The Left Side of the Bracket Is Packed

The Sweet Sixteen is set, and the difficulty of paths…varies. Here are how the sixteen teams line up from best to worst, with their KenPom ranking and adjusted efficiency margin (in points per 100 possessions) in parentheses:

1. Gonzaga (1st, +37.04)
2. Baylor (2nd, +29.73)
3. Michigan (3rd, +29.29)
4. Houston (4th, +29.28)
5. USC (6th, +26.82)
6. Alabama (8th, +25.61)
7. Loyola (9th, +25.56)
8. Villanova (11th, +23.54)
9. Florida State (13th, +23.29)
10. Arkansas (15th, +22.09)
11. Oregon (17th, +21.66)
12. Creighton (19th, +21.01)
13. UCLA (24th, +20.27)
14. Syracuse (36th, +17.60)
15. Oregon State (50th, +14.83)
16. Oral Roberts (128th, +3.87)

And here are those teams grouped by region:

West

1. Gonzaga (1st, +37.04)
5. USC (6th, +26.82)
11. Oregon (17th, +21.66)
12. Creighton (19th, +21.01)

East

3. Michigan (3rd, +29.29)
6. Alabama (8th, +25.61)
9. Florida State (13th, +23.29)
13. UCLA (24th, +20.27)

South

2. Baylor (2nd, +29.73)
8. Villanova (11th, +23.54)
10. Arkansas (15th, +22.09)
16. Oral Roberts (128th, +3.87)

Midwest

4. Houston (4th, +29.28)
7. Loyola (9th, +25.56)
14. Syracuse (36th, +17.60)
15. Oregon State (50th, +14.83)

***

It’s not an even distribution of ability, and that, of course, impacts the championship picture. But how exactly is it impacted?

A quick way to answer this would be to take the average adjusted efficiency margin (adjEM) of each region. But that may be misleading. After all, Oral Roberts’s presence in the South doesn’t exactly help Baylor or Villanova, given the Golden Eagles are the Sweet Sixteen’s biggest underdog, pegged at just 11% likely to beat Arkansas.

A better way is this:

Take the likelihood a given team would beat a roughly median Sweet Sixteen opponent, followed by a roughly median Elite Eight opponent, followed by a roughly median Final Four opponent, and then compare that probability of said team making the championship to a simulated probability given the real bracket.

So that’s what we’ll do.

For the median Sweet Sixteen opponent, we’ll choose Florida State, on the mean side of the split halfway through the remaining field. For the median Elite Eight opponent, we’ll choose Loyola, knowing the likelihood of upsets on Saturday or Sunday. For the median Final Four opponent, we’ll go with Michigan, in the middle of a tight pack of non-Gonzaga regional favorites adjEM-wise.

This is a ballpark median path. It isn’t perfect. But it’s going to be close to what that number would be, and the important thing is to take something at least roughly resembling the median. Here’s how each team would do against it:

TeamChampionship Game Likelihood (Median Path)
Gonzaga45%
Baylor22%
Michigan21%
Houston21%
USC15%
Alabama13%
Loyola12%
Villanova9%
Florida State9%
Arkansas7%
Oregon6%
Creighton6%
UCLA5%
Syracuse3%
Oregon State1%
Oral Roberts0%

Now, here’s how each team did in 1,000 quick simulations today of the real bracket (that isn’t a ton of simulations, so this too is rough, but fine for our purposes):

TeamChampionship Game Likelihood (Real Path)
Gonzaga49%
Houston31%
Baylor29%
Michigan17%
Loyola17%
USC11%
Alabama10%
Villanova10%
Arkansas10%
Florida State5%
Oregon3%
UCLA3%
Creighton2%
Syracuse2%
Oregon State1%
Oral Roberts0%

And here’s the difference, expressed as a percentage of how much harder (positive values) or easier (negative values) a team’s real path is than the median path:

TeamPath Difficulty (Median vs. Real)
Creighton67%
Oregon50%
Florida State44%
UCLA40%
Syracuse33%
USC27%
Alabama23%
Michigan19%
Oregon State0%
Oral Roberts0%
Gonzaga-9%
Villanova-11%
Baylor-32%
Loyola-42%
Arkansas-43%
Houston-48%

And finally, here’s how each team’s real path differs from how difficult their path should be, based on their position on the selection committee’s overall seed list (higher values constitute a more difficult path than seeding would imply, lower constitute an easier path):

TeamReal Path DifficultySeeded Path DifficultyDifference
Creighton198
Florida State3107
Michigan8146
Alabama7136
Gonzaga11165
Oregon264
Baylor13152
USC671
Syracuse54-1
UCLA43-1
Houston1612-4
Arkansas1511-4
Villanova128-4
Oregon State92-7
Loyola145-9
Oral Roberts101-9

A few takeaways:

  • Of course, this is not through some fault of the selection committee. The committee made its mistakes, but this is more just how things have played out.
  • While Loyola and Oral Roberts have the easier paths relative to expectations, a lot of that is because their paths have been the most difficult so far, with them knocking off two of the three top-2-seeds already eliminated.
  • Four of the eight toughest paths relative to expectations belong to teams in the West Region, and that’s despite Iowa’s elimination.
  • Three of the four toughest paths relative to expectations belong to teams in the East Region.
  • The six toughest paths relative to expectations are all on the left side of the bracket.
  • Baylor, despite a friendly Elite Eight draw (Arkansas, most likely), has a tough Sweet Sixteen matchup, dealing with a Villanova team somehow managing to survive without Collin Gillespie. That shows up here.
  • Contrarily, Villanova has a surprisingly easy path, as they would’ve expected a much more difficult Elite Eight opponent based on their seeding, and that’s enough to outweigh the bad draw in the Sweet Sixteen.
  • Gonzaga does not have it easy. Unless you take into account the fact they don’t have to play Gonzaga, in which case they’re blessed beyond measure.
The Barking Crow's resident numbers man. Was asked to do NIT Bracketology in 2018 and never looked back. Fields inquiries on Twitter: @joestunardi.
Posts created 3026

Leave a Reply

Your email address will not be published. Required fields are marked *

Related Posts

Begin typing your search term above and press enter to search. Press ESC to cancel.