Brackets are finally out. we now got some updating to do
note from on his OP this year
Welcome to the 7th year (technically 8th year, but Covid canceled the 2020 NCAA tournament) of my research into pre-tournament stats that can help determine which teams most closely resemble prior Ken Pom era champions. For people who are familiar with my previous year’s posts, it’s great to talk NCAA Tournament basketball again! For those who are new, get ready for a deep dive into which teams are the championship contenders for this year’s tournament.
Prior to last years tournament I collected 12 stat categories, scraped various websites for data, used index/matching in excel, and spent way more time than I’d like to admit to come up with my infamous spreadsheet. Last year I removed the “Frontcourt player averaging over 12 ppg” and “4 players averaging double figures” stat categories to get to 10 stat categories. This year I am removing Opponent Two PT % < 46% and Opponent Free Throw Rate < 31%, and Top 90 in offensive rebounding %.
That means for 2024 I am down to seven stat categories. My reasoning for this is over the years I have identified which stats have the strongest correlation to all champions in the Ken Pom era (since 2002) and they are:
@DavistonTiger
here is the picking guide from last year
going through the seeds on upset watch
1seeds:
-very low sample size but slow temp0 has been a theme
this year. Uconn (325th) and Houston (349th) slowest tempo 1seeds
2seeds:
- adjO or adjD <50 are 6-4, rest are 41-2
this year. iowa st adjO 51
3seeds:
-14 seeds with adjO and adjD within 40 places of each other = 5-9. kennesaw st almost pulled it off last year
this year. MoreheadSt (vs Illinois) and Oakland (vs Kentucky)
4seeds:
-adjO or adjD outside top 40, + 13seed with adjO or adjD inside top 50. 13seed is 6-5 in that scenario, everyone else 7-38. last year furman over virginia and louisiana over tennessee (tennesse won by 3)
this year. Alabama (121 adjD) and Kansas (52 adjO). Charleston is close (adjO 53) and Samford (adj67). bama is the closest here, charleston has really good guards too...
5seeds:
-adjO or adjD outside top 60 are 5-15, everyone else is 28-8 + 12seeds inside top 60 overall are 14-11 and the rest are 8-23
this year. SDSU (adjO 69th) but UAB is 109 overall. next closest is wisconsin (adjD 54) v JMU (69 overall) and Zags (adjO 50) v McNeese (81 overall). i like the JMU > Wisconsin pick. 12 seeds are pretty weak this year cause of bid stealers though
6seeds:
-11seed is better or within 5 spots overall of 6seed (12-5 record)
-11seed is 25spots overall or more behind the 6seed (ALWAYS 1 per year) look for 6seeds with adjO or adjD outside top 60 (3-7 record)
-coin flips on others
last year every single matchup was in category 2. iowa st was the one who lost
this year:
TTU vs NC St (2nd category), USCe v Oregon (coin flip), Clemson v UNM (1st category), BYU v Duquense (2nd category)
- i like nc st over ttu, duquense should not be an 11 seed but its in that true upset category that always happens each year
7seed:
-just take higher kenpom/trank. they are 39-16. better team wins 3/4 times
@Kaiiu
note from on his OP this year
Welcome to the 7th year (technically 8th year, but Covid canceled the 2020 NCAA tournament) of my research into pre-tournament stats that can help determine which teams most closely resemble prior Ken Pom era champions. For people who are familiar with my previous year’s posts, it’s great to talk NCAA Tournament basketball again! For those who are new, get ready for a deep dive into which teams are the championship contenders for this year’s tournament.
Prior to last years tournament I collected 12 stat categories, scraped various websites for data, used index/matching in excel, and spent way more time than I’d like to admit to come up with my infamous spreadsheet. Last year I removed the “Frontcourt player averaging over 12 ppg” and “4 players averaging double figures” stat categories to get to 10 stat categories. This year I am removing Opponent Two PT % < 46% and Opponent Free Throw Rate < 31%, and Top 90 in offensive rebounding %.
That means for 2024 I am down to seven stat categories. My reasoning for this is over the years I have identified which stats have the strongest correlation to all champions in the Ken Pom era (since 2002) and they are:
- 4 seed or better
- Ken Pom ranking in the top 25
- Top 25 Ken Pom adjusted offense
- Top 40 Ken Pom adjusted defense
- 3 PT % > 33%
- Win regular season title (RS) or win conference tournament (CT)
- Have a head coach who has been to the elite 8 before
@DavistonTiger
here is the picking guide from last year
going through the seeds on upset watch
1seeds:
-very low sample size but slow temp0 has been a theme
this year. Uconn (325th) and Houston (349th) slowest tempo 1seeds
2seeds:
- adjO or adjD <50 are 6-4, rest are 41-2
this year. iowa st adjO 51
3seeds:
-14 seeds with adjO and adjD within 40 places of each other = 5-9. kennesaw st almost pulled it off last year
this year. MoreheadSt (vs Illinois) and Oakland (vs Kentucky)
4seeds:
-adjO or adjD outside top 40, + 13seed with adjO or adjD inside top 50. 13seed is 6-5 in that scenario, everyone else 7-38. last year furman over virginia and louisiana over tennessee (tennesse won by 3)
this year. Alabama (121 adjD) and Kansas (52 adjO). Charleston is close (adjO 53) and Samford (adj67). bama is the closest here, charleston has really good guards too...
5seeds:
-adjO or adjD outside top 60 are 5-15, everyone else is 28-8 + 12seeds inside top 60 overall are 14-11 and the rest are 8-23
this year. SDSU (adjO 69th) but UAB is 109 overall. next closest is wisconsin (adjD 54) v JMU (69 overall) and Zags (adjO 50) v McNeese (81 overall). i like the JMU > Wisconsin pick. 12 seeds are pretty weak this year cause of bid stealers though
6seeds:
-11seed is better or within 5 spots overall of 6seed (12-5 record)
-11seed is 25spots overall or more behind the 6seed (ALWAYS 1 per year) look for 6seeds with adjO or adjD outside top 60 (3-7 record)
-coin flips on others
last year every single matchup was in category 2. iowa st was the one who lost
this year:
TTU vs NC St (2nd category), USCe v Oregon (coin flip), Clemson v UNM (1st category), BYU v Duquense (2nd category)
- i like nc st over ttu, duquense should not be an 11 seed but its in that true upset category that always happens each year
7seed:
-just take higher kenpom/trank. they are 39-16. better team wins 3/4 times
@Kaiiu
