- The Season's Over -

Advanced Basketball Statistics 101

December 11, 2009

Phoenix Suns Offense

Phoenix Suns Photo Credit: Andrew D. Bernstein/NBAE via Getty Images

By Zachariah Blott

Advanced statistics is one area that baseball is lightyears ahead of basketball. Bill James was a baseball fan who popularized sabermetrics, the analysis of baseball through objective evidence, back in the 80′s through his tome-like Historical Abstracts. Sabermetrics is named after SABR, the Society for American Baseball Research.

Since then, more meaningful statistics—like OPS and WHIP—have entered the public consciousness and appear on basic player stats pages, Michael Lewis’s well-read “Moneyball” (which is being made into a movie starring Brad Pitt) has shown millions how some clubs use sabermetrics to their advantage, and two pitchers who combined for 31 wins and 0 saves can win the 2009 Cy Young Awards without ruffling many feathers.

In hoops, however, the basic stats still rule what are supposed to be in-depth discussions. Points, rebounds, and assists are the end-all, be-all for many fans and analysts of the game. It’s so bad, All-Defensive Teams are usually made up of a) whoever lead the league in blocks, b) whoever lead the league in steals, and c) household names who average over 20 PPG and are considered good defenders. Rarely does the phenomenal defensive talent who doesn’t fill up the stat sheet in some area make the First Team (Shane Battier still has only made two Second Team lists [Ed. Shameful]).

As much as I love the aesthetic beauty of the game that first attracted me to basketball, I try to find and use meaningful statistics whenever I have a real conversation about what makes a team or a player good or bad. The basic stats do paint a fuzzy picture of what’s happening, but it’s some more advanced stats that add the details we should be looking for.

This article is meant as a primer to help basketball fans understand a little more deeply how to evaluate team and player performance. The first two stats, ORating and DRating, are the most necessary tools in evaluating how well a team scores and defends, and the Four Factors get into the nuts and bolts of what affects ORating and DRating. At the end of the article, I’ll list a few resources where you can find additional information about these numbers or the more advanced statistics that are out there.

Offensive Rating (ORating)

What it measures: How many points a team scores per 100 possessions

Formula: (Points scored * 100) / Possessions

Current NBA leader: Phoenix Suns, 114.5

League average: 106.7

A little more depth: ORating gets at the heart of the question: How good is this team’s offense? Whereas the Warriors average 108.8 PPG and the Trail Blazers only score 95.8, Portland actually has a better offense. Golden State runs the fast break all game and scores 106.5 points per 100 possessions, while Portland is content to walk it up the floor and score 109.1 per 100. ORating eliminates the pace at which a team plays, so that we can see how efficient they are at scoring.

If you’re curious how you can figure out the amount of possessions a team used in a game or season, here’s the common formula that estimates it pretty close: 0.96 * (Field Goal Attempts – Offensive Rebounds + Turnovers + (.44 * Free Throw Attempts))

Defensive Rating (DRating)

What it measures: How many points a team gives up per 100 possessions

Formula: (Points surrendered * 100) / Possessions

Current NBA leader: Boston Celtics, 99.1

League average: 106.7

A little more depth: Similar to ORating, DRating tells us how good a team’s defense is regardless of pace. Pacers’ opponents score 101.7 PPG, whereas the Pistons’ opponents only manage 94.9. However, because Indiana keeps things moving quickly and Detroit barely moves, guess who has a 104.5 DRating and who’s at 107.5. If you can’t follow what I’m getting at, it’s actually harder to score against the Pacers on a given possession.

The Four Factors to NBA success, after the jump …

The Four Factors

Mathematician Dean Oliver pinpointed four specific team statistics in his 2003 book “Basketball on Paper” that winning teams thrive at over the teams they beat: Effective Field Goal Percentage, Turnover Percentage, Offensive Rebounding Percentage, and Free Throw Rate. The order I’ve listed them is their order of importance, and every mathematician who’s looked at these four always ranks them the same way. Here’s what exactly the Four Factors are and how they affect the game.

