The oddity of the expected-yet-unexpected 2025 NCAA Tournament

How do we find a balance between the appropriate outcome, yet feeling a little bored by how we got there?

An odd sensation came about during the final Elite Eight game for me yesterday: impatience. Not over a monitor review or some excessive whistling or even teams taking shots they shouldn’t. Not even over a result of a game that wasn’t that mission critical to my moral fibers. No, I felt impatience because for the third time in four games, it was basically over before it started.

That might be the feeling I remember most from the 2025 NCAA Tournament, for better or for worse. In Eamonn Brennan’s chat yesterday I expressed what a weird feeling on the whole this NCAA Tournament was producing for me. On one hand, I am satisfied because for the entire season, the foursome of Duke, Houston, Florida, and Auburn were the four best teams. They’ve been in the top four at KenPom in some order every day since February 12, and it only took that long because Florida started out 26th. (Houston, Duke, and Auburn were Ken’s preseason top three in that order.)

The number of tournaments where the top four pre-Tournament teams on our nation’s leading metrics site have even made the Elite Eight, much less all gotten to the Final Four, is very few. Prior to 2025, only 2007, 2008, and 2019 saw all of the top four making the Elite Eight, with only 2008 seeing all of them in the Final Four. My pre-Tournament analysis post had 2007 and 2019 as two of the three most similar Tournaments to 2025, though it didn’t quite spot 2008 as among the top.

On that end, a Final Four with the four best teams in the sport is a thoroughly satisfying outcome. At an average of +9.4 AdjEM, the gap between the 1 seeds and the 2-4 seeds is the highest I have on record across 25 years of data. The only Tournaments close were 2021 (1 vs. 1 title game), 2015 (1 vs. 1 title game), and 2002 (shrug). This was a historically dominant year at the top of the sport, the likes of which we are all wondering if we’ll ever see again or if it’s the new normal.

But: it was a little boring. The data, from Neil Paine below, backs it up.

I took Neil’s data and turned it into a four-quad graphic similar to that of the work Micah McCurdy does for the NHL. As such:

Perhaps I’ve gone the wrong way of looking at it; I assume for the average person 2014 was more memorable and weirder than 2015. But simply put, 2015 felt right in a way 2014 didn’t. A title game featuring a 7 seed playing an 8 is, well, a bit much. 1990’s Tournament is an underrated wheel of chaos, too: Elite Eight games of 4 vs. 10, 4 vs. 6, and 1 vs. 11. 2015, for what it’s worth, did reward the best teams for the most part and had a healthy amount of close games, including an instant classic in the Final Four between Wisconsin and Kentucky.

Even last year was pretty nutty. A lot of blowouts belied a strange collection of events: the two best teams in the sport making the Final Four, along with 4 and 11 seeds. The first 1 vs. 2 game in the Elite Eight since 2019, but also 14 opening weekend wins by lower-seeded teams. (There’s been 11 in the entirety of the 2025 Tournament.) To get here after going through 2022 (top left quadrant) and 2023 (top right) is a little bit of a swing we’ve got to get used to.

Still, the number of close games this year is pretty small. Connor Hope noted that just 25% of the 64 games thus far have finished within two possessions through the Elite Eight. Removing the two First Four games from that dataset, the 23% rate of close games is the smallest we’ve seen since 20% (12 of 60) in 1999. Even last year, which had a lot of blowouts, saw 17 of 60 (28.3%) end up being close.

Some of this can be assigned to unfortunate late-game variance. For instance, not counting Duke over Arizona (a 100-93 win for Duke) or Arkansas over Kansas (a 79-72 win with a late-game Arkansas comeback) makes things look worse, while counting Michigan State losing 70-64 in the Elite Eight in a game that was largely around a 10-point Auburn lead for the entirety of the second half does seem a little fishy. Plus, winners outshot losers 37.4%-29.3% from three in this Tournament on average. I’m sure Houston appreciates that their opponents combined to shoot 24% from deep.

By pure margin of victory, this isn’t even one of the five biggest blowout tournaments, per Neil. 1993, 1996, 1999, 2008, and 2024 were all more boring in terms of the average game being interesting, and the stat I mentioned earlier about 1999 having even fewer close games does ring true. (Yet it had so many upsets! If only I’d been not five years old. I wonder what the takes must have been like.) And again, this was the year with the largest gap I have on record from the top four to everyone else.

I also think it’s largely fine if, once every 20 or so years, we have all the 1 seeds make the Final Four. Plus, this wasn’t the expected outcome; prior to the Tournament beginning, the KenPom odds for the four 1 seeds gave them a 5% chance of all making it to San Antonio. In an average year, there’s about a 2-2.5% chance of all the 1 seeds getting there. Considering prior to 2025 we’d had 39 Tournaments and only one of them had all the 1 seeds get to the Final Four, a 1-in-20 hit rate isn’t that offensive. People upset about this should consider watching Ligue 1, for instance, to see a sports structure that really is uncompetitive.

Still, I wouldn’t deny anyone - let alone myself - the ability to complain that the 2025 NCAA Tournament has been pretty dull. I would put it alongside 2008 as probably the least-exciting I have seen, at least prior to the Final Four. This is fine. They can’t all be winners, and it’s good to have a Dullsville, USA Tournament every few years that rewards the absolute best of the best, especially when we have Tournaments every few years that are a complete crapshoot and end up with a VCU/Butler playoff to play on Monday night. (Or, you know, San Diego State/FAU.)

So, yeah, it hasn’t been great. On one hand, this was technically the most likely individual outcome and the one that is the most appropriate based on season-long outcomes. On the other, the path to get there was unfortunately low on entertainment, and like 2008, this year really could have used a Western Kentucky in the Sweet Sixteen or a San Diego/Siena causing Round of 64 chaos. (Or, yes, a Davidson. I guess Arkansas was our Davidson this year? Oof.)

I guess I ultimately land somewhere in the middle: this is neither the greatest nor worst NCAA Tournament in my lifetime, just like 2023 was neither the greatest nor worst. It has its upsides and its downsides. Now, hopefully, the Greatest Final Four in Modern History will deliver like the 1993 Final Four did, after the actual dullest first two weekends in history:

Two instant classics and one 78-68 game that was a 68-65 UNC lead with three minutes to go. Those were nice rewards for an NCAA Tournament low on surprises and close games before. Could history repeat itself 32 years later? You’ve got an ACC team playing a Big 12 team, and an SEC team playing…okay, not another SEC team, but bear with me. It’s not that nuts. I guess this means my pick is Duke over Florida after Florida defeats Auburn in overtime, yet calls a timeout they do not have two nights later. I don’t think we’d put that in Dullsville, USA, would we?