Explaining some essential basketball statistics

In most sports today, simple statistics have been supplemented and, for many, supplanted by various forms of advanced metrics. Some people find this profusion of analysis helpful and fascinating, yet others find it confusing and a bit overwhelming.

Our purpose here is to explain some of these basic and advanced metrics and clarify why the latter can provide additional insight into team and individual performances. This article is by no means meant to be comprehensive, but rather to provide a baseline for you to better understand how the statistics are calculated and used.

The Raw Numbers

Raw per game or season statistics are fairly simple. Generally, it is the target statistic divided by the number of games played. For example, broadcasters often refer to Maryland's prolific offense by noting that the Terps average 82.8 points per game. Most of us are familiar with this simple calculation. After 19 games, the Maryland women's basketball team had scored 1,573 points (1,285 on field goals and 288 on free throws) so 1,573/19 = 82.8 points per game. Stories and broadcasters might also cite their shooting percentage. In this case, Maryland has made 596 of the 1228 shots they've attempted, or 48.3 percent.

What does this tell us?

In one instance, you might quickly glance at Maryland's average of nearly 83 points and see that the Terps scored only 74 points in their game against Illinois. You might then think, "Oh, Illinois held Maryland to almost nine points below their season average. They must have done a good defensive job." This might be true and, for some support and confirmation, you might think to look at how Maryland's shooting percentage for this game compared to their shooting percentage for the season. Already you can see an instance where a single bit of raw statistical information provides a limited view of a team's or an individual's performance and might need to be supplemented by looking at some other piece of information. But even this might not always be the most informative way to look at that performance.

For example, how might you look at Maryland's 74 points if you saw that Illinois had slowed the pace of the game so that Maryland had only 60 possessions rather than their usual 74? Or, what if they had accelerated the pace so that the Terps had 85 possessions? The former would lead you to believe Maryland had scored very efficiently, whereas the latter might indicate Maryland shot poorly or committed lots of turnovers. Thus, it is often more helpful to look at possession-based (sometimes called "tempo-free") statistics than it is to look at simple raw numbers.

What's a Possession?

Before we can establish this tempo-free metric, we need to define a possession. It seems simple, right? When a team has the ball, it has a possession. But is that the end of the story? For example, when a team takes a shot and misses but gets an offensive rebound, do you count that as one possession or two? Or, if Team A has possession of the ball and their coach is assessed a technical foul and what ensues is this: (1) one free throw attempt for Team B and (2) the ball is then returned to Team A. In this instance, did Team B have a possession? Or, did they have a half of a possession since they only had the chance to score one point? So you see, defining possession is a bit more complex than you thought, no?

Like him or not, Ken Pomeroy's method for determining possessions is among the more widely used. It's also fairly simple. This is the formula that underlies possession-based statistics used in our stories. That formula is:

(Field Goals Attempted - Offensive Rebounds + Turnovers + .475*Free Throws Attempted)

Thus, for Maryland: 1,228-285+306+.475*381 = 1,410. Divide this by their 19 games and you have 75.3. Note that the formula includes a minor adjustment for "team" or "deadball" rebounds (most of which occur when a player misses the first of two free throws and thus does not affect possessions).

If you're wondering why .475*FTA? This is an adjustment factor for occasions when a single free throw is attempted, such as in the situation described above or for a single free throw after a made basket for the "traditional" three-point play. As you'll see, this adjustment factor can change slightly depending on the measurement you're trying to make, but the principle remains consistent.

Before we start, we need to note that there are a number of other advanced analyses that are beyond the scope of this article. The most frequently cited of those, Offensive Rating or ORtg, is usually applied to individuals rather than teams. ORtg was developed in 2004 by Dean Oliver and is meant to measure points produced per 100 possessions. It involves a significantly more complex set of statistics and calculations than any metric we cite. (Oliver's explanation requires about seven pages in Appendix 1 of his seminal book Basketball on Paper.) Our explanation is limited to some of the team-based statistics we expect to see most often.

Let's get down to those stats

Offensive Points Per Possession (OPPP) is the average number of points a team scores divided by the average number of possessions per game. Example: Maryland averages 82.7 points per game and has 75.3 possessions. 82.8/75.3 = 1.10.

Defensive Points Per Possession (DPPP) is the average number of points a team has scored against them divided by the average number of possessions per game. Example: Maryland's opponents average 59.4 points per game and 75.3 possessions. 59.4/75.3 = 0.79.

Effective Field Goal Percentage (EFG%) accounts for the fact that a three-point shot has 50 percent more value than a regular basket. The formula is (FGM+.5*3FGM)/FGA. Example: 93 of Maryland's 596 made field goals are three-point shots. Thus, for Maryland, (596+.5*93)/1228 = 52.3 EFGP. We now see that if Maryland had not attempted a single three-point shot, they would need they would need to have made 52.3 percent of their 1,228 shots (642) to score the same number of points (1,285) on their FGA.

Points Per Weighted Shot (PPWS) measures how efficiently a team (or player) turns field goal attempts plus free throw attempts into points. This formula is: Points/(FGA+(0.475*FTA)). Continuing with Maryland as the exemplar, 1,573/(1228+(0.475*381)) = 1.11.

Free Throw Rate (FTR), a possession-based metric, measures how effectively a team scores from the free throw line. The formula is: FTM/(FGA+TO+(0.44*FTA)). For Maryland: 288/(1228+306+(0.44*381)) = .169.

Free Throw Productivity (FTP) measures the ability of a team to reach the free throw line and how effectively they score once they get there. The simple calculation is FTM/FGA. Maryland's is: 288/1228 = .234.

Turnover Rate (TORate) measures the number of possessions that end in a turnover. When Maryland has the ball: 306/1431 = .214.

Rebound Percentage or Rebound Rate (Reb%) is intended to measure the percentage of an opponent's shots that were successfully rebounded. While the arithmetic for calculating this metric is straightforward, it requires several steps. You need to separately calculate a team's offensive and defensive rebound percentages. The formula for the former is OReb/(OReb+OppDReb). Conversely, the formula for the latter is DReb/(DReb+OppOReb). Through 19 games, Maryland has 305 offensive rebounds and 521 defensive rebounds. Their opponents have 217 offensive rebounds and 384 defensive rebounds. Plugging the numbers into the two formulas we have Maryland's offensive rebound percentage as 305/(305+384) = 44.3% and the defensive rebound percentage as 521/(521+201) = 72.1%. To calculate the rebound rate, add the two percentages and divide by two: (OReb%+DReb%)/2. So, for Maryland, (44.3+72.1)/2 = 58.2. (Note: This calculation also adjusts for "deadball" rebounds.)

Selected Glossary

(Single and plural are interchangeable):

2FGA = 2 point Field Goal Attempt

2FGM = 2 point Field Goal Made

2FG% = 2 point Field Goal Percentage (same as 2FGP)

3FGA = 3 point Field Goal Attempt

3FGM = 3 point Field Goal Made

3FG% = 3 point Field Goal Percentage (same as 3FGP)

AST = Assist (sometimes just A)

BLK = Block

DR = Defensive Rebound (sometimes DRB or DReb)

EFG% = Effective Field Goal Percentage (sometimes EFGP)

FG = Field Goal (includes 2 point and 3 point field goals)

FG% = Field Goal Percentage

FGA = Field Goal Attempt

FGM = Field Goal Made

FT = Free Throw

FT% = Free Throw Percentage

FTA = Free Throw Attempted

FTM = Free Throw Made

OR = Offensive Rebound (sometimes ORB or OReb)

R = Rebound (sometimes RB or Reb)

STL = Steal

TO = Turnover (sometimes TOV)