Anthony Morrow: Shooter

Effective Field Goal Percentage (eFG%)

What it measures: It’s a shooting percentage that factors in the extra points scored on 3-pointers.

Formula: (Field Goals + .5 (3-pt Field Goals)) / Field Goal Attempts

Current NBA leader (Team): Phoenix Suns, .553

Current NBA leader (Player): Anthony Morrow, Golden State, .650

League average: .497

A little more depth: eFG% is like shooting percentage, but it provides a bump for points scored from behind the arc. For example, Morrow’s real shooting percentage is .536, but he is such a prolific 3-pt shooter, his eFG% gets quite a boost (players who play close to the hoop, like Dwight Howard, will always have very similar FG% and eFG%). A team’s ability to hit its shots, and its ability to prohibit opponents from hitting theirs on the other end, is the single most important factor in winning and losing games. This makes sense because hitting shots is what scoring points is all about.

Let me show you how three totally different players all ended up with a similarly average eFG% last year.

Player A did his damage inside, shooting 49% from the field with only 12 trifectas for the year – he had a .492 eFG%.

Player B is known as a gunner, shooting 42% from the field and from deep, where he attempted over a third of his shots – he had a .491 eFG%.

Player C can do a little of everything, hitting 47% overall and 35% from behind the arc, where he took about a fifth of his shots – he had a .502 eFG%.

Three different playing styles, but all three players were right around last year’s league average of .500. Who are they? Here’s a hint: all three averaged over 22 PPG. The answer appears at the end of the column.

Anthony Morrow photo credit: Icon SMI

Turnover Percentage (TO%)

What it measures: How many turnovers a team commits per 100 possessions

Formula: (Turnovers * 100) / Possessions

Current NBA leader: Atlanta Hawks, 11.2

League average: 13.8

A little more depth: Not turning the ball over is obviously important because a team needs the ball in order to score. Teams give up the ball without even getting a shot off once every 7 to 8 possessions, so you can imagine how well an offense would do if it never lost the ball and always had a chance to shoot. This is why guards who can dominate the ball without coughing it up, like Chris Paul, are invaluable to a franchise.

You could use the possessions formula above to calculate TO% for individual players, but they won’t tell you anything because it benefits players who shoot a lot compared to their turnover totals. Due to this flaw, every forward (let alone almost every guard) ends up looking better than Jason Kidd, despite his outstanding 8.8-2.8 A/TO rate.

Ben Wallace

Ben Wallace Photos Credit: Icon SMI

Offensive Rebound Percentage (ORB%)

What it measures: How often a team rebounds its own missed shots

Formula: (Offensive Rebounds * 100) / (Field Goal Attempts – Field Goals)

Current NBA leader (Team): Memphis Grizzlies, 31.6

Current NBA leader (Player): Greg Oden, Portland, 16.0, Ben Wallace, Detroit, 16.0

League average: 26.8

A little more depth: Rebounding a miss is the polar opposite of a turnover. Instead of not having a chance to shoot the ball for two or three possible points, the team now has a chance to shoot again, and often from very close to the hoop where the rebound was secured. An individual player’s ORB% measures what percentage of his own team’s missed shots he personally rebounded while he was on the floor (not for the entire game). This allows for a seldom-used player like Spurs rookie DeJuan Blair to have a 15.7 ORB% even though he only plays 15 MPG.

On the flip side, keeping an opponent from rebounding their own missed baskets is also important for a team. It is simply called Defensive Rebound Percentage (DRB%), and the league average is 73.2 (100 – 26.8).

Free Throw Rate (FT/FGA)

What it measures: How well a team gets to the free throw line and converts the freebies

Formula: Free Throws Made / Field Goals Attempted

Current NBA leader: Denver Nuggets, .330

League average: .231

A little more depth: Getting to the free throw line gives a team a chance to earn points without the opposing team’s defense being a factor, and it eliminates the possibility of turning the ball over. The shooter also has to make them for it to be an effective trip, however, that’s why the formula only includes made free throws. Teams like the Nuggets who play aggressively and love contact will gladly crash into defenders while getting to the hoop, that’s why they do so well in this area (by the way, Denver is ahead of everyone by a mile in FT/FGA).

