Network News

X My Profile
View More Activity
On Twitter: TerpsInsider and PostSports  |  Facebook  |  E-mail alerts: Redskins and Sports  |  RSS

Efficiency margin, a little-used stat, says Terps near top of the ACC

As Maryland ventures through the final two weeks of the regular season, it will do so as one of the most efficient teams (per possession) in the ACC. We've discussed on many occasions this season how potent the Terrapins have been on offense this year, how crisp their execution has been most of the time and how that portends well for the down the line. But in terms of quantifying Maryland's efficiency level, this forum had relied solely on observations and stat lines.

The folks at Basketball Prospectus, however, have used a decades-old, yet seldom-used, formula to determine precisely how efficient Maryland has been this season. Using a measuring stick first used by Dean Smith during his days roaming the sidelines at North Carolina, Basketball Prospectus has calculated that the Terrapins possess an efficiency margin of plus-0.11, which ranks second in the ACC behind Duke (plus-0.16).

What, exactly, constitutes a team's efficiency margin? According to John Gasaway, a writer for Basketball Prospectus, it's a calculation that measures a team's points per possession against the points per possession tallied by its opponents.

"A college basketball game is going to have something between 50 possessions if it's really slow, or 75, 80 on the other hand if it's really fast," Gasaway said. "The only way to make sense of any of the stat totals in there is if you know which of those extremes the game was played at."

For instance, Maryland is averaging 68.6 possessions per game in ACC play this season (the sixth-highest average among ACC teams) and is scoring 1.08 points per possession. The Terrapins' opponents, meantime, are scoring 0.97 points per possession. Simple subtraction (points per possession minus opponents' points per possession) gives Maryland a plus-0.11 efficiency margin.

"Of course, it all depends on the quality of your opposition, but speaking of major-conference teams, if they are playing conference games against opponents in their own league, generally speaking, if you score one point or better per possession, that's good," Gasaway said in a telephone interview. "If you score 1.1 point per possession, that's great. And if you score 1.2, that's simply stratospheric, out of this world, historic, fantastic. That would be a general rule of thumb."

In case you're curious, here are the efficiency margins of each ACC team:

1) Duke -- plus-0.16
2) Maryland -- plus-0.11
3) Clemson -- plus-0.05
4) Virginia Tech -- plus-0.03
5) Florida State -- plus-0.02
6) Wake Forest -- 0.00
7) Georgia Tech -- minus-0.01
8) Virginia -- minus-0.03
9) North Carolina -- minus-0.08
10) Boston College -- minus-0.08
11) NC State -- minus-0.09
12) Miami -- minus-0.10

And here are the points per possession averages of each ACC team:

1) Duke -- 1.11
2) Maryland -- 1.08
3) Virginia Tech -- 1.00
4) Wake Forest -- 1.00
5) Boston College -- 1.00
6) Miami -- 1.00
7) Florida State -- 0.99
8) Clemson -- 0.98
9) Georgia Tech -- 0.98
10) Virginia -- 0.97
11) NC Stae -- 0.97
12) North Carolina -- 0.96

How far has North Carolina fallen this season? Well, last year the Tar Heels led the ACC in efficiency margin (plus-0.15) and in points per possession (1.16). North Carolina then went on to win the national title. This season, UNC is at or near the bottom of the league in both categories. Yikes.

So what do all these numbers mean? Well, Gasaway says they can be used to help predict how a team will fare in the NCAA tournament. He said Maryland, for instance, will be ranked too low if it receives the No. 7 seed many prognosticators currently are giving them. According to Gasaway, the Terrapins are comparable to Purdue (plus-0.11 efficiency margin) and Syracuse (plus-0.11).

But are the efficiency margins of teams from different conferences comparable, given that these calculations are based purely on how teams fare against foes from only their specific conference?

