Busting Brackets
Fansided

2019 NCAA Tournament: 11 teams analytically who can win it all

LEXINGTON, KENTUCKY - FEBRUARY 16: Nick Richards #4 and Reid Travis #22 of the Kentucky Wildcats celebrate after the win against Tennessee Volunteers at Rupp Arena on February 16, 2019 in Lexington, Kentucky. (Photo by Andy Lyons/Getty Images)
LEXINGTON, KENTUCKY - FEBRUARY 16: Nick Richards #4 and Reid Travis #22 of the Kentucky Wildcats celebrate after the win against Tennessee Volunteers at Rupp Arena on February 16, 2019 in Lexington, Kentucky. (Photo by Andy Lyons/Getty Images) /
facebooktwitterreddit

The history of the NCAA Tournament can help predict who will cut down the nets in March, but let’s make sure we’re using history correctly.

As “Bracket Season” approaches, expect to see more and more analysis regarding which teams will perform best in March. One of the main topics of conversation will likely be which teams have the best chance to win the national championship. The following tidbit will almost undoubtedly be floated around: Going back to the 2001-02 season (as far as KenPom’s data base goes back), no team has won a national championship with a defensive efficiency ranking worse than 18th.

(All statistics used in this article were found via KenPom and are up to date through the games completed on Feb 18.).

This is accurate and definitely very interesting, but this information is often cited without the proper context. Each season’s rankings at KenPom.com reflect teams’ rankings AT SEASON’S END. Rankings headed into each season’s NCAA Tournament are provided as well, but since they’re provided in an excel file, they’re harder to work with and therefore often overlooked.

Here’s why this is important. There are several past national champions that entered the NCAA Tournament with defensive rankings worse than 18th, but played solid defense on their way to a title run and had their defensive rankings shoot up as a result. So based on historical precedent, national champions haven’t had to have the 18th or better defense at the beginning of the tournament. They just have had to be 18th or better by the END of the tournament. 

In case this isn’t making sense, here it is illustrated below. The first chart reflects the FINAL KenPom rankings for each season. 2008-09 North Carolina is the national champion with the worst final KenPom ranking (18th). 10/17 champions finished with a top 10 defense. On the offensive side of things, 16/17 champions finished with a top 20 offense, with 2010-11 UCONN as the exception. 14/17 champs finished with top ten offenses.

def1
def1 /

The next chart reflects the rankings headed into the tournament. Remember 2008-09 UNC, which had the worst final KenPom defensive ranking of any champion at 18th? They actually entered the tournament with the 37th best defense. Among the 17 champs, three entered the tournament with a defense ranked worse than 30th, and six entered the tournament with defenses ranked worse than the “magical 18th” number. On the offensive side of things, UCONN was again the outlier with an offense ranked 57th. But outside of that, 16/16 champs had top 25 offenses, and 10/16 had top 10 offenses.

def2
def2 /

So what can be taken from all this?

  1. The common tidbit regarding no champion having a defense worse than 18th is definitely very flawed, but it still makes a worthwhile point. National champions have had to play at a top 18 defense level to cut down the nets. But even if a hypothetical team hasn’t been a top 18 level defense all season, maybe there is reason to think they can get to this level in the tournament. Maybe their defense was especially good late in the regular season, and the rankings are thus underrating them a bit. Maybe they got a good defensive player back from injury? If there is no reason to anticipate defensive improvement, maybe a championship simply isn’t in the cards.
  2. Although a top-18 level defense might be too strict of a cutoff, it it’s still valuable to know that no champion has entered the tournament with a defense worse than 37th. This would theoretically cross off quite a few teams, including some fairly highly seeded ones.
  3. The “common tidbit” has focused on the defense of national champions, likely because 2013-14 UCONN’s low ranked offense shattered any “minimum offensive ranking” requirement for a championship. But it’s very possible that this was simply an outlier, and thus shouldn’t really be overly incorporated into future predictions. Outside of UCONN, 21st (2010-11 UCONN, funny enough) was the worst offense of any champion entering the tourney. Therefore, I’d proposed a new and fairly conservative tidbit that of course could be taken down by the rare outlier: A national champion will enter the tournament with an offensive ranking no worse than 21st and a defensive ranking no worse than 37th.

The below chart reflects this season’s teams that currently meet this criteria. Notable exceptions include Michigan, Tennessee, Texas Tech, Purdue, and Wisconsin. At 53rd, Purdue‘s defense kept it out, while the other four teams fell short due to their offenses.

def3
def3 /

History doesn’t always repeat itself, so it’s definitely possible a major outlier cuts down the nets and all of this was “worthless” for those filling out their brackets.

Ranking the last 25 NCAA Champions. dark. Next

The least we can do is make sure we’re using history the “right way” and hope for the best.