Well, you clicked on it, so I won't belabor it: if you like upsets, it's probably gonna be a bad year for upsets. It would be awesome if this turns out to be the stupidest thing I write all year, and that in about six weeks, we are talking about a Sweet Sixteen with a 14 seed in it and a Final Four with Utah State. I would love that, as it would restore some of the lack of sanity that reminds us of the NCAA Tournament at its very best.

But: I cannot tell you to expect these things, not based on the data. When I initially ran the numbers for this year a month back, there were three major takeaways:

  1. The return of Cinderella probably isn't happening in 2026.
  2. That being said, we probably won't see all four 1 seeds in the Final Four this April.
  3. Sadly, we're probably going to see our second straight year without a major upset...in the opening round.

Considering the two most correlated Tournaments were 2025 (you remember this one) and 2017 (1 seed vs. 1 seed final, no wins by 13-16 seeds), it all made sense. Now, with an additional month of data in tow, we have a bit more insight into what the 2026 NCAA Tournament is shaping up to be.

However, compared to last year at this time, it seems like people are pretty much in agreement that, yes, it's going to be top-heavy. Last year I had a whole big row about my annoyance with heads being buried in sand:

This news hasn’t stopped the usual hordes of people, some of whom I genuinely like, doing the Please Like My Sport thing. Run a search and you’ll find “it’s really wide open this year”, “the college basketball field is as wide open as it’s ever been”, and my favorite, “all 364 teams have weaknesses”.

This year, we have passed the denial stage in record time. Our guy Evan Miyakawa says the actual Final Four is on Saturday. In preparation for this article, I searched every combination of "NCAA Tournament," "college basketball", and "wide open" I can think of. Not a single post since early March 2025. Seemingly everyone has the same list of title contenders, which is just KenPom's top six plus Texas Tech. Even guys I normally would never run to for an interesting take on college basketball are saying things like "the Big 12 Tournament will be better than the opening weekend of the NCAA Tournament," and honestly...maybe? Probably not, but maybe.

This isn't me saying that the 2026 NCAA Tournament will suck. Far from it. Last year's Final Four produced three fantastic games, and the Sweet Sixteen had three games decided by three points or less. The 2017 Tournament had a 7 seed make the Final Four and produced the Florida/Wisconsin Sweet Sixteen game, one of the wildest three-hour experiences in recent Tournament history. Even 2019 and 2000, a pair of comparisons I'll get into, produced a 12/13 game (2019) and two 8 seeds in the Final Four (2000).

The good news: you're unlikely to have all four 1 seeds make the Final Four again and for the Elite Eight to be all 1 vs. 2/1 vs. 3 games. The bad news: you may have to suffer through a weak Round of 64 to get there.

Much of what I wrote about last month is still more or less true. The top 50 is still, on average, the strongest it's ever been. Teams ranked in the top 5, top 10, top 25, and top 50 are all at their strongest average efficiency margins ever recorded. The only group that isn't is the teams ranked 51-100, which...boy, I can't say that fills me with a ton of excitement for the opening Thursday and Friday!

But, hey, we've overcome this before. The distances between these groups would be more educational to you, though they're also not the most encouraging in modern history. Here, you'll see that no field on record has larger gaps from its top 5, top 10, and top 25 to the back half of the top 100 ever. If you thought 2025 was an extreme case of haves versus have nots, you should've waited a year to see worse.

For the discussions of the Round of 64, I do think it's educational to look at the only three years within three points of 2026. 2000, 2017, and 2025 had a combined zero wins by 13-16 seeds, and in 2000, the nightmare scenario happened: one singular upset, 11-seeded Pepperdine defeating Indiana. When 2017's big upset is a game none of you remember (12-seed MTSU over Minnesota in a game they were favored), well, there's your story.

Still, the next few comparisons are a little rosier. The other years with at least a 12-point gap or higher (2004 is awfully close, as you'll see) both featured a 13-seed winning. If it makes you feel better, there are actually some areas in which 2026 isn't the most extreme Tournament to date. Just probably not in the Round of 64. Considering the five closest comparisons had a combined two wins by 13-16 seeds and an average of three true upsets (almost entirely by 11 and 12 seeds), don't get your hopes up for a renewed Cinderella well past midnight.

So: you probably won't get the most chaotic Round of 64 in Tournament history. What about the Round of 32 and further on? There's some educational aspects to the charting if you play with it above. Notably, the gap from 1-5 to 26-50 is only the fifth-largest we've seen, more in line with years like 1998 and 2001. However, the years that offer the closest collective Z-score distancing are unsurprisingly 2000, 2015, and 2017. (1998 does finish fourth.)

