If you listen to analytics-minded basketball handicappers, you’ll often hear the term “efficiency” being bandied about. If you’re not using offensive and defensive efficiency in your own handicapping process, you’re really behind the curve. The modern betting marketplace is driven by analytics. Efficiency is what makes the motor run in basketball.
What is offensive or defensive efficiency? The simplest way to think about it is “scoring adjusted for pace.” Basketball fans are no longer at the mercy of TV or radio announcers saying a team has a “good defense” just because it happens to hold on to the ball for a long time on offense. Conversely, you won’t be misled by announcers saying that a fast-paced team plays “bad defense” when that’s actually an illusion created by possession counts.
If you’re of a certain age, you’ll remember that the media used to rave about Princeton’s defense before the shot-clock era because the Ivy League power played such low-scoring games. If your games are ending 52-50, that must be due to great defense, right? Made sense until you watched Princeton on TV and realized it didn’t have great shot blockers or defensive rebounders or swift sprinters disrupting passing lanes. Princeton’s defensive skill sets were mortal. Its games were low-scoring because the offense held on to the ball forever trying to create back-door cuts for layups.
It’s one of the ironies of the semi-recent era that the media raved about Princeton’s defense, but it was ultimately the principals of the Princeton offense that spread through the sport.
On the other side of the floor, way too many in the mainstream media derided Golden State’s defense because the Warriors played such high-scoring games. They were a team with rebounders, shot blockers, speed and the ability to force turnovers by taking opponents out of their comfort zone. When you adjust for possession count during this recent dynasty, the Warriors always graded out as fantastic on defense. The games were high-scoring because it was racehorse basketball. Golden State’s defensive skill sets were a big reason it used to dominate the spot.
How do you calculate offensive and defensive efficiency?
— Offense: Points divided by the number of possessions.
— Defense: Points allowed divided by the number of possessions.
Sometimes you’ll see it expressed numerically as “points per 100 possessions.” Other times, it’s simply “points per possession.” If you prefer your handicapping diet to have numbers that say 1.023 or 0.984, then use per possession. If you’d rather see 102.3 or 98.4, use 100 possessions. What matters most is what you learn from the numbers, not where the decimal point goes.
How do you calculate the number of possessions?
The standard formula for estimating possession counts used by the field (as first presented publicly in “Basketball on Paper” by Dean Oliver in his landmark 2004 book) is shot attempts minus offensive rebounds plus offensive turnovers plus 0.44 times the number of free throw attempts. If you’re not comfortable using a spreadsheet, you can get close enough for learning purposes at a glance by using half the number of free throw attempts rather than 0.44 — a difference so slight it won’t influence the conclusions you’ll draw.
Of course, the easiest thing to do is to visit websites that do all the math for you!
NBA: “Hollinger Team Stats” at ESPN show offensive and defensive efficiency on the far right side of the chart. You can click on the column header to sort from best to worst. The first column is pace, if you’re just interested in possessions per game for each team.
College: Ken Pomeroy’s home page at kenpom.com goes a step further and adjusts everyone’s efficiency based on the caliber of opposition. There’s also a column for adjusted tempo that accounts for the pace of opponents.
The smartest influences in the betting markets have been using variations of these concepts for years. Dean Oliver’s book came out well over a decade ago. And he was the first author to publish the math for public consumption. Some sharp syndicates were betting on pace, particularly on Over/Unders, in previous decades. It has become publicly known in recent years that legendary North Carolina coach Dean Smith used the concept of adjusting stats for possessions as far back as 1959.
Here’s an example of possible illusions created by not adjusting for pace.
Let’s say Saint Mary’s had these points-per-game rankings:
— Saint Mary’s on offense: No. 42 in the nation.
— Saint Mary’s on defense: No. 32 in the nation.
Because even current-day announcers are prone to use raw stats rather than ones adjusted for pace, you’ll still hear analysts suggesting that Saint Mary’s has a strong defense that helps maintain its status as a power out west. The offense isn’t quite as good, but this is a “balanced” team led by its defense.
The problem is, Saint Mary’s is like Princeton. The Gaels are very patient with the ball, which can create misleadingly low scores because of a slow pace.
Saint Mary’s rankings after pace adjustments:
— Saint Mary’s on offense: No. 1 in the nation.
— Saint Mary’s on defense: No. 121 in the nation.
This isn’t a balanced team that’s slightly better on defense than offense. Not at all! This is actually a very smart offense that consistently gives itself good shots — but the Gaels are awful on defense by the standards of quality teams. They’re not even in the top 100 nationally in points allowed per possession.
We’re not going to suggest this happens to all college or pro teams. Many play near the national or NBA average in pace, so their scoring stats aren’t misleading. But teams on the extreme end, either very slow or very fast, will have illusions created by their scoring averages that are compounded by the media’s tendency to use any positive stat it can find to hype a corporate partner. As handicappers and bettors, you must dig to learn the truth about the real skill sets of teams on both sides of the floor so you can make smart bets.