Basketball Per Game and Per 36 Minutes Calculator
Basketball Per Game and Per 36 Minutes Calculator
Points, rebounds or any basketball counting statistic per game, per minute, per 36 minutes and per 40 or 48 — and the specific, one-directional way a per-36 projection misleads when it is applied to a reserve.
Any counting statistic per game and per 36 minutes
Total ÷ minutes × 36A rotation player: 420 points in 1,116 minutes across 62 games
Formula, and what the 36 is
Per game = total ÷ games played
Per 40 = (total ÷ minutes) × 40, and per 48 = (total ÷ minutes) × 48
- 36 is a convention, not a rule
- no governing body’s game is 36 minutes long. FIBA’s Official Basketball Rules 2024 give “4 quarters of 10 minutes each” at Art. 8.1, a 40-minute game; the NBA’s Rule 5 Section II-1 states that “All periods of regulation play in the NBA will be twelve minutes”, a 48-minute game. 36 was adopted because it is roughly what a heavily used starter plays
- it is a rate and behaves like one
- double the total and the minutes together and it does not move; halve the minutes alone and it doubles; set the minutes to zero and it refuses. Those three are the test of any per-minute figure, and this page is asserted against them
- the direction of the error
- one way, which is why the page exists. A reserve’s minutes are the easier minutes — against other reserves, in the stretches where the game is not being decided, before fatigue and foul trouble arrive. Tripling them does not triple the production at the same rate, so a per-36 figure flatters the player with few minutes and almost never the other way round
- what could not be established
- no published study quantifying how per-36 projections fared for players whose minutes actually did increase was found while this page was written. The direction of the effect is widely described; the size of it is not printed here, because a figure would imply a precision this page does not have. Nor does any governing body set a minimum minutes total below which the figure should not be quoted, so none is printed — statistics sites that rank by per-36 impose their own and state it
Worked example
A rotation player: 420 points in 1,116 minutes across 62 games
Minutes per game: 1,116 ÷ 62 = 18.0. Per game: 420 ÷ 62 = 6.77 points
Per 36: 420 ÷ 1,116 × 36 = 13.5 points, almost exactly double the per-game figure, because the minutes are almost exactly half of 36
Per 40, a FIBA game: 15.05. Per 48, an NBA game: 18.06. Same player, three bases, three numbers
Double both the total and the minutes and the per-36 figure is unchanged at 13.5, which is the check that it really is a rate
Halve the minutes to 558 with the total unchanged and it doubles to 27.1 — correct, and also the warning: the smaller the denominator, the wilder the projection
What the 13.5 is not: a prediction that this player would score 13.5 a game as a starter. It is what their current rate would produce at a starter's workload, and the rate would change
The same 420 points, five normalisations
| Normalisation | Figure | What one unit of the denominator is |
|---|---|---|
| Per game, over 62 games | 6.77 | One appearance, however long |
| Per minute played | 0.376 | One minute on the floor |
| Per 36 minutes | 13.55 | A convention for a starter’s workload; nobody’s rule |
| Per 40 minutes | 15.05 | A FIBA game: four quarters of 10 minutes, Art. 8.1 |
| Per 48 minutes | 18.06 | An NBA game: four periods of twelve minutes, Rule 5 Section II-1 |
What a per-minute rate does and does not survive
| Test | Expected | Why it matters |
|---|---|---|
| Double the total and the minutes together | The figure does not move | This is what makes it a rate, and the reason a substitute can be compared with a starter |
| Halve the minutes, leave the total | The figure doubles | The same property read as a warning: a small denominator makes a large and unreliable number |
| Set the minutes to zero | No figure at all | A rate with no denominator does not exist. A zero here would be a different and false claim |
| Give the player three times the minutes in reality | The figure falls | The one the arithmetic cannot do. The rate is not a property of the player independent of their workload, and treating it as one is the error this page is written against |
Per 36 is a projection, and it is wrong in a known direction
This page is about basketball, the five-a-side game played to a basket ten feet above the floor. It has no single rule-maker: the Fédération Internationale de Basketball (FIBA) publishes the Official Basketball Rules used by most of the world, the NBA publishes its own rulebook for its own competition, and the NCAA publishes separate rules for United States college play. Where a number on this page is one of theirs rather than plain arithmetic, the body and the edition are named beside it.
A raw total answers the wrong question when playing time differs, and in basketball it differs enormously: a rotation player may be on the floor for fifteen minutes and a starter for thirty-eight in the same game. Dividing by minutes fixes that, and multiplying back up by 36 puts the answer on a scale a reader recognises. A basketball rule belongs to a competition and to a season. The NBA plays “periods of regulation play … twelve minutes” (Official Rules, Rule 5 Section II-1) where FIBA plays “4 quarters of 10 minutes each” (Official Basketball Rules 2024, Art. 8.1), so a 48-minute and a 40-minute game are both normal and a figure normalised to one does not compare with one normalised to the other. 36 belongs to neither, which is why this page also prints the per-40 and per-48 figures.
