# Tempo-Free Stats Primer

*Updated: 3/19/14*

Over the past two seasons, we’ve increased our adoption of tempo-free statistics into many of our game preview and recap stories, but we didn’t spend any time really explaining them. So, we created this page to give you a breakdown of how tempo-free statistics work, why they’re preferred (in some cases) and which ones can still be (in our opinion) highly misleading.

First, **why should we use tempo free statistics**? Tempo simply skews many normal statistics. A team that plays up tempo is expected to score more than a team that plays a slower tempo. They’ll also likely get more rebounds, force and have more turnovers; more of almost everything.

So, how can you compare the two statistically? That’s where tempo-free comes in. As the name suggests, these statistics look to strip tempo from the numbers, to give a better picture of how the teams actually stack up.

They’re far from perfect. But in many ways they’re a better measure of long-term average strengths and weaknesses.

#### Possessions

Possessions are the basis of most tempo-free formulas. While you could go back and pull up box scores and count how many possessions teams had in every game, there’s a mathematical formula that gives a rough estimate.

*Field Goal Attempts – Offensive Rebounds + Turnovers + (0.475 x Free-Throw Attempts)*

…or in stats language:

FGAs – orebs + TOs + (0.475 x FTAs)

The explanation of the formula is mostly straightforward, until you get to end. Every possession will end one of three ways: A shot, a turnover, or a free-throw. If a team gets an offensive rebound, they get another chance at the same possession, hence why those are subtracted.

But then there’s the free-throw multiplier. While a possession could end in free-throws, the number of free-throws could vary. As is the slight possibility it could end in a free-throw.

The number, .475, comes from one of the largest proponents of tempo-free statistics, **Ken Pomeroy**. He determined back in 2005 that on average 47.5% of free-throws result in a change in possession.

Being an average, that number isn’t perfect. And it means that the number of possessions for a single game will likely end in a fraction, that has to be rounded to the nearest whole number. In the long run, the numbers are likely to average out.

There are some weaknesses. Possessions doesn’t take into account overtime games. Obviously, if a team plays multiple overtime games, they will have more possessions, which can skew statistics. Also, an “and-one” counts as 1.475 possessions in the formula above. That’s factored in to the free-throw factor, but if there are multiple “and-ones” in a game it could result in an extra calculated possession or two. (This is especially noteworthy given the new emphasis on foul calls)

Possessions per game averages tend to range between about 60 and 72 over the course of a season. In a single game, the number can range from closer to 50 to well over 80 in an up-and-down game.

#### Points per Possession (O-PPP)

and Defensive Points per Possession (D-PPP)

These are two of the easiest to understand statistics, and in my opinion two of the most useful. Calculating these statistics are easy:

O-PPP: Points / Possessions

D-PPP: Points Allowed / Opponents Possessions

These can, and should replace points per game and points allowed per game, as it simply strips those numbers of misleading tempo.

But again, they’re not perfect. The caliber of opposing offenses and defenses could cause a less talented team to have better numbers than a more talented team because of their competition.

With the new rules in place this past season, the average for both numbers was near 1.06. These numbers run between 1.00 to 1.15 as an average.

#### Effective Field Goal Percentage (eFG%)

Effective field goal percentage is a way to combine two-point and three-point field goal percentages in a way to place more emphasis on three-point shooting, since those shots are worth more points. The formula used to calculate it is:

*(Field Goals Made + 3-point Field Goals Made * 0.5) / Field Goals attempted*

Or in stats speak…

(FGM + 3PTFGM * 0.5) / FGA.

This is one of the **Four Factors**, and is considered to be the most important of the factors.

While it can be a good tool to simply a team’s offensive effort, it does strip out some important information: Why. eFG% tells me which team is more likely to score if there’s not a turnover. And if offensive rebounds are taken out of the picture.