"It is tricky," Gasaway said. "You're dealing with teams that are playing two entirely different sets of opponents. But what I would offer based on a couple of years of working with this stuff is that generally speaking, and, you know, with the case of the Pac-10 this year or the SEC last year, they need an asterisk by them. They constitute exceptions.

"But generally speaking, the differences in quality of play between these top six major conferences are really not going to be very great. We tend to invest a lot of discursive and polemic energy into questions like is the ACC better than the Pac-10? Or is the Big 12 better than the ACC? And it's fun to do. I'm not above doing that myself, but if you look at the wide sweep of college basketball, you can fit those top six conferences qualitatively into about an inch. Then the rest of the conferences below them, that takes in more of one or two or even three feet of qualitative space, so the difference between teams in different conferences, you know, their relative levels of performances is going to usually be a larger difference than the quality of the opposition. But it's something you definitely have to be careful about."

Last season, Maryland owned a minus-0.07 efficiency margin entering the postseason and then proceeded to defeat California (plus-0.02) in the first round of the NCAA tournament. The Terrapins fell to Memphis (plus-0.27) in the second round, though the Tigers' rating surely was inflated by its far inferior competition in Conference USA. For a look at the efficiency margins of other teams at the conclusion of the regular season last year, click here.

As for the ACC this season, Gasaway said he is not as down on the conference as are many other observers, primarily because of the performance of the teams at the bottom of the standings.

"Well, the ACC is down in the sense that they don't have that critical mass of, say, two or even three great teams that are big and scary and that everybody points to," Gasaway said. "Ironically, by some measures, the ACC, however, is actually the strongest conference in the country for the exact opposite reason. It's also true that they don't have any really bad, struggling teams.

"The teams at the bottom of the conference like Miami or N.C. State, whoever you want to pick, are clearly superior to comparable bottom-of-the-conference teams, such as DePaul or Nebraska or Indiana. And that's not a part of the conference that people usually look at, but when you're talking about a season's worth of conference games and then trying to come up with a number for the quality of the competition, when you've got zero really bad teams and you never play a really bad team, that means your strength of schedule is actually good."

By Steve Yanda  |  February 24, 2010; 10:04 AM ET
Categories:  Men's basketball  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   StumbleUpon   Technorati   Google Buzz   Previous: Padgett seeing earlier (if not more) minutes
Next: Terps finally back on somewhat normal schedule


As Archie Bunker once famously said, "Don't confuse me with statistics, man, I want the FACTS!"

Posted by: Rational_Man | February 24, 2010 10:34 AM | Report abuse

I like this gassaway dude.

Posted by: jpfterps | February 24, 2010 11:26 AM | Report abuse

These stat freaks just don't seem to get it.
It's the score differential that matters. The efficiency is a similar result. No coach is going to look at that and teach from it. Here is what matters: Defensive stops, rebounding differential, turnover differential, shooting percentage. You can teach defense, rebounding, ball control and shooting.

Posted by: petecard | February 24, 2010 11:32 AM | Report abuse

Interesting article for stat geeks. Funny that HGR picked up that quote comparing md to purdue/syracuse even befor this entry came up.

Posted by: JoeMama11 | February 24, 2010 11:42 AM | Report abuse

Steve Yanda's column, as usual, is spot-on. The use of numeric criteria to grade teams sharpens arguments and helps to force some objectivity.

There are at least two 'rpi' indices and the Pomeroy index. Neither the RealTime rpi nor the other one seem to make clear their criteria. Pomeroy does make his criteria explicit.

The RealTime rankings of some ACC teams today are: #2 - Duke; 21 - Wake; 31 - GT; #34 - MD; 35 - Clem; 40 - FSU; 45 - VT.

Pomeroy's take is - wait for it - #1 - Duke; #12 - MD; 18 - FSU; 19 - Clem; 21 - GT; 37 - WF.

I've followed the two foregoing indices this season and their predictions have been nearly identical and very good, even though their rankings seem to differ greatly. But the difference between their rankings is actually small when examined over the 340-odd teams in Division I.