For the mathematically curious, a .231 means that a team should make 2 free throws for every 8.7 field goal attempts, which should result in 8.7 points because the average eFG% is right around .500. That means free throws make up 19% of all points (2 / (2 + 8.7)).

Additional Resources:

The Wikipedia page explaining APBRmetrics: Basketball’s answer to sabermetrics (they rhyme, by the way).

Basketball Reference’s Yearly Summaries: Provides all sorts of great stuff – just scroll down to the Miscellaneous Statistics section.

Basketball Value: Provides a ton of in-depth statistics to evaluate every facet of the game. You might want to start with the Glossary link at the top.

The APBRmetrics Discussion Board: This can keep you abreast of the most recent ideas in the field.

82games: Contains a whole slew of everything; it’s fun just stumbling through the site checking out all that it has.

Hoops Numbers: A relatively new site that does a decent job of finding new ways of looking at the game and how to evaluate it mathematically.

Answers to the eFG% example:

Player A was Chris Bosh.

Player B was Kevin Martin.

Player C was Kobe Bryant.

Zachariah Blott cannot recommend Rick Telander’s “Heaven Is A Playground” enough.

Get more value with your NBA betting when you compare the NBA odds to find the best number available.

9 Comments »Posted by ETB Contributor on Dec. 11, 2009 at 9:01 am in ETB Articles, NBA

9 Responses

Four Factors visualized version here,

http://www.nbastuffer.com/2009-2010_NBA_Regular_Season_Advanced_Stats.html

Posted by: Serhat on December 11th, 2009 at 10:18 am

Fantastic primer.

Posted by: Vic De Zen on December 11th, 2009 at 11:26 am

Dang I guessed correctly for Bosh and Kobe, but instead of Kevin Martin I guessed J.R. Smith. oh well.

Posted by: purdman on December 11th, 2009 at 3:55 pm

Great post. And if it was Advanced NBA Stats 101, then this Basketball-Reference post must be Advanced Stats 201 then:

http://www.basketball-reference.com/blog/?p=1120

Posted by: James on December 11th, 2009 at 6:37 pm

[...] in the league he’s posted defensive ratings of 110 (2009) and 109 (2010). For reference the league average is 106.7 so he hasn’t shown a drastic improvement to warrant a trend of improvement. That [...]

Posted by: Omar Casspi Wants Out Of Sacramento — A Potential Hornets Target? « Swarm & Sting | A New Orleans Hornets Blog on April 5th, 2011 at 8:19 am

Those are all shit stats, the hard stats are all that matters. Niggas kill me with these nerd stats and arbitrary calculations.

Posted by: jay on July 6th, 2011 at 7:14 pm

Let’s give a guy more credit for jacking up 3s at a terrible percentage than someone who’s solid from within the arc. LMAO @ all those bogus stats, that’s why the NBA doesn’t and will never recognize them.

Posted by: jay on July 6th, 2011 at 7:15 pm

[...] Empty the Bench put together a very good primer on advanced statistics. [...]

Posted by: Points in the Paint | JockSpin.com on January 24th, 2012 at 9:03 pm

@Jay: Use your brain. Suppose Tim Duncan shoots 6 layups and makes 3 of them, while J.R. Smith shoots 6 three pointers and hits only 2. Both players have scored 6 points on 6 shots, yet Duncan’s FG% is 50 percent, while Smith’s FG% is only 33 percent. As follows, J.R. Smith now looks like a terrible shooter who “jacks up 3s at a terrible percentage” even though he has scored just as many points on just as many shots as Tim Duncan “who’s solid from within the arc”. Effective field-goal percentage corrects for this by accounting for the extra point that 3-pointers are worth.

Posted by: anomiemnemonical on September 13th, 2013 at 2:37 pm

Leave a Comment



(will not be displayed)