If you were to read one book to understand the foundation of basketball analytics, it should be Dean Oliver’s Basketball on Paper. He’s widely viewed as the Grandfather of Basketball Analytics. Oliver has made numerous contributions, however his introduction of the Four Factors is both simple and informative.
What are the Four Factors?
Oliver examined derivatives of the box score and concluded teams were good or bad based on four main factors:
1. Shooting percentage from the field.
2. Getting offensive rebounds.
3. Committing turnovers.
4. Going to the foul line often and making those shots.
The next step, of course, was to how to best measure each of the factors. He determined the following as the best fit:
1. Effective Field Goal percentage = (Field Goals Made + 0.5 x 3pt Field Goals Made)/Field Goals Attempted. This simple metric gives the team “extra” credit for a three-pointer made. Thus, it’s a much better measure then field-goal percentage.
2. Turnover percentage = the number of turnovers a team commits per possession.
3. Offensive rebounding percentage = offensive rebounds / (offensive rebound + opponents defensive rebounds).
4.Free throws per field goal attempted = FTA/FGA.
Intuitively these should make sense. A team may not have to excel at all four to be a consistent winner, but certainly at least a couple. For example, perhaps a team generally doesn’t shoot well (eFG%), but could see their fair share of wins if they: A) do not turn the ball over more than normal, B) generate a significant number of offensive rebounds (i.e. rebounding a good number of the “extra” misses) and C) get to the free-throw line a lot if the shots are not falling. Or perhaps a team that turns the ball over often, but may be able to pull games out by shooting well. If you view a game with these metrics in mind, it becomes clear what factors are leading to a win or a loss.
The telling factors
Evan Zamir of The City blog found that “The four factors (eFG%, TOR, FTR, & ORR) explain about 96 percent of point differential.”
I am always looking for a combination simplicity and informative when it comes to analytics and the four factors give us just that.
So how are teams doing so far this year?
It is quite clear the teams with the best records are excelling in most of these factors on both sides of the ball:
If we look at Oklahoma City, for example, its clear why they are off to a great start. Notably, the team has the second-highest eFG% in the league. They also get to the free-throw line a lot. Defensively, the Thunder are forcing tougher shots as evidenced by the fourth-best-eFG%-allowed metric. So, despite turning the ball over at a high rate (highest in the league), this aggressiveness is more than offset by high percentage looks and trips to the line. By looking at these factors, it is easy to see why Oklahoma City is 15-4.
At the other end of the spectrum, a solution to the Raptors issues is not clear. On the offensive end the team ranks 24th in eFG%. (The main culprit here is Andrea Bargnani, who takes the most field goal attempts per game for the team, with a poor 45.4% eFG percentage.)
The picture is not any more resolute on the defensive end, where the team has slipped significantly from a year ago. A solution may be elusive. For example, if defensive eFG% was the lone major issue, the coaches could deploy more aggressive defensive strategies and accept a (hopefully) small tradeoff by committing more fouls. However, the Raptors are both allowing a high eFG% and putting opponents on the line a lot.
One will also see that turnovers are not a problem as several fans have speculated. I see a number of requests to “slow it down” to reduce them.
Two problems with this:
1. Turnovers are not strongly correlated with pace. Of the 13 teams playing at above average pace so far this year, eight of the 13 have below average turnover percentage.
2. A team could be sacrificing eFG% at times with a slower pace. Again, it’s not strongly correlated, however you can envision scenarios to run hard when you gain a possession, thereby increasing the odds for “numbers” or a mismatch on defense. Not having the defense “set” creates higher-value scoring opportunities. The key is, of course, not to force it when it’s not there and bring the ball back out into a half-court set.
With two smart point guards, one suggestion would be to have the Raptors increase their pace dramatically. When you consider the Raptors are a young, athletic team, this creates their best opportunity to improve one key weakness (eFG%). As well, when the defense is not set, you could see many additional opportunities for offensive rebounds (via mismatches and opponents having difficulty finding their man). The turnover percentage may rise, but the offset should be much greater. Oh, and if you score more often, opponents have less opportunity to run against you.
The Raptors have the fast horses. Use them.