Two things that *are included* when looking at points per possession.

eFG% also doesn’t explain why teams are good shooters. Are they exceptional inside and alright on the perimeter? Are the lights out from outside, and so-so in the paint?

Up until late in this past season, we didn’t use eFG% on the site. But given it’s adoption in the greater statistical circles, we have decided to begin implementing it’s use.

#### Rebounding Percentage, (Reb%)

Offensive Rebounding Percentage (OReb%)

and Defensive Rebounding Percentage (DReb%)

Rebounds are one of the statistics skewed by more than just tempo. The number of your opponents missed shots has a big factor in the number of defensive rebounds available, and the number of your missed shots plays into the number of offensive rebounds available.

These statistics try to rectify both of those issues by making those numbers a percentage of the total rebounds available. Here’s the math:

OReb%: Offensive Rebounds / (Offensive Rebounds + Opponent’s Defensive Rebounds)

DReb%: Defensive Rebounds / (Defensive Rebounds + Opponent’s Offensive Rebounds)

Reb%: (OReb% + DReb%) / 2

OReb% is considered one of the **Four Factors**.

According to Pomeroy, even though the skillsets are similar, there’s no correlation between a team’s OReb% and their DReb%. That means a team could by more skilled in one over the other.

50% is the average for total rebounding percentage. For the all-important offensive rebounding percentage, the average is about 31%.

#### Turnover Rate (a.k.a. Turnover percentage)

This is simple enough: The turnover rate is the percentage of possessions that end in turnovers. High-tempo teams are going to have more turnovers generally because there are more chances to turnover the ball. The math is simple:

TORt: Turnovers / Possessions

There is also an opponent’s turnover rate, which is the percentage of time you force a turnover on defense. It’s a very similar formula:

OppTR: Opponents Turnovers / Opponents Possessions

Turnover Rate is another of the four-factors.

#### Free Throw Production (a.k.a Free Throw Rate)

Free throw production is simply meant to show a team’s ability to get to the free throw line. The formula is as follows:

FTProd: Free Throws Attempted / Field Goals Attempted

If this formula looks different to what you’re used to, that’s because it can also sometimes be free throws made / field goals attempted.

Free-throw production is the last of the four-factors. But, if a team isn’t a good free-throw team, the number above isn’t as important.

This is one where the numbers are superficial: They don’t mean a whole lot by themselves, but are good for comparison purposes The number rose this past season, with the average tends to be around 39, ranging from 30 to 50 over the course of a season.

For the most part, that’s it. There are a ton more tempo-free statistics, but the ones listed above are the major ones we use on the site.

There are a few other questions that have popped up throughout the year, which I’ll answer below.

#### Frequently Asked Questions (FAQ)

**Will you replace individual game stats with tempo-free Stats?** Not entirely. While tempo-free are solid when looking over the long term, I don’t like using them by themselves for individual games. Tempo is an important part of the game, and is a factor in it’s outcome, and regular statistics often to provide a better picture for individual games. But in the long term, especially when you have teams of different tempos matching up, trying to compare using standard statistics is difficult.

**Why didn’t you include (insert tempo-free stat here)? **I’m also not wanting to completely confuse readers not used to tempo-free statistics. I chose the ones I feel help provide the best overall picture of individual teams.

And that’s really what I’m trying to do. I’m not trying to necessarily predict performance. I’m using the numbers to help paint the picture: show what these teams are great at, and not so great at.

**What are the Four Factors you keep mentioning? **You can trace the beginnings of the popularization of tempo-free statistics to a book called “Basketball on Paper,” by Dean Oliver. In it, Oliver, who was 19 when he wrote it, indicated four factors that determine a teams success. They are:

- Effective Field Goal Percentage (eFG%)
- Turnover Percentage (ToRt)
- Offensive Rebounding Percentage (OReb%)
- Free Throw Production (FTProd)

These are weighted, with eFG% the most important of the bunch, although it can differ somewhat from team-to-team.

**You’re a dork.** That’s not a question.