 2.82
7. Arizona 2.72
8. Iowa St 2.57
9. Texas Tech 2.46
10. Missouri 2.37
11. Alabama 2.36
12. Kansas 2.13
13. UConn 2.12
14. Maryland 2.09
15. Clemson 2.0781
16. Kentucky 2.0778
17. St John’s 2.067
18. Purdue 2.02
19. Michigan 1.9794
20. Mississippi St 1.9788
21. Michigan St 1.93
22. BYU 1.91
23. Marquette 1.86
24. UCLA 1.83
25. Mississippi 1.77
26. Wisconsin 1.66
27. Arkansas 1.64
28. Vanderbilt 1.612
29. Louisville 1.608
31. Georgia 1.54
38. Texas A&M 1.32
43. Oklahoma 1.19
45. Texas 1.18
48. N Carolina 1.14
53, Indiana 0.96
73. LSU 0.66
88. S Carolina 0.44

Nice to see us continue to improve game over game. Playing at a Sweet 16/Elite 8 level, with some obvious upside potential beyond that.

Duke is starting to remind me of some Gonzaga teams of recent years. Not in how they play, but in the way they are rolling over a weak conference. They are good, but the last time they faced a team with a pulse we saw what happened.

 

[ImTheVillageIdiot]

1. Auburn 4.35
2. Duke 3.74
3. Tennessee 3.17
4. Houston 2.95
5. Gonzaga 2.87
6. Florida 2.82
7. Arizona 2.72
8. Iowa St 2.57
9. Texas Tech 2.46
10. Missouri 2.37
11. Alabama 2.36
12. Kansas 2.13
13. UConn 2.12
14. Maryland 2.09
15. Clemson 2.0781
16. Kentucky 2.0778
17. St John’s 2.067
18. Purdue 2.02
19. Michigan 1.9794
20. Mississippi St 1.9788
21. Michigan St 1.93
22. BYU 1.91
23. Marquette 1.86
24. UCLA 1.83
25. Mississippi 1.77
26. Wisconsin 1.66
27. Arkansas 1.64
28. Vanderbilt 1.612
29. Louisville 1.608
31. Georgia 1.54
38. Texas A&M 1.32
43. Oklahoma 1.19
45. Texas 1.18
48. N Carolina 1.14
53, Indiana 0.96
73. LSU 0.66
88. S Carolina 0.44

Nice to see us continue to improve game over game. Playing at a Sweet 16/Elite 8 level, with some obvious upside potential beyond that.

Duke is starting to remind me of some Gonzaga teams of recent years. Not in how they play, but in the way they are rolling over a weak conference. They are good, but the last time they faced a team with a pulse we saw what happened.

UConn > Kentucky ????
If it weren’t a computer, my feelings would be hurt on that one

 

[Blueheart32]

Illinois has no pulse?

 

[mcwildcats04]

I'm selling Gonzaga, Arizona, and Kansas at those rankings.

 

[HoneyMoon155]

UConn > Kentucky ????
If it weren’t a computer, my feelings would be hurt on that one

I’m just wondering how Kansas is ahead of us

 

[ImTheVillageIdiot]

I’m just wondering how Kansas is ahead of us

Idk metrics have liked Kansas for most of the year. They have been dropping off lately though, can’t speak for this model

 

[Aike]

Illinois has no pulse?

Correct

 

[Aike]

I’m just wondering how Kansas is ahead of us

Biggest issue we have is that we haven’t been disruptive enough. We were better vs. Bama, generating 13 steals.

But think how often we’ve let teams look too comfortable and just run their stuff all night unimpeded.

That’s something this model is picking up on.

 

[HoneyMoon155]

Idk metrics have liked Kansas for most of the year. They have been dropping off lately though, can’t speak for this model

I guess I just don’t understand the formula that goes into the metrics, they’ve not looked good all year

 

[Aike]

I'm selling Gonzaga, Arizona, and Kansas at those rankings.