That would indicate a fairly interesting Round of 32 lies ahead, at least for me. In those three years, the following happened:

  • 2000: Two 8-over-1 upsets, as well as two 10-over-2s.
  • 2015: An 8-over-1 and two 7-over-2s.
  • 2017: An 8-over-1, two 7-over-2s, and an 11-over-3.

If you add 1998, which had three Round of 32 upsets, that's a Round of 32 with three or four true upsets happening. Considering we had one in the 2025 Tournament and none after the Round of 32, that's promising to me. I think an active and entertaining Round of 32 is a good thing for the Tournament, especially if the Round of 64 is trending towards being a bit of a dud.

Beyond that, the closest comparisons for your later rounds are all already listed. The commonalities between your top five years - 2000, 2015, 2017, 2019, and 2025 - are listed below.

  • 11 of the 20 Final Four spots were taken by 1 seeds.
  • Four of the five years had at least one spot taken by a 5 seed or lower.
  • Four of the five years saw at least three of the four 1 seeds make the Elite Eight.
  • But! Four of the five years saw a Certified Weird Elite Eight Game: an 8 vs. 6 and 8 vs. 7 in 2000, 4 vs. 7 in 2015, 4 vs. 7 in 2017, and a 2 vs. 5 (okay, stretching the definition here) in 2019.

All this being said, it's notable to me that the two least-similar years by far are 2006 and 2023, two of the five seasons with a 3 seed or higher champion. 2003 (Syracuse) and 2014 (UConn) also rank pretty low on the Z-score distance list. Of the four least upset-filled years in the last three decades (2007, 2009, 2019, and 2025), all are among the top eight years in strength of correlation. It's nice that 2021 is in the mix, too, but I would argue it's more likely that year was a one-off thanks to...y'know, pandemics and such. If you really want to narrow it down to a top three, 2017, 2019, and 2025 - three years where the 13-16 seeds went a combined 1-47 and put a total of 7 1 seeds in the Final Four - are the three strongest correlations by some distance.

Still, while the second-most boring Tournament (by percentage of games ending in upsets) is indeed the top comparison, it's definitely notable that three of the next four had three or more Round of 32 upsets, and all had at least one upset from the Sweet Sixteen onwards. Perhaps patience provides chaos.

What would all this look like in the future NCAA Tournament? Let's plot it out.

On the opening day of the NCAA Tournament, the overwhelming response from the populace was a heavy sigh. Just one upset happened - an 11-over-6, featuring a team from the First Four - and most games featuring 14, 15, and 16 seeds were blowouts. A 13 seed did threaten a 4 for about 35 minutes, but it wasn't meant to be.

Friday didn't provide much more. Two additional upsets happened, including the beloved 12-over-5, but despite a tremendous challenge from a 14 seed in a 5-point loss to a 3, this was the second Tournament in a row without a major upset that grabbed the public's attention. Many thinkpieces were written about the continued death of Cinderella, how the Round of 64 is dying out, and how bad NIL and the transfer portal are for the future of college basketball.

Then the Round of 32 happened. Not only did it end with an 11 seed in the Sweet Sixteen, it ended with the loss of a 1 seed (the first Round of 32 loss for a 1 seed since 2023!) and the departure of a pair of 2 seeds, one of whom lost in overtime. Suddenly, the NCAA Tournament was fixed, thanks to more upsets happening on Saturday and Sunday than on Thursday and Friday. People became extremely excited for the second weekend of the Tournament again, because the thing we had lost was back once more: chance. All you need is chance.

As it wound down in the Sweet Sixteen and Elite Eight, two key trends emerged: a pair of 1 seeds surviving to Indianapolis, but a pair of surprises, somewhere amongst the 3-7 lines, making it as well. After all, the sport is at its best when it mixes actual good teams winning games along with some chaos. While the games in Indianapolis would prove to be a little less star-studded than a year ago, those in attendance were secretly thankful that a little chaos had been re-introduced to the party after a year of utmost calm.

Much like its contemporaries, the 2026 NCAA Tournament would produce a 1 seed as its champion, the third year in a row with such and the seventh of the last eight Tournaments to do so. The formula, after a year off, was back: 75% correct and agreeable results to 25% confusion.

Maybe that ratio is a little higher or lower when this actually happens and I do/don't look like an idiot, but this is my best guess based on what I'm seeing in the data: boring Round of 64, enthralling Round of 32, and a Wacky Final Four Team once it's all said and done. So! Congrats to Champion 1 Seed, Runner-Up 1 Seed, and Surprise 5-7 Seed in Indianapolis in about six weeks. I'm sure you know who you are, because it will be April and not February.