Tourney Mythbusting: Does defense win the NCAA championship?

(This blog appeared originally in early December. I’ve updated it to reflect the current state of college hoops.)

You’ve heard the old adage: defense wins championships. Does that apply to the NCAA tourney? If you look at the raw points scored and allowed data, you would probably conclude that offense matters more in the dance. Take a look at this chart, which shows the average points scored and allowed by the tourney field in each succeeding round of advancing teams:

The blue line tracks scoring offense and the red tracks defense. Notice that, with each succeeding round of the dance, the teams get significantly higher scoring. Final Four teams are 4.8 points more prolific than the overall field, and champions have the highest-powered offenses. On the other hand, the points that advancing teams allow doesn’t change much from round to round. In fact, champions give up more points than any other round of advancers.

Okay, that means offense wins tourneys, right? Well…not exactly. By now, most everyone is familiar with Ken Pomeroy’s possession-based statistics. (Check out www.kenpom.com if you aren’t.) Think of Ken’s stats as the WAR of college hoops—only more grounded and definitive. They take the bias of playing tempo out of the equation and concentrate solely on the number of points per possession that a team scores or allows. After all, you could look at a team like Princeton, which scores 64.0 points a game and gives up just 57.6 and conclude that they were lock-down defenders…but offensively challenged. But the fact is, the Tigers like to play at a glacial pace (343h slowest out of 347 D-I teams). They’re actually the 65th most efficient offensive team in the land…and just the 138th best defenders.

So what if we redid our analysis of tourney advancers, using the teams’ rankings within the field in terms of offensive and defensive efficiency? This gives a much clearer view of the value of offense and defense. Take a look:

2013_offdef

The offense and defense efficiency rankings of tourney advancers are pretty close to each other all the way to the finals. (Remember: a higher number here is worse.) Defense holds a slight edge over offense in four of the first five rounds. Only Sweet 16 advancers have a better ranking in offense than defense. But the number that really jumps out here is the relative rankings of the champions. Tourney winners are much higher ranked offensively than defensively—a gap of five positions. In fact, the gradually flattening curve for defense suggests that, while you need a solid defense to advance in the dance, defense can only take you so far. On the other hand, you have to have an elite offense to cut down the nets.

Let’s look at it another way. I evaluated the tourney performance of one through six seeds above and below their median offensive and defensive efficiency values. The numbers show that better defensive teams defy seed expectations to a higher degree—but better offensive teams win more hardware. First, the defensive breakdown:

The 108 1-6 seeds that have better than median defensive efficiency numbers own a +.166 PASE—about a sixth of a game per dance above expectations. And they’ve won seven of the nine tourneys since 2004 (the first year that I began tracking Pomeroy stats).

Now take a look at the comparison of high seeds above and below their offensive efficiency median:

Above-median offensive teams also overachieve in the tourney (+.072), but at a lower rate than their defensive counterparts. On the other hand, eight of the nine tourney champs have owned better-than-median offensive efficiency numbers. Only UConn in 2011 fell below the median (and they were on wrong side of the defensive numbers too).

I think there’s a reasonable case to be made that offense trumps defense in the NCAA tourney—at least when it comes to crowning the champion. The numbers say that you need to play solid defense. But that will only take you so far. When it comes to the big prize, you better be proficient at scoring.

Of course, the best teams excel at both ends of the court. Just look at the 21 teams that have been ranked among the top ten in both offensive and defensive efficiency heading into the dance. Nearly half of them (10) have reached the Final Four and five have cut down the nets. More impressively, these 21 squads own a whopping +.537 PASE.

Right now, only one team is among the top ten in both offense and defense. That would be Florida. There are a handful of teams in the top 20 of both offense and defense–Indiana, Louisville, Pittsburgh, Ohio State and Syracuse–but they overachieve at a less dramatic rate (+.236 PASE)

Keep your eyes on these efficiency numbers when it comes time to fill out your bracket. They may help point you to the squads that are more likely to make a deep run in the dance.

This entry was posted in Tourney Trends. Bookmark the permalink.

6 Responses to Tourney Mythbusting: Does defense win the NCAA championship?

  1. Jbessa says:

    Agree. What about adding a 3rd dimension to the analysis on tempo. So for two good offensive teams with similar offensive efficiency, would you rather have a fast or slow tempo? Ive been wondering about this. My sense is faster tempo is better because you can exploit your offensive efficiency more fully with more possessions. So using similar analysis of kenpom tempo does tempo get faster with advancing teams in the tourney?

    • ptiernan says:

      I stopped tracking tempo in my tourney database–and that was a mistake. I did, however, do a blog last year, where I detected a correlation between slow-paced teams and KenPom’s “bad luck” stat. Bad luck is basically the variance between what KenPom efficiency numbers say a team’s results should be and what they actually are. I found it odd that so many slow-paced teams–like Wisconsin and BYU and Pitt (this was last year, I think)–had high numbers for bad luck. I think it’s because, as these teams limit possessions, their scores are naturally lower…and they’re more prone to being victimized. The other thing is think is true about KenPom data: it doesn’t factor in WHEN a team is efficient. My contention is that slow-paced teams might struggle to maintain their efficiency into the late, frantic stages of close games…when they either need to move fast and score, or need to hold up under desperate pressure from teams trying to come back.

  2. Tom says:

    Princeton Badgers? You might be mixing your glacial teams.

    • ptiernan says:

      I did indeed mix my glacial squads. I originally had Wisconsin in there, thinking their offensive numbers were better than they really were. I switched to Princeton and missed the mascot name change. Nice catch.

  3. Jake says:

    I wonder if there’s a way to blend in “defensively converted offense” into these statistics? I would venture that a large part of Louisville’s offense is “converted” from opportunistic transition defense (as opposed to traditional half court), and that without that factor they would not score as well in the offensive categories.

    • ptiernan says:

      No question that teams like VCU and Louisville have offenses that thrive on defense. We saw how Temple was able to short-circuit the Ram attack just by getting the ball over half court with consistency. I don’t know a way to calculate that though.

Leave a Reply

Your email address will not be published. Required fields are marked *

* Copy This Password *

* Type Or Paste Password Here *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>