Might be right. Think it’s obligatory here to remind folks that this isn’t a ranking. I’m not trying to rank who has played the best. I’m trying to assess how a team plays, and whether that style is conducive to tournament success.

Those two things are typically highly correlated, but not necessarily the same thing. Hence the differences between this and other models.

 

[HoneyMoon155]

Biggest issue we have is that we haven’t been disruptive enough. We were better vs. Bama, generating 13 steals.

But think how often we’ve let teams look too comfortable and just run their stuff all night unimpeded.

That’s something this model is picking up on.

That’s true, seems like we do always just let our opponents play to their strengths

 

[HoneyMoon155]

Might be right. Think it’s obligatory here to remind folks that this isn’t a ranking. I’m not trying to rank who has played the best. I’m trying to assess how a team plays, and whether that style is conducive to tournament success.

Those two things are typically highly correlated, but not necessarily the same thing. Hence the differences between this and other models.

Can you run your model to just include games played during conference play?

 

[mcwildcats04]

Might be right. Think it’s obligatory here to remind folks that this isn’t a ranking. I’m not trying to rank who has played the best. I’m trying to assess how a team plays, and whether that style is conducive to tournament success.

Those two things are typically highly correlated, but not necessarily the same thing. Hence the differences between this and other models.

Yeah, fair enough. Does the model weight more towards recent games or more equally-weighted across the entire year?

 

[Aike]

I guess I just don’t understand the formula that goes into the metrics, they’ve not looked good all year

Except when they beat Michigan St. and Duke. And Iowa St. and took Houston to 2 overtimes.

They’ve looked really bad at times, but also shown that they can hang with, if not beat literally anyone on a given night.

 

[HoneyMoon155]

Except when they beat Michigan St. and Duke. And Iowa St. and took Houston to 2 overtimes.

They’ve looked really bad at times, but also shown that they can hang with, if not beat literally anyone on a given night.

I forgot they even played duke

 

[Aike]

Can you run your model to just include games played during conference play?

Not easily. It’s not that hard to run a date range, but unfortunately some non-conference games might get mixed in.

I do tend to run a last month of the season only model if I have time.

 

[Aike]

Yeah, fair enough. Does the model weight more towards recent games or more equally-weighted across the entire year?

It’s weighted equally, but I usually run a last month of the season model right before the tournament.

Sometimes I’ll fill out brackets splitting that 50/50 or 67/33 or whatever. Results are typically similar whichever way I do it.

The last month model is usually good for identifying outliers. Like Oregon and Auburn were both really hot in 2019. Florida Atlantic was hot two years ago.

BYU had gone cold last year.

 

[Blueheart32]

Correct

Come on man, that's just silly. They are ranked higher in kenpom than 6 teams that Ky has lost to, they beat Arkansas and Missouri as well as Oregon, UCLA, and Wisconsin. I'm not trying to sell them as some really good team, but to say no pulse is silly.

 

[Aike]

UConn > Kentucky ????
If it weren’t a computer, my feelings would be hurt on that one

Biggest thing keeping UConn afloat is that they block a ton of shots. Tends to be advantageous in the tournament. Also stil run nice offense with a high assist/turnover rate.

Obviously they have some issues.

 

[Aike]

Come on man, that's just silly. They are ranked higher in kenpom than 6 teams that Ky has lost to, they beat Arkansas and Missouri as well as Oregon, UCLA, and Wisconsin. I'm not trying to sell them as some really good team, but to say no pulse is silly.

Nobody has a sense of humor anymore. Is your sister their coach or something?

They are 47 in the model. About as much pulse as Oklahoma, Texas, or UNC. Certainly good enough to win a game or even a couple with the right draw.

Feel better?

 

[Blueheart32]

Nobody has a sense of humor anymore. Is your sister their coach or something?

They are 47 in the model. About as much pulse as Oklahoma, Texas, or UNC. Certainly good enough to win a game or even a couple with the right draw.

Feel better?

OH YES! YES! Thank you Thank you!!!