Ya know, the irony is not lost on me that we as bracketologists obsess over the top-16 bracket preview show as much as we do, trying to forecast every minute detail and split every hair, then the grand reveal finally comes, and… it’s already outdated a few hours later.
That’s the consequence of holding the preview show on a Saturday, especially one as massive as the Saturday we just had, which may have very well featured not one but two games between Final Four teams.
As such, today’s Bauertology bracket already looks quite a bit different than what the selection committee showed us on Saturday—though the lessons learned from Saturday have been applied just as much as the new results.
For my money, these were my biggest takeaways from the top-16 show that I’ll be keeping in mind for my process going forward:
The committee loves when you beat top teams, especially away from home. This isn’t really anything new, but it’s a fact that has perhaps been somewhat lost in a world where metrics seem to dominate the majority of discussion. Houston was a better team metrically than both Iowa State and UConn entering Saturday, yet the Cougars found themselves at #6 overall behind those two. The reason? Top-end wins: The Cyclones crushed a 2 seed in Purdue 81-58 in West Lafayette, while the Huskies have neutral-court victories over Illinois and Florida and a 61-56 victory at Allen Fieldhouse on the team sheet. Houston’s best road/neutral win in comparison is at BYU. Same logic applies to Texas Tech as a 4 seed, surpassing a number of teams they were metrically worse than, as a result of those neutral and road victories over Duke and Arizona, respectively.
Metrics are still very important, but they’re not binding. While teams still mostly landed within the range of where their metrics said they should be, there were a couple key deviations. Iowa State didn’t have a single top-4 metric on Saturday morning, yet the Cyclones landed on the 1 line as a consequence of having numerous victories over other protected seeds, while still being within a reasonable metric range. Alabama’s résumé metric average would have landed the Crimson Tide as a 3 seed naturally, but they were among the committee’s first two teams on the outside; perhaps more reserved efficiency numbers and the fact that Bama is 1-6 against NET top-17 teams had something to do with it. So we should continue to use metrics as a baseline, just don’t fall into the habit of ordering teams solely on their metrics when other important factors are at play.
Head-to-head results seem to matter a bit more than before. Another one of the biggest reasons that Iowa State landed on the 1 line ahead of Houston, as noted by committee chair Keith Gill, is because of the Cyclones’ victory over the Cougars last Monday night. This same reasoning appears to have been applied elsewhere: Illinois at #7 overall ahead of #8 Purdue, and Vanderbilt present while Alabama was absent. Still have to be careful about when and where to apply this focus on head-to-head results, otherwise you could have argued Arkansas to be included over Vandy or Texas Tech. But it does seem that if teams are close enough to each other on the overall seed list, then it’s a consideration that will be accounted for.
Teams were not punished for key injuries… yet. The one team I got wrong out of the 16 shown, Texas Tech, was on the outside of my top-16 guess merely because I figured that a season-ending injury to a critical piece like JT Toppin would knock the Red Raiders down in the selection committee’s estimation quite a bit. And yet, the committee maintained respect for Tech’s huge wins over Duke, Arizona, and Houston, despite the fact that those wins came with Toppin on the floor, only moving TTU down from a 3 seed to a 4 upon hearing the injury news, as Gill noted. That said, I still think there’s potential for Tech to tumble further if they look bad without him in any of their final four games. But given that we hadn’t seen what the Red Raiders looked like without Toppin prior to the top-16 reveal, it’s more than fair to leave them largely where they were until we get more data. (And if Saturday’s destruction of Kansas State is any indication, then Texas Tech may be completely fine.)
And most importantly: by and large, do the same thing I’ve been doing! I got 15 out of 16 teams correct, my only miss being Alabama in place of Texas Tech—an easily explainable miss—and I had nine teams down to the exact overall seed, too. Basically, I just need to remember to keep emphasizing big wins away from home, use metrics as a guideline without being fully dependent on them, and take note of important head-to-head results when it comes to putting the teams in 1 to 68 order. Otherwise, my process should essentially be the exact same! Always a big relief when I’m able to draw this kind of conclusion.
So, how do these lessons apply to today’s bracket? You’ll see that I’ve moved Duke past Michigan up to #1 overall as a result of Saturday’s game in DC, despite the fact that the Wolverines remain ahead of the Blue Devils in NET and five of the six team-sheet metrics. If there were ever a time to invoke the head-to-head clause, it would be here!
You’ll also see that I’ve moved Tennessee up to my final 4 seed ahead of a whole hodgepodge of SEC teams that also includes Alabama, Arkansas, and Vanderbilt—one that is really hard to sort out since they’ve all basically beaten each other. The Volunteers are the only one of that group to nab a road win against the others, having two of them in beating both the Crimson Tide in Tuscaloosa and the Commodores in Nashville. Pair that with a win in Vegas over Houston, and I think I can easily justify choosing Tennessee as the protected seed out of that bunch.
Finally, I’m making a move that other bracketologists might find risky by kicking Indiana out of the field for the time being. The Hoosiers possess metrics that would typically garner inclusion, but at just 4-10 in Quad 1/2, there just isn’t enough meat on the bones for me. Their only weighty road/neutral win is at UCLA, i.e., another bubble team, while a metrically worse team like USC at least went and won at Wisconsin, while also beating the Hoosiers head-to-head in early February. Meanwhile, other bubble-team inclusions that lack those crucial road victories, like Santa Clara and New Mexico, make up for it by being both metrically sound and having good marks in Q1/2 (8-5 and 9-5, respectively). And I do think that there’s a general sentiment of leniency for non-power conference teams who don’t get the same sheer number of opportunities to accrue huge road/neutral wins as P5 teams do, so that logic is applies to my thinking as well.
Now for the rest of the bracket! As you’re probably aware of at this point, my automatic bids are chosen by which team leads their conference in my résumé metric BRCT. HOWEVER, I am adding a stipulation that if it’s really close, I can use my discretion to make a more sensible choice. Each of the Southland, Sun Belt, and Ohio Valley automatic bids are within a point of each other in BRCT right now. When it’s that tight, close enough that minor shifts in the NET rankings could very well be enough to flip a team into the conference lead, I’m going to be a bit more open-minded on which team to take as my projected automatic bid.
That’s it for Monday’s bracket; hope to see you on Wednesday for the next edition of Bauer’s Bubble Watch! By this time next week, we’ll already be in March. How the time flies.