It's interesting and telling that Pomeroy ranks the ACC as the strongest conference, but we already knew that, didn't we?

Posted by: AncientTerp | February 24, 2010 12:13 PM | Report abuse

"Efficiency margin, a little-used stat, says Terps near top of the ACC"

W-L record, a heavily used stat, also says Terps near top of the ACC.

Posted by: elowe2 | February 24, 2010 12:40 PM | Report abuse

I would hardly call tempo-free stats "seldom-used". Maybe by the average joe or the average joe journalist/commentator. But I assure you that coaches, basketball analysts who have been paying attention, and most importantly, the NCAA Tourney Selection Committee uses them heavily. Which is good for us. Remember how shocked everyone was when the Terps got a high seed in the 2004 Tourney after the miraculous ACC tournament run? Well, if you looked at the Ken Pomeroy rankings we were a much better team than our 7-9 ACC record showed. Same goes for the 2007 team, which also got a higher seed than most expected.

And petecard, I'm not sure what your point was about "stat freaks" that "just don't get it"...but after having read the columns and blogs from Gasaway, Pomeroy, and the like over the years, I would argue that they have a great deal of credibility on what matters and doesn't matter to winning basketball games...more so, dare I say, than you do. They are reasonable analysts who don't just hide behind the numbers, and are constantly looking for ways to improve their various formulas.

Posted by: Jeremy76 | February 24, 2010 12:56 PM | Report abuse

ACC to me is still clearly a top 2 or 3 conference this season. But of course, the popularly held belief is you only judge a conference by the teams at the top. The truth is, that's really only one part of it. You look at the teams in the middle, and the teams at the bottom as well, relative to other conferences. From top to bottom, there is no question the ACC is still one of the top conferences even in what many call a "down" year.

We don't have any 8-18 Depauls in the ACC. We don't have any Rutgers teams that lost NINE straight games by double figures. What we have is a conference in which every single team is at least .500 overall, the only conference in America that can say that.

Arguably the worst team in our conference this year won the national championship last year. They beat Michigan State this year, but got crushed by everyone in the ACC schedule.

Arguably the second worst team in our conference Boston College was a solid 10-5 heading into ACC play, with wins over a Big Ten school and a Big East school, but they too go crushed by everyone in the ACC.

But for some people, they think just because UNC isn't good, the ACC must be terrible.

Posted by: Barno1 | February 24, 2010 1:19 PM | Report abuse


I am not saying at some level stats are not useful. But people get so carried away with how to parse performance indicators to determine rankings. The only rankings that matter are the one's done by the selection committee. And it should give everyone pause when one of the members talks about how important he thinks the "eye test" is.

Here is my question, if you took the possession efficiency rating and ranked teams that had played each other (i.e. ACC), how different would the rankings be than the standings? My answer, not much.

As a practical matter, what does it really do for you? How important should it be as a tool for the selection committee?

What's great about the tournament is that we play the games to see who the last team standing is. The 66th team may want to use possession efficiency to make their claim to the 65th spot for a play-in game.

But in the end, just win baby.

Posted by: petecard | February 24, 2010 2:42 PM | Report abuse

I remember one year recently where UNC was not good, but the ACC had two final four teams including the champion. And a different team was champion the next...

Posted by: BrokenClipboard | February 24, 2010 3:21 PM | Report abuse

Hah! BrokenClipboard with the comment of the day.

Btw, speaking of UNC. They are so insanely due right now, if I were a betting man I'd be all over them tonight against an FSU team that struggles on the road.

Posted by: Barno1 | February 24, 2010 5:15 PM | Report abuse

I'd hate to see the Terps wint the efficiency margin stat tonight against Clemson but lose the game.

Regards -


Posted by: HughGRection | February 24, 2010 5:49 PM | Report abuse

The comments to this entry are closed.

RSS Feed
Subscribe to The Post

© 2010 The Washington Post Company