Using possession-based data to forecast the next champion

For the last four or five years, I’ve been using a set of statistics to forecast the teams most likely to cut down the nets in March. Some of those stats, like coaching experience and conference affiliation, can be applied before a game is even played. Others, like scoring average, margin and strength of schedule, are only relevant as the season plays out—particularly the conference schedule, which we’re just about to start.

Later today, once the AP rankings come out, I’ll post my standard “champ check” that uses eight stats to identify teams with the characteristics of the past 12 champions. None of those stats, however, are with possession-based, or tempo-free, numbers that I’ve been using in other analyses. I’ve been hesitant to use them because I was waiting until I had ten years of samples. But I got a recent comment asking if I’d look into it. So I’ve decided to provide an alternative champion forecast leveraging this in-depth game-play data.

I have Ken Pomeroy’s possession-based data, as of Selection Sunday, for the last nine dances. The ultimate champion in those tourneys had no lower than a 115.1 offensive efficiency value (points per 100 possessions) and no higher than a 92.2 defensive efficiency value. UConn set both these upper and lower limits in 2011. The 2009 North Carolina squad was equally soft defensively, but their OE was 123.8.

Let’s assume that this year’s champion will fall into this range, with and OE no lower than 115.1 and a DE no higher than 92.2. Based on Pomeroy’s calculations, which teams are the best candidates to win the tourney? Take a look at this chart; the teams in all caps meet both the offense and defense criteria:


There are nine teams are efficient enough on both ends of the court to be champions. That’s more teams than will meet the standard criteria I use later today. For one thing, the standard stats kick out Mid-majors like Gonzaga and Creighton. But at least two teams that will appear on my standard list do not appear here. That would be Louisville and Syracuse. Both teams meet the “points scored” criteria I track through the basic champ check, but they aren’t actually efficient offensive squads.

So by this analysis, the teams most likely to be crowned the 2013 tourney champ are: Indiana, Florida, Duke, Kansas, Pitt, Ohio State, Michigan, Gonzaga and Creighton.

Interestingly, all these teams met the defense criteria, while 11 fell short of the offense limit. I don’t have the data to confirm it, but I’m guessing that offensive efficiency is down overall this year while defense is up. If that’s the case, it might not be useful to use the raw efficiency numbers as our limits. We may be better off filtering the teams by their OE and DE ranks.

So let’s do the analysis again. The last nine champions have owned an offensive efficiency rank no lower than 17 and a defense rank no lower than 25. Here’s what the same teams we just studied rank for offense and defense possession-based efficiency (I’ve crossed out the teams that fall outside our limits):

  • TEAM             OE       DE
  • Indiana           1          10
  • Louisville        17       1
  • Florida            3          7
  • Duke               2          13
  • Kansas            7          6
  • Pittsburgh      5          19
  • Ohio State      12       15
  • Syracuse         25       4
  • Michigan        4          37
  • Minnesota      15       20
  • Arizona           16       18
  • Kentucky        21       12
  • VCU                 24       9
  • Gonzaga          11       35
  • Creighton       9          42
  • Wisconsin       22       23
  • Oklahoma St  77       3
  • Michigan St    44       14
  • Cincinnati       74       5
  • Wichita St       60       16
  • LIMIT             17       25

By this analysis, three teams that didn’t make the grade by raw numbers are now on the list: Louisville, Minnesota and Arizona. Conversely, three teams fall off this list: Michigan, Gonzaga and Creighton.

There are six teams that satisfied both the possession-based methods we employed—Indiana, Florida, Duke, Kansas, Pitt and Ohio State. Interestingly, three of these teams didn’t meet the criteria for our most recent basic champ check. Florida was too low scoring, Pitt wasn’t in the AP top 20 and the Buckeyes played too soft a schedule.

So would you put a bet down right now that Indiana, Duke or Kansas will be the 2013 champion? It’s too early to get that bold. For one thing, OSU’s strength of schedule will get better and Pitt ought to climb in the AP rankings, maybe even later today. Plus, there’s the whole “small sample size” issue with possession-based data. Still, it will be interesting to track both the basic and these advanced credentials for potential champions. Come Selection Sunday, fewer than a handful of teams will satisfy both of them.

This entry was posted in Champ Credentials. Bookmark the permalink.

One Response to Using possession-based data to forecast the next champion

  1. Matt says:

    Thanks for doing this. I appreciate you taking the time to look at it after my comment.

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>