True Shooting Percentage, and Why Field Goal Percentage Misleads

What field goal percentage actually counts

FG% = FG / FGA. Makes over attempts. It is a hit rate, and a hit rate is only meaningful when every hit is worth the same. In basketball, hits are not worth the same: a shot from behind the arc scores three, a shot inside it scores two. A player who makes four of nine threes has produced twelve points; a player who makes four of nine layups has produced eight. Field goal percentage records both as 44.4% and reports nothing about the fifty percent difference in output.

The second omission is larger. Free throws are not field goals, so they do not appear in FG% at all — neither in the numerator nor in the denominator. A player who draws seven free throws a game and makes six of them has scored six points that the metric cannot see, and has done so without spending a field goal attempt. For anyone whose game is built on contact, field goal percentage is measuring the minority of their scoring.

eFG%: paying the three-pointer what it is worth

Effective field goal percentage is eFG% = (FG + 0.5 x 3P) / FGA. Every made three counts as one and a half made shots. People often treat the 0.5 as a convention someone chose; it is forced. Points from field goals are 2 x FG + 3P, which factors to 2 x (FG + 0.5 x 3P). Divide by 2 x FGA and you have points from field goals per attempt, halved: eFG% = (2 x FG + 3P) / (2 x FGA). The half-credit is the only number that puts the metric back on the same scale as FG%, so that 50% still means one point per attempt.

Run the wing through it. He makes 5.0 of 12.0, of which 4.0 are threes, so his field goals produce 2 x 5.0 + 4.0 = 14.0 points. Divide by 2 x 12.0 = 24.0 and eFG% = 58.3%. His FG% is 41.7%. The gap of 16.6 percentage points is the entire three-point premium, and it is exactly what field goal percentage was throwing away. To reach the same eFG% shooting only two-pointers he would have to make 7.0 of 12.0 — two more baskets a game.

TS%: folding in the free throws

True shooting percentage is TS% = PTS / (2 x (FGA + 0.44 x FTA)). The numerator is every point the player scored, from anywhere. The denominator is an estimate of how many scoring chances he used to get them, doubled so the result stays on a percentage scale that reads like FG%. The quantity FGA + 0.44 x FTA has a name: true shooting attempts. Double the TS% and you get points per shooting possession directly — 60.2% means 1.205 points every time this player finishes a possession with a shot.

The centre is the case that makes the point. He converts 6.9 of 12.0 inside, an excellent 57.5%, and he draws 5.2 free throws — but he makes only 2.6 of them. His 16.4 points come out of 12.0 + 0.44 x 5.2 = 14.288 true shooting attempts, for a TS% of 57.4%. He leads the league table on FG% and trails both team-mates on the metric that counts everything. Hand him an 80% free-throw stroke and nothing else changes: 4.16 makes, just under 18.0 points, TS% 62.8% — first of the three. That single skill is worth more than five percentage points of true shooting.

Where the 0.44 comes from

The 0.44 is the weakest joint in the formula and the one worth understanding. It answers a single question: how much of a possession does one free throw attempt consume? If every trip to the line were two shots, the answer would be 0.5, because two attempts would together consume one possession. The coefficient sits below 0.5 because several kinds of free throw do not fit that pattern.

A three-shot foul on a three-point attempt gives three attempts for one possession, so each is worth about 0.33. An and-one gives a single free throw on a possession that the made field goal already ended, so that attempt consumes nothing at all. Technical and flagrant free throws end nobody's possession. Averaged across a full league season, these cases pull the figure down to roughly 0.44, which is why Basketball Reference uses 0.44 in its possession estimate and in true shooting attempts.

Two consequences follow. First, 0.44 is a league average applied to an individual, so it is slightly wrong for everyone. A player who lives on and-ones truly consumes less than 0.44 of a possession per attempt; a player whose free throws all come in clean two-shot pairs consumes more. Second, because the coefficient sits in the denominator, an error in it moves TS% by very little. Swap 0.44 for 0.5 on the centre and his denominator rises from 14.288 to 14.600, moving TS% from 57.4% to 56.2% — real, but far smaller than the 16-point gaps the metric was built to expose.

