Pace and Possessions: Why Per-Game Stats Mislead

A team that drops 120 points a night sounds like a juggernaut. Sometimes it is. Just as often, it's a team that simply plays fast — racing up and down the floor, taking quick shots, and piling up raw totals that have almost nothing to do with whether it's any good. Points per game is the first stat anyone learns and one of the most misleading in the sport, because it quietly bundles together two things that ought to be kept apart: how well a team plays and how fast it plays. Untangling them starts with a single concept — the possession.

Why per-game totals deceive

Scoring is part skill and part opportunity. Every trip down the floor is a chance to put points on the board, so a team that takes more trips will score more points even if it's no better at converting any one of them. That's the whole problem with per-game totals: they reward volume of opportunity, not quality of execution. Two teams can be playing identically efficient basketball — getting the exact same value out of every possession — and still finish the night ten or twelve points apart on the scoreboard purely because one of them played at a sprint and the other at a crawl.

The same trap catches defenses. A team that "allows only 105 a night" might be a brilliant defense, or it might just walk the ball up and limit the total number of possessions in every game it plays. Points allowed per game, like points scored per game, is contaminated by tempo on both ends. To compare teams honestly you have to scrub the pace out, and to do that you first have to count the thing pace is measured in.

What a possession actually is

A possession is one team's continuous control of the ball — it begins when you gain the ball and ends when you give it up, whether that's by making a shot, missing one the other team rebounds, or turning it over. Crucially, an offensive rebound does not start a new possession; it extends the same one. Grab your own miss and put it back, and that whole sequence — miss, rebound, putback — counts as a single possession, because you never surrendered control of the ball.

That definition is clean, but the box score doesn't hand you a "possessions" column. So we estimate it from the events the box score does record. The standard estimate has been a fixture of basketball analytics since Dean Oliver's work:

Possession estimatePoss ≈ FGA − ORB + TOV + 0.44 × FTA

Read it event by event. Every field goal attempt (FGA) ends a possession — unless your own team rebounds it, which is why offensive rebounds (ORB) are subtracted back out, since they extend a possession rather than end one. Every turnover (TOV) ends one too. And trips to the free-throw line have to be folded in carefully: most are two shots, but some are one-and-ones, three-shot fouls, or and-ones, so multiplying free-throw attempts (FTA) by 0.44 converts them into possession-equivalents using the league's long-run average. The result isn't exact — it's an estimate — but it's accurate enough that the whole edifice of per-possession analysis rests on it.

Pace: possessions per 48 minutes

Once you can count possessions, pace falls out immediately. Pace is just the number of possessions a team uses per 48 minutes — a full game's worth of regulation time.

PacePace = 48 × Poss / (team minutes / 5)

The minutes adjustment handles overtime and lets you compare teams on a level 48-minute footing. A high-pace team churns through more possessions per game; a low-pace team grinds out fewer. That's all pace is — a count of opportunities, a description of tempo. It is emphatically not a measure of quality. A fast team isn't a good team and a slow team isn't a bad one; they're just teams that have chosen different speeds. Pace is a stylistic choice, the way a band chooses a tempo, and a fast song isn't a better song.

The fair comparison: per 100 possessions

If per-game totals are contaminated by tempo, the fix is to change the denominator from "per game" to "per possession" — and by convention, per 100 possessions, which keeps the numbers in a familiar range. That's exactly what offensive and defensive ratings do, and it's why I lean on them so heavily; I broke down the full machinery in offensive and defensive rating. Offensive rating is points scored per 100 possessions, defensive rating is points allowed per 100, and the gap between them — net rating — is the cleanest one-number summary of team quality there is. A possession is a possession whether it takes four seconds or twenty-four, so once everything is expressed per 100 of them, the sprinters and the grinders finally stand on the same scale.

A clearly-illustrative example

The cleanest way to feel this is with round, made-up numbers. The following figures are a hypothetical illustration, not real teams. Picture two clubs that are exactly as good as each other — both score 1.15 points per possession, every single night. The only difference is tempo. The Sprinters play at 105 possessions per game; the Grinders play at 90.

Multiply it out: 1.15 × 105 lands the Sprinters near 121 points a game, while 1.15 × 90 leaves the Grinders around 104. That's a seventeen-point gulf in scoring average between two teams that are playing identically efficient basketball. Per-game points would crown the Sprinters as a runaway offense and bury the Grinders as mediocre. Per-possession, they're twins. Now flip the same logic onto defense and you see how a slow team can post a gaudy "points allowed" figure without ever forcing a difficult shot — it just hands out fewer possessions to score on.

Pace is a style, not a verdict

None of this makes fast basketball good or slow basketball bad. A team might push the tempo because it has the legs and the shooters to thrive in chaos, or it might slow things to a halt to shorten the game and protect a thin bench or a defense-first identity. Both can win titles; both can miss the playoffs. Pace tells you a team's preferred speed, which is genuinely useful context — it shapes matchups, foul trouble, and how variance plays out — but it is not a quality grade, and treating a high scoring average as proof of a great offense is the single most common mistake in casual analysis.

This is also why pace matters when you read the rest of the box score. A fast team will accumulate more raw rebounds, assists, steals, and turnovers simply by having more possessions, which is why the smarter versions of those stats are expressed as rates. The same instinct underlies the four factors, which measure shooting, turnovers, rebounding, and free-throw generation per opportunity rather than per game — so that tempo never sneaks in and inflates the picture.

The takeaway

When a broadcast flashes "averaging 119 a game," the honest follow-up questions are at what pace, and against what defenses? Per-game totals answer how fast a team plays at least as much as how well. Possessions are the true currency of basketball, pace is how many of them you spend, and per-100-possession ratings are the exchange rate that lets you compare any two teams fairly. Learn to think in possessions and the scoreboard stops fooling you.