There was a lot of fire last night during the announcement of the bracket. The major beefs that folks had with the bracket were threefold:
- "Harvard over Hofstra?"
- "Denver's a five-seed? For real?"
- "Pennsylvania's a four-seed? Penn is good, but a four?"
These issues are too granular. The bigger issue -- the one that actually matters -- is this: Did the selection committee apply the selection and seeding criteria in a way that was consistent with prior iterations of the selection committee? The ability of the selection committee to construct a workable bracket is heavily influenced by the mandatory selection and seeding criteria pushed on the selection committee by the NCAA. As a result, the only question surrounding the veracity of the bracket is whether the committee applied the criteria in a way that doesn't deviate from prior bracket compositions.
Luckily, LaxPower published a guide to the NCAA Tournament selection process. In that study (covering years 2003 through 2012 (the last ten years of play)), LaxPower discovered that -- in its method -- all teams with a sum of its algorithm less than 21.00 received an invitation 100 percent of the time ("Locks"); teams with a sum between 22.00 and 33.00 received an invitation 88 percent of the time ("In"); teams with a sum between 34.00 and 51.00 lied in the bubble with 34.00-38.00 yielding a 79 percent invitation ("Bubble In") and 39.00-51.00 yielding a 21 percent invitation ("Bubble Out"). Teams with a greater than 51.00 sum received an invitation less than 1 percent of the time ("Out"). (This obviously correlates to the chance to get an at-large invitation, but its principles are broader than that (even though there are potential gaps).)
Using that study as the foundation to examining the selection committee's bracket this season, LaxPower published a prediction of the field prior to the release of the bracket. This is how LaxPower saw the bracket -- seedings and inclusions/exclusions -- shaking out:
HYPOTHETICAL SEED | TEAM | VALUE | AQ OR AT-LARGE | ACTUAL SEED |
1. | Duke | 3.50 | At-large | 1. |
2. | Syracuse | 4.50 | At-large | 2. |
3. | Loyola | 12.00 | AQ | 3. |
4. | Pennsylvania | 13.00 | AQ | 4. |
5. | Virginia | 15.00 | At-large | 8. |
6. | Notre Dame | 17.00 | AQ | 6. |
7. | North Carolina | 19.50 | At-large | Unseeded |
8. | Cornell | 21.00 | At-large | Unseeded |
9. | Maryland | 21.50 | At-large | 7. |
10. | Denver | 24.50 | AQ | 5. |
11. | Johns Hopkins | 29.00 | At-large | Unseeded |
12. | Harvard | 30.50 | At-large | Unseeded |
13. | Drexel | 33.00 | AQ | Unseeded |
14. | Albany | 37.50 | AQ | Unseeded |
15. | Bryant | 58.50 | AQ | Play-In // Unseeded |
16. | Air Force | 72.50 | AQ | Play-In // Unseeded |
17. | Siena | 74.50 | AQ | Play-In // Unseeded |
18. | Richmond | 139.00 | AQ | Play-In // Unseeded |
Some thoughts on this relative to the actual bracket:
- The selection committee -- for the most part -- acted in a similar fashion as it had in the past. This is a good thing: It means that the committee didn't deviate significantly from its prior behaviors, utilizing similar tools as in years past and composing a bracket that is rooted in the committee's extended history. The tent pole marks were -- in general -- the tent pole marks, and the bracket lined up well with what was predicted from the group. Again: It's the selection criteria that people get angry about, not necessarily how the committee applies the criteria. The committee does a good job of putting information out there regarding what they're looking at and what the benchmarks are, but there are issues with the actual benchmarks and criteria that the NCAA mandates that the selection committee utilize.
- The last predicted at-large -- Harvard -- was right on the money with the selection committee's call (and it was in-line with how the committee would have made that decision in the past). Hofstra, based on LaxPower's method was right in the belly of the "Bubble Out" section with a value of 42.00. The committee was on-point with its behavior building into the bracket's release.
- The most surprising decision in the bracket was the seeding of Denver. Jim Siedliski, the chair of the selection committee, has been very open about the committee's evaluation of the Pioneers (links to interviews with Siedliski are below), noting that Denver ended up getting comparable treatment to Loyola. With respect to how Denver was positioned relative to the committee's prior behavior, though, there was a notable deviation in expected and actual result: The Pioneers were a tournament team (squarely within the "In"), but were -- based on past committee actions -- targeted for an unseeded lifestyle, well below the team's actual five-seed. That's okay as blips happen (the existence of special cases is real, and Siedliski's comments imply that Denver was unique this season given the team's overall profile). But it doesn't mitigate the surprising nature of the Pioneers rising up the bracket despite having a profile that the committee has -- in the past -- sent to the unseeded side of the field. Whether Denver deserves a five-seed is a completely different conversation from one that asks whether the Pios' placement this year looks like an outlier relative to prior committee behavior.
- Outside of Denver, every other team in the bracket falls relatively near where they were expected to fall. Pennsylvania -- for all the fire the team took on selection night -- fell where a Pennsylvania team was supposed to fall based on how the criteria has been applied in the past. Virginia is slightly underseeded and that hurt them a bit with their draw (getting Hopkins instead of Harvard (assuming perfect bracket integrity (i.e., no flight rule, etc.))). North Carolina is underseeded, but the Tar Heels drew the opponent expected (except it will take place in Colorado instead of Chapel Hill, an important change that could carry impact). Hell, Maryland got the Cornell team they were expected to get. Otherwise, this shook out exactly as the selection committee has performed in the past (save for the high seeding of the Pioneers).
Overall, the selection committee stepped in almost the same footprints that were laid in the past. So, if you dislike the bracket, you dislike the selection criteria, not necessarily how the selection committee applied the widely-circulated criteria.
ADDITIONAL READING
DI Bracket Reaction: Selection Chair Siedliski, Coaches' Podcast -- Inside Lacrosse
Four Big Questions with NCAA Selection Chair Siedliski -- Lacrosse Magazine
College lacrosse analysis: Denver's seeding most curious part of Selection Sunday -- The Post-Standard