Saturday, January 14, 2012
Teams Revealing Who They Really Are
"You are what your record says you are."
That's a quote from former NFL coach Bill Parcells some years back in response to a question about whether his team was really better than its record indicates. The basic point — it doesn't matter how "good" you are. It only matters whether you won or lost when you played the game.
In college basketball, things aren't so simple. Actually, records can be quite deceiving. Not only is there a much wider disparity in who those records are built against (strength of schedule), but "talent" seems to be far less consistent in college hoops. Sometimes a team can look unbeatable. Other times it can look like straight dog poop. With some teams, you just never know.
One thing that does seem to be a great equalizer, though, is team efficiency.
For the uninitiated, efficiency is basically a tempo-neutral measurement of how a team performs on a per-possession basis. There are plenty of stat nerds on the internet who can explain it better than I can, so look it up if you need more information. My interest isn't so much the process of finding the stat, but rather how to use the stat for my own personal well-being, in this case my NCAA tournament bracket.
(What? NCAA tournament bracket? But we barely just began our two month bitch-fest about bubbles and seeding. How can we even think about the bracket?)
(Because you can use the efficiency ratings to figure out which teams to watch. It still comes down to watching the games and figuring out matchups. The stats just help you figure out where to look).
Now that we've got that settled, now for some proof that this stuff actually matters:
* In the five years I've been tracking this, no NCAA tournament team with an efficiency split over 30 or more has lost in the first round. (Efficiency split is a team's offensive efficiency minus its defensive efficiency.)
* Of the 26 teams who have entered the tournament over the past five years with an efficiency split of 30 or higher, 22 have made at least the Sweet 16, 17 have made at least the Elite Eight, 11 have made the Final Four, and four have won the title (with two others losing in the title game).
* On the flip side, of the 78 teams that have come into the tournament over the past five years with an efficiency split under 10, 71 of them lost their first game and only one has made it out of the second round. That was VCU, which somehow found its way into the Final Four with an 8.8 split. The next lowest to make it out of the first weekend was last year's Butler team that made it to the championship game with a split of 15.
(I'm considering last year an aberration. We'll see if this year repeats or not.)
So anyway, I hope those few facts convince you of the power of the efficiency split. (Note: It's actually adjusted efficiency from KenPom.com.) Now it's time to see which of this year's teams fall into which bucket.
* As of games played on Tuesday, only four teams are at the 30+ level: Ohio State (34.37), Syracuse (32.34), Kentucky (31.16), and Wisconsin (30.48).
I'd say that makes the Buckeyes, Orange, and Wildcats pretty good bets for an Elite Eight run or better, but you can't trust Wisconsin no matter what the stats say. Twice in the past five tournaments the Badgers have gone into the tournament with a 30+ efficiency split. Two other times they were over 25. They have yet to make it out of the Sweet 16, and didn't even get there twice. Until Bo Ryan shows his style can actually get the job done in the Tourney setting, I'm viewing their potential with a heavy dose of skepticism.
* I said you can't trust Wisconsin. Same goes with Georgetown. In 2007, their efficiency split was 35+ and they made the Final Four. In 2008, it was 30+ and they lost in the second round. The past two years, it was 25+ and 18+ and they got bounced in round 1. This year the 11th-ranked Hoyas are at just past 21. So even if they end up a 3- or 4-seed, beware of some serious upset potential.
* Baylor is on course for a possible one seed, but their 23.83 efficiency split would be by far the lowest of any top seed in the past five years. In fact, only one two seed has had a lower split, and that was the 2010 Villanova team that got upset in the second round.
Working in Baylor's favor perhaps is last year's championship Connecticut team had a 22.85 split as a three seed. I'm not sure Baylor has anybody approaching Kemba Walker and his ability to take over a game, though. (Actually, I'm sure they don't. Nobody does.)
* I mentioned Kentucky as a good bet. I know they've had a few spots of inconsistency this year, but not only do they have the second-highest split in the country, John Calipari has never failed to make the Elite Eight as a one seed (four tries, two Final Fours, and one gut-wrenching loss in the 2008 title game).
(It also pays to watch coach trends. You can trust in Sean Miller to get more out of his teams. You can trust Kevin Stallings to get less.)
* Speaking of the 2008 title game, the team that beat Calipari's Memphis squad that year was Bill Self's Kansas Jayhawks, who have failed to make it past the Sweet 16 their last three trips as a non-one seed. So while Self's team is sitting with a very respectable 29.88 split, they have "underachievers" stamped all over their tournament future.
* Beware the team with a middling seed (4-6) but a high efficiency split (25+). Of the 21 teams meeting that criteria over the past five years, one made it to the Final Four (last year's Kentucky team — told you it was a weird year). Of the other 20 teams, five made the Sweet 16, 12 lost in the second round and five fell to first-round upsets by double-digit seeds.
That's 17 of 21 failing to make it out of the first weekend, so watch out Florida (25.32), Gonzaga (22.48), and Alabama (22.34).
* Of the 100 teams to enter the tournament with a sub-15 split over the past five years, only VCU last year made it to the Sweet 16. Teams who should be concerned about that: Stanford (15.80), Louisville (15.55), Illinois (15.43), Dayton (14.13), Harvard (13.97), Mississippi State (13.27), San Diego State (10.55), and Arkansas (10.09).
So when you see somebody pumping up Harvard or San Diego State as sleepers, remember they are going for the 1%. Personally, I'm going with the 99%. But maybe that's just because I like money. (Mitt Romney for President!)
So there you go. These are by no means final projections of what will happen in two-plus months. Players will get better or worse, healthy or injured, quit or otherwise somehow change the fortunes of their teams. What we see now in the first few days of conference play is not what we will necessarily see come championship Week.
But if you're looking for a place to start scouting, you could do a lot worse than following the map efficiency ratings provide.
(Just kidding about the Romney thing.)
(I think.)