clock menu more-arrow no yes mobile

Filed under:

Predicting a Champion with Kenpom

How does Villanova compare to past champions?

NCAA Basketball: Big East Conference Tournament Championship-Villanova vs Providence Noah K. Murray-USA TODAY Sports

About a month ago I wrote an article here analyzing some of the top teams’ adjusted-offensive efficiency and adjusted-defensive efficiency. Based on the analysis from the beginning of February, Villanova had some work ahead of them if they wanted to fit the trend of past champions. Now that the season has finished, let’s take a look at how the past champions were ranked entering the big dance.

Note: 2014 UConn was not included in the graph (80 AdjO, 11 AdjD) for the sake of compactness. Duke in 2010 and Kansas in 2008 were not included because accurate data could not be acquired from the WayBack Internet Archive. For explanations on how rankings are calculated please refer to Kenpom.

As you can see, the largest clump hovers in the top 5 AdjO and top 30 AdjD. Every champion had a top 15 AdjO entering the tournament, and except for Duke in 2015, a top 30 AdjD. Due to the limited data, a trendline would be useless for analyzing a trend so we are stuck simply eyeballing the graph.

Now let’s throw some of the top teams from this year onto the same graph.

I know what you’re saying, “Where’s Virginia?” While Virginia currently has the number one ranked defense, their 21st ranked offensive doesn’t fit on this graph. Obviously, there is always a chance of them becoming an outlier and winning it all, but based on the current trend you wouldn’t expect them to. Which raises the question, “How does the number one ranked defensive team do in the tournament?

How do #1 defenses fair the tournament?

2017: Virginia (37th offense) - 5 seed, lost in second round

2016: Wichita State (81st offense) - 11 seed, lost in second round

2015: Virginia (27th offense) - 2 seed, lost in second round

2014: Arizona (35th offense) - 1 seed, lost in Elite 8

2013: Louisville (15th offense) - 1 seed, won Championship **vacated**

2012: Ohio State (10th offense) - 2 seed, lost in Final 4

2011: Texas (21st offense) - 4 seed, lost in second round

It is not uncommon for a team to have a number one defense and a solid offense as well. The best defensive team in the county has been bounced in the second round for three years in a row (although one was an 11 seed). Prior to that, they had some real success, reaching the Elite 8 in 3 of the 4 seasons and winning a championship - that was later vacated.

Based on Kenpom rankings and the past, which teams are most likely to win it all?

Rankings of Past Champions

Team AdjO AdjD
Team AdjO AdjD
'17 UNC 4 25
'16 Villanova 11 7
'15 Duke 3 57
'13 Louisville 15 1
'12 Kentucky 2 5
'11 UConn 12 27
'09 UNC 1 28
'07 Florida 3 11
CHAMP AVG. 6.38 20.13
CHAMP STD. DEVIATION 5.39 18.29

Current AdjO - AdjD

Team AdjO AdjD
Team AdjO AdjD
Virginia 21 1
Villanova 1 22
Kansas 6 46
Xavier 7 59
UNC 4 34
Purdue 2 29
Duke 3 7
Cincinatti 54 2
Michigan St. 9 9
Gonzaga 12 15
WVU 14 39
Auburn 16 44
Michigan 29 4
Texas Tech 46 3
Tennessee 40 5

Please remember the following is based purely on past Kenpom rankings and trends. For the sake of this analysis I am purely basing a team’s chances on how they fit in with past champions.

Favorites to win -Villanova, Duke, Purdue, MSU

According to Kenpom, Duke and Michigan St. have the most well-rounded teams in the country. Duke has made a tremendous turnaround since my last article, moving from the 69th best defense to the 7th ranked defense. Michigan St. has the 9th ranked offense and defense, but did not exactly have the toughest schedule with some quality wins.

Purdue and Villanova also fit into the trend very well, both similar to UNC’s 2009 squad. ‘Nova’s offense has consistently been dominant all season, and their defense has certainly been improving in crunch time. Purdue is very similar to ‘Nova in their dominant offense and strong defense, but them being in the same region could make a potential Elite 8 matchup very interesting.

Don’t count them out - Virginia, UNC, Kansas, Gonzaga, WVU, Michigan

None of these teams winning would be as unexpected as UConn in 2014, but they are just on the outskirts of the dominant trend of the past. It seems wrong putting the overall number one seed team with only two losses on the season in this category but I’m doing this based purely on numbers. Most people wouldn’t raise an eyebrow at the number one overall seed winning the tournament, but based on rankings it would be slightly unexpected. Gonzaga actually has well-rounded team as well but that strength of schedule really raises questions about how they will fare against other dominant teams. UNC and Kansas have similar rankings and are just on the outskirts of that top 4 - top 30 trend, making them not completely out of left field. If anything, either UNC or Kansas winning would actually expand that clumping and make it more accurate. WVU is just hanging on the outskirts of that second clump (Villanova, Louisville, UConn) but they’re outside the top 30 in AdjD, off the trend of years past. Michigan is fairly similar to Virginia’s rankings, with the 5th best AdjD and the 29th best offense.

Probably not - Xavier, Cincinnati, Auburn, Texas Tech, Tennessee

Texas Tech, Tennessee, and Michigan are ranked 3rd, 4th, and 5th respectively in AdjD. However, their 46th, 40th, and 29th ranked AdjO are widely off the trend of past champions. Xavier is similar to Duke in 2015 with one of the best offenses (7th) but not good on the defensive side (59th). Cincinnati has the 2nd best defense in the country, but their 54th ranked offense is not good enough based on years past. Auburn barely squeaks in on the offensive side but their 44th ranked defense is a little too far off to consider them true contender purely based on rankings.

As we all know, March Madness doesn’t always follow the laws of logic and rankings. A team I didn’t even talk about might win it all. But hey, what’s March Madness without some “expert” analysis?