ACC Efficiency Rankings (Week 2)

[loran16]

I posted this originally in the ACC Thread for the week from last week, but I think it could use it's own thread since it's not talking about the individual games.

Below are the current efficiency numbers for each ACC team in conference play thus far. These numbers used to be kept track of publically by basketball prospectus' John Gasaway, but he's currently putting them behind the site's new pay-wall, so I'll be posting them here all season.

These numbers go beyond win-loss record to, over time, show who truly is the best, and then second best team in the ACC team, and so forth. They're better at doing so than win-loss records, as they truly show which teams are actually beating up on opponents and which are lucky to skate by. Early on of course, they're skewed by outliers (games against Wake Forest), but they're still quite interesting.

The relevant numbers are as follows:
Tempo: Possessions per Game (A measure of how up-pace a team plays.)
Offensive Efficiency: Amount of points scored by a team's offense per possession.
Defensive Efficiency: Amount of points allowed by a team's defense per possession.
Efficiency Margin: The amount a team will outscore it's opponent per 1-possession-each (basically O Efficiency minus D Efficiency)

For ease of reading, I've added a final column, which shows how the Efficiency Margin of each team translates into that team's average margin of victory (or defeat) in an average ACC game (67 Possessions). This is just for display purposes...a team like Duke which plays faster than that will win by more, while a team like UVA which plays slower will win by less.



Thoughts:

1. There are good ACC Teams. There are bad ACC Teams. And then there's Wake Forest. The Deacons are EPICLY bad, as you can see by the gap in average-margin-of-victory/efficiency-margins. For reference, the worst major conference team last year, Indiana, had a -0.20 efficiency margin, which means Wake is 1.5 times WORSE than that team. YIKES! 0-16 is a definite possibility for the Deacs.

2. Because Wake is so bad, teams that play them are inflated temporarily. Currently, that's VT (TOP OF THE STANDINGS!), MD, and NC State. Of course, the fact that NC State is still negative in efficiency despite having a game against Wake is NOT a good sign for State's long-term prospects (The same can be said of Maryland). Still, VT is NOT quite as good as their efficiency ranking implies, but at least they've taken advantage of beaten Wake in a way that MD and NC State haven't.

3. Clemson and BC are surprise teams in the top 1/3. Clemson's ranking is mainly because of an insane shooting night (83.3 eFG%) against GT, but well GT isn't amazingly terrible, and while it was probably a fluke, it's still a pretty damn good performance. BC's been doing it the same in every game...great offense, terrible D. (The same appears true of Clemson).

4. Only 3 teams in the ACC - Duke, VTech, and FSU - have above average Offensive and Defensive Efficiencies. Average O and D efficiencies in college basketball tend to be around 1.00. These three teams are the only ones who are both scoring at a greater rate and preventing opponents from doing so. And really, none of the three are doing so by much (except maybe VT, who has the edge in having played against Wake).

Overall, so far it's a wide open race between 11 teams in the conference. Only Wake stands out as truly terrible and as unlikely to upset anyone.

[Saratoga2]

While there is one weak sister in the ACC, by and large the teams are competitive. Is there any breakdown of how teams fare at home versus away. It would be expected that home games would result in superior numbers and those away inferior. Also, teams like Duke, which has played a lot of home games should have an advantage, although I don't know how many ACC teams have played a lot on the road.

I presume, and may be wrong in this, that all the scheduled games are included in these numbers, not just ACC games. Can you confirm that?

[Saratoga2]

I went back and reread your post and noticed these were just conference games, so ignore my question about that. Also it makes moot my question about home and away games, since that will balance over the season.

[gw67]

loran16,

Thanks for posting. This approach is an indication of how well teams are playing and appears to be similar to Pomeroy. There are several factors that can skew the data at this time (NOT just playing Wake) - small data base; playing one game against a hot team; playing on the road, etc. These numbers will be more meaningful toward the end of the season when these factors tend to even out. Pomeroy is just starting to make sense, IMO. The bottom line; however, is still wins and losses. Playing well and losing doesn't help you finish higher in the league standings.

gw67

[loran16]

loran16,

Thanks for posting. This approach is an indication of how well teams are playing and appears to be similar to Pomeroy. There are several factors that can skew the data at this time (NOT just playing Wake) - small data base; playing one game against a hot team; playing on the road, etc. These numbers will be more meaningful toward the end of the season when these factors tend to even out. Pomeroy is just starting to make sense, IMO. The bottom line; however, is still wins and losses. Playing well and losing doesn't help you finish higher in the league standings.

gw67

Of course. Tiny Sample size is a major issue, that should be noted. But you can see trends starting to take shape (multiple teams being only O, no defense; Wake being godawful), so I thought posting would be useful anyhow. Besides, Gasaway used to keep track from this early on anyhow, so I thought I would too :-P.

Yes Wins and Losses is the bottom line. The point generally though is that if you lose close games and blow out victories, you're more likely to win in the future than a team that gets blown out in losses and wins close games.

[loran16]

Quick Update...Duke solidly jumped to #2 by beating NC State solidly...but Virginia tech has not only kept their #1 ranking, but IMPROVED IT with a huge win over Maryland.

Clemson of course dropped, and right now the only 3 teams with above average O and D performances are Duke, VT, and Georgia Tech (Thanks Wake and Carolina!).

Oh, and I didn't think it was possible, but Wake got worse. My goodness.

---------------------------------------------------------------------------
With Duke playing Wake, Duke SHOULD emerge as #1 at the end of this weekend, though by how much is a question. VT is inactive against the ACC this weekend, so they're stuck where they are.