basketball-men-d1 flag

Daniel Wilco | NCAA.com | November 7, 2018

A bracket geek's take-it-to-the-bank 2019 March Madness predictions

Even in March, there are few guarantees you can make about the NCAA tournament with certainty. But, to honor the start of a new season, here are five March Madness predictions we’re (fairly) sure of:

A 16 seed won’t win again this year

The conversation around the 16 seed will never be the same after UMBC pulled off the historic upset of 1-seed Virginia last year. Yes, we now know it’s actually possible. But that game did not signal the start of a new trend. A 16 seed is just as likely to win in 2019 as it was in any of the past 34 years of 16 seeds. And that is very, very unlikely.

Even after UMBC won by 20 points, 1 seeds still hold the margin in the matchup by 24.4 points over the 136 games played. Only 15 of the past 136 games have been decided by fewer than 10 points. No 16-v-1 game has been decided by one possession since 1996. We could list a new insurmountable fact every day from now until Selection Sunday, but we know it won’t stop the fans. Over the past eight years, players in our Bracket Challenge Game have been picking 16 seeds to win their first-round game more and more, with a record 7.95 percent of our millions of brackets predicting one last year. That number will skyrocket this year. And they’ll all be wrong.

A blue team will play in the championship game

Of the 33 national championship games since 1985 (not counting Louisville’s win in 2013), only four did not have a blue team playing (1987, Indiana beat Syracuse; 2000, Michigan State beat Florida; 2002, Maryland beat Indiana; 2007, Florida beat Ohio State). What’s more, 22 of the 33 champions have been blue squads. Blue teams have won nearly 60 percent of their games since 1985 and holds a winning record against every color except orange.

Here is the breakdown of championships won by primary uniform color in the last 33 years:

Championships by primary color

No one’s bracket will stay perfect through the Sweet 16

It goes without saying that no one will pick a perfect bracket this year. We can never get enough of how ridiculous the odds are, so here’s a refresher from our post on the history of March Madness brackets:

Virtually all NCAA tournament brackets disregard the First Four and only pick games starting with the first round. Since there are 64 teams in those brackets, the most basic calculation is the number of possible outcomes for 63 games picked randomly. That would be 2 (the number of potential winners for each game) to the 63rd power (the number of games in the bracket). More simply, that's 2 times 2, 63 times, which is equal to roughly 9.2 quintillion. 

For reference, if you filled out 1 billion random brackets every single second for 100 straight years, you would still be 6 quintillion brackets shy of 9.2 quintillion.

But that only applies if every game is a coin flip. In practice, there’s a lot of information that usually goes into picking brackets. The most basic is seeding, which we discussed earlier. Since every team is seeded 1-16 in its region — with the highest-ranked team receiving a 1 seed, and the lowest a 16 — even someone who has no basketball knowledge at all can make a somewhat educated guess on which team is favored in each matchup. 

The late DePaul professor Jeff Bergen broke down the odds for someone making informed decisions for each game and came up with the very optimistic odds of 1 in 128 billion. Much better that 1 in 9 quintillion for sure, but still almost so low as to be negligible.


But we’ll make our claim about something that seems a bit more reasonable. No one will stay perfect through the Sweet 16. In order prove us wrong, a bracket would have to get 48 games correct in a row. If we did our calculation for a bracket that flipped a coin for each of those (again, this is not a perfect estimate), the odds would be 1 in 281,474,976,710,656. That’s 281 trillion. For perspective, that’s roughly the same number of seconds in nine million years. Good luck.

The best defensive team in the country will not win the title (especially if it’s Virginia)

While when they lost was surprising, the fact that Virginia didn’t win the national title last year was not. The Cavaliers were ranked as the best defensive team in the nation using the Defensive Simple Rating System, which ranks teams based on a combination of average point differential and strength of schedule. In the 34 years of the modern NCAA tournament, only one team ranked No. 1 in DSRS reached the Final Four, and none have ever won the championship. In fact, Virginia last year became the 10th team to finish ranked No. 1 in DSRS and lose in the first round. Of the 34 teams, 24 have either missed the NCAA tournament or lost before the first weekend.

NCAA TOURNAMENT RESULT COUNT
DNP 8
Lost in First Round 10
Lost in Second Round 7
Lost in Sweet 16 6
Lost in Elite Eight 3
Lost in Final Four 1

Not all of them have been seeded as high as the Cavaliers last year. Only three teams have finished No. 1 in DSRS and received a 1 seed. They were Virginia in 2014 (lost to 4-seed Michigan State, 61-59 in the Sweet 16), Virginia in 2016 (lost to 10-seed Syracuse, 68-62 in the Elite Eight), and Virginia in 2018 (lost to 16-seed UMBC, 74-54 in the first round). So while we’re very confident in saying that no team that finishes with the best DSRS will win the title, we’re even more sure if that team is Virginia.

One team will make its first-ever NCAA tournament

As of this season, there are 353 teams in Division I. Of those, 44 have never made the NCAA tournament. And there are some solid teams that are trending up among the 44. Two years ago, Northwestern was on that list before making its first NCAA tournament (and picking up a win). Last year, Lipscomb earned its first invitation. This year, it’ll probably be on one of the following four teams:

  • Bethune-Cookman
  • Grambling 
  • South Dakota
  • Hartford

For more detail, read why we think each one of those has a chance.