Post by bluepenquin on Mar 13, 2017 10:05:23 GMT -5
Since this article references Men's college basketball - and the selections occurred yesterday - I was curious how seeding and the bracket for college basketball compares to volleyball.
One big difference is basketball will seed all 68 teams (there are 4 more teams in basketball) while volleyball only seeds the top 16. Geography has an impact on the final seedings - particularly in determining the regional matchups. An example - Baylor came in as the #12 seed, but got moved up to the #9 line because of the geography rules in place to put more teams closer to the regional host. Still - the impact of geography is greater in volleyball.
Below is the top 16 seeds for this year's college basketball tournament followed by their RPI ranking and their ranking in Pomeroy ratings. Pomeory ratings for basketball are essentially the equivalent of the Pablo rating in volleyball. It is a measure of future success - predictive/projection as opposed to RPI.
1. Villanova - 1, 2
2. Kansas - 3, 10
3. North Carolina - 5, 3
4. Gonzaga - 8, 1
5. Kentucky - 4, 4
6. Arizona - 2, 20
7. Duke - 6, 12
8. Louisville - 7, 6
9. Oregon - 9, 16
10. Florida State - 14, 19
11. UCLA - 16, 18
12. Baylor - 11, 13
13. Butler - 15, 26
14. Florida - 10, 9
15. West Virginia - 24, 5
16. Purdue - 19, 15
Clearly, RPI had way more impact on the seedings than Pomeroy (if Pomeroy had any impact). At a high level - I am seeing the exact same complaints with how basketball did their seedings as we have seen with volleyball. There are the issues like Kansas - a team that won a large number of last minute close games - which shows up well in RPI, but not as well in predictive ratings like Poemeroy.
Here are some seeding snubs per Pomeroy:
17. Virginia - 18, 7
21. SMU - 13, 11
25. St. Mary's (Calif.) - 17, 14
38. Wichita State - 32, 8
Wichita State is the Creighton/Hawaii in volleyball. They play in a crappy conference that absolutely kills their RPI - but Pomeroy loves them as the 8th best team overall.
Here are the RPI snubs - teams that were in the top 16 in RPI, but didn't get a top 16 seed:
21. SMU - 13, 11
22. Cincinnati - 12, 22
To me, this looks like something we would see in volleyball. Teams outside the power conferences with a high RPI (like a Western Kentucky in VB) that is going to lack the other stuff to get a seed.
Here are the top 3 RPI teams to not get a bid:
Illinois State 33
Texas Arlington 45
Monmouth 49
Again, something we would see in volleyball - teams from not very good conferences.
One of the biggest 'improvements' that people in volleyball want for their bracket is to seed 1 through 64. Basketball does this - but you may be surprised how this probably doesn't solve the problem. If this was volleyball - we would have the Kentucky subregional which would include #4 Kentucky and #8 Wichita State (in terms of Pomeroy ranking). Or the Florida subregional with #7 Virginia and #9 Florida. And then at the other end we would have the Florida State subregional with #19 Florida State, #40 Xavier, and #45 Maryland. Or the Butler subregional with #26 Butler, #33 Minnesota, #48 Middle Tennessee State.
One big difference is basketball will seed all 68 teams (there are 4 more teams in basketball) while volleyball only seeds the top 16. Geography has an impact on the final seedings - particularly in determining the regional matchups. An example - Baylor came in as the #12 seed, but got moved up to the #9 line because of the geography rules in place to put more teams closer to the regional host. Still - the impact of geography is greater in volleyball.
Below is the top 16 seeds for this year's college basketball tournament followed by their RPI ranking and their ranking in Pomeroy ratings. Pomeory ratings for basketball are essentially the equivalent of the Pablo rating in volleyball. It is a measure of future success - predictive/projection as opposed to RPI.
1. Villanova - 1, 2
2. Kansas - 3, 10
3. North Carolina - 5, 3
4. Gonzaga - 8, 1
5. Kentucky - 4, 4
6. Arizona - 2, 20
7. Duke - 6, 12
8. Louisville - 7, 6
9. Oregon - 9, 16
10. Florida State - 14, 19
11. UCLA - 16, 18
12. Baylor - 11, 13
13. Butler - 15, 26
14. Florida - 10, 9
15. West Virginia - 24, 5
16. Purdue - 19, 15
Clearly, RPI had way more impact on the seedings than Pomeroy (if Pomeroy had any impact). At a high level - I am seeing the exact same complaints with how basketball did their seedings as we have seen with volleyball. There are the issues like Kansas - a team that won a large number of last minute close games - which shows up well in RPI, but not as well in predictive ratings like Poemeroy.
Here are some seeding snubs per Pomeroy:
17. Virginia - 18, 7
21. SMU - 13, 11
25. St. Mary's (Calif.) - 17, 14
38. Wichita State - 32, 8
Wichita State is the Creighton/Hawaii in volleyball. They play in a crappy conference that absolutely kills their RPI - but Pomeroy loves them as the 8th best team overall.
Here are the RPI snubs - teams that were in the top 16 in RPI, but didn't get a top 16 seed:
21. SMU - 13, 11
22. Cincinnati - 12, 22
To me, this looks like something we would see in volleyball. Teams outside the power conferences with a high RPI (like a Western Kentucky in VB) that is going to lack the other stuff to get a seed.
Here are the top 3 RPI teams to not get a bid:
Illinois State 33
Texas Arlington 45
Monmouth 49
Again, something we would see in volleyball - teams from not very good conferences.
One of the biggest 'improvements' that people in volleyball want for their bracket is to seed 1 through 64. Basketball does this - but you may be surprised how this probably doesn't solve the problem. If this was volleyball - we would have the Kentucky subregional which would include #4 Kentucky and #8 Wichita State (in terms of Pomeroy ranking). Or the Florida subregional with #7 Virginia and #9 Florida. And then at the other end we would have the Florida State subregional with #19 Florida State, #40 Xavier, and #45 Maryland. Or the Butler subregional with #26 Butler, #33 Minnesota, #48 Middle Tennessee State.