Three players, three different leaders

Line them up on identical volume, 12.0 field goal attempts each, and watch the order move. On FG% the centre wins at 57.5%, the guard is second at 45.0% and the wing is last at 41.7%. On eFG% the wing takes it at 58.3%, the centre falls to 57.5% and the guard drops to last at 51.7%. On TS% the guard leads at 61.3%, the wing is second at 60.2% and the centre is last at 57.4%. Nobody's shooting changed between those three sentences. Only the question did.

The guard is instructive because he is a poor shooter by both field goal measures and the most efficient scorer on the floor. He converts 45.0% of his shots and only 36.4% of his threes, which is why his eFG% is the worst of the three. But he makes 6.3 free throws a game at 85.1%, and those points arrive for 0.44 x 7.4 = 3.256 attempts of denominator instead of 7.4. His 18.7 points over 15.256 true shooting attempts come to 1.226 points per shooting possession, the best number on the team.

How to use all three without fooling yourself

TS% is the right default for the question most people are actually asking: how many points does this player get out of a scoring chance? It is not the right tool for every question. FG% still tells you something eFG% and TS% cannot — how often the ball goes in — which matters when you are scouting finishing at the rim or diagnosing a shooting slump. Split it by zone and it becomes useful again; used as a single season-long number, it is a summary of a mixture and summarises nothing.

The second discipline is volume. Efficiency and workload trade off: a player taking six shots a game inside a comfortable role is not comparable to one taking twenty-two with the shot clock at four. That is what usage rate is for, and it belongs beside TS% rather than after it. A 61% true shooting on low usage and a 58% on very high usage are different achievements, and no single number resolves them. Read them as a pair.

Frequently asked questions

What is a good true shooting percentage?

Read it against the league, not against a fixed threshold, because league-wide true shooting drifts with rule changes and shot selection. The usable rule is relative: a player several points above the league average is efficient, one several points below is not, and the same raw number means different things in a high-scoring era and a low-scoring one. Doubling it gives points per shooting possession, which is the version worth quoting: 60% is 1.20 points every time that player finishes a possession.

Can true shooting percentage go above 100%?

Yes, in small samples. Because the metric is points divided by twice an estimate of shooting possessions, anything that produces points cheaply pushes it up. A player who makes one three-pointer on his only attempt scores three points on one true shooting attempt: 3 / (2 x 1) = 150%. Two made free throws with no field goal attempts give 2 / (2 x 0.88) = 113.6%. Over a full season the arithmetic settles, but for a single game the ceiling is not 100%.

Why is the centre's eFG% identical to his FG%?

Because he attempts no three-pointers. The 0.5 x 3P term in the numerator is zero, so eFG% collapses to FG / FGA — the definition of FG%. That is the honest way to see what eFG% adds: it is FG% plus a bonus that only exists for players who shoot threes. For any player with zero three-point attempts the two numbers are the same by construction, which is why comparing a centre to a wing on eFG% is fairer than on FG% but still incomplete.

Does true shooting percentage account for turnovers?

No, and that is a real limit. TS% counts only possessions that ended in a shot or a foul, so a player who turns the ball over on one drive in six is measured as though those possessions never happened. To see the full cost of a player's ball-handling you need a metric whose denominator includes turnovers, such as points per possession or offensive rating. TS% answers how efficiently he shoots, not how efficiently he plays.

Should I use TS% or eFG% when comparing two shooters?

Use eFG% when the question is strictly about shot-making from the floor — comparing two spot-up shooters, or judging whether a shot diet is producing. Use TS% whenever free throws are part of the story, which is nearly always for guards and centres. The two players in this article who differ most on eFG%, the wing at 58.3% and the guard at 51.7%, are only 1.1 points apart on TS%, and in the opposite order. The free-throw line is doing all of that work.