Per 36 minutes is a projection, not a measurement. A per-game average describes what happened; scaling a bench player’s production to 36 minutes describes what would have happened if their rate had held while their workload roughly tripled, against starters rather than other reserves — and it would not have held. The direction of the error is known and it is one way: per-36 flatters the player with few minutes, because the minutes a reserve gets are the easier ones and because fatigue, defensive attention and foul trouble all scale against them. No published study quantifying this for players whose minutes actually increased was found while this page was written, so no figure for the size of it is printed.
What that means in practice. A reserve scoring 420 points in 1,116 minutes is at 13.5 points per 36. That is not a forecast that they would score 13.5 as a starter, and it is not evidence that the coach is wrong to play them eighteen minutes: it is the rate achieved in the minutes they have had, projected onto a workload they have not. The projection earns its place through the comparison it makes possible — this reserve’s rate against that starter’s, which raw totals cannot compare at all — and stops being useful the moment it is read as a prediction. The panel’s own arithmetic shows that halving the minutes doubles the figure, which is correct and is also the problem.
Two further cautions. No published minimum number of minutes exists below which a per-36 figure should not be quoted, so this page sets none and names the gap instead; sites that rank by per-36 impose their own and say what it is. And the figure knows nothing about the quality of the minutes or the opposition. No governing body publishes a good-or-bad threshold for this figure, and none is invented here. It is interpretable only against a position, a competition and a season, so the honest use of it is comparing players in the same role in the same competition, from one source. For efficiency rather than volume use the true shooting percentage calculator or the effective field goal percentage calculator; for a team rate normalised by possessions, the possessions and pace calculator; and for the period lengths above with their article numbers, the three-point line and shot clock interpreter.
Frequently asked questions
How do you calculate a per-36-minutes figure?
Divide the total by the minutes played and multiply by 36. For 420 points in 1,116 minutes: 420 ÷ 1,116 × 36 = 13.5 points per 36 minutes. The same arithmetic works for rebounds, assists, steals, blocks or turnovers.
Why 36 minutes and not 40 or 48?
Because it is roughly a heavily used starter’s workload, and for no other reason. No rulebook mentions 36 minutes: FIBA’s game is four quarters of 10 minutes under Art. 8.1 of the Official Basketball Rules 2024 and the NBA’s four periods of twelve under Rule 5 Section II-1. This page prints per 40 and per 48 too.
Is a per-36 figure a prediction of what a player would do as a starter?
No, and treating it as one is the commonest misuse. It is the rate already achieved in the minutes already played, projected onto a workload the player has not had. A reserve’s minutes are the easier minutes — against other reserves, before fatigue and foul trouble — so the projection flatters the low-minutes player, in one direction, almost always.
How many minutes are enough for a per-36 figure to mean anything?
More than you would like, and nobody publishes a number. Halving the minutes doubles the figure, so a player with 120 minutes can be made to look like anything. Sites that rank by per-36 set their own minimum and state it; if you are comparing two players, pick one and apply it to both.
Can I compare a per-36 figure with a per-40 one?
Not directly — they differ by a ninth for the same player, and the panel prints both so the conversion is visible rather than assumed. The same applies across competitions: a FIBA 40-minute game and an NBA 48-minute game give different per-game totals for identical players, a regulation difference rather than a performance one.
Related calculators
References
- Fédération Internationale de Basketball. Official Basketball Rules 2024, v1.0 — approved at Mies, 26 April 2024, “Valid as of 1st October 2024”; read 12 Oct 2026. Art. 2.1 (the 28 m by 15 m court); Art. 2.5.4 (the 6.75 m three-point arc and the 0.90 m sideline setback); Art. 8.1 and 8.7 (four 10-minute quarters, five-minute overtimes); Art. 16.2.1 (the 1, 2 and 3 point values); Art. 29.1.1 and 29.2.2/29.2.3/29.2.7 (24 seconds, reset to 14). The court diagrams returned as captions only, so no figure here is from one.
- National Basketball Association. Official Rules, Rule No. 5 Section I items 2-3 (two points inside the arc, three outside) and Section II items 1-2 (“All periods of regulation play in the NBA will be twelve minutes.”, five-minute overtimes); Rule No. 7 Section I (“The shot clock shall start at 24 seconds”) and Section IV-d-1 (reset to 14 seconds after a missed attempt “that contacts the basket ring”). official.nba.com, read 12 Oct 2026.
- National Collegiate Athletic Association. 2025-26 NCAA Men’s Basketball Rules, dated August 2025, read 12 Oct 2026; its preface states that “All NCAA member institutions are required to conduct their intercollegiate contests according to these rules”. Rule 1 Section 7 Art. 1 (the three-point line); Rule 2 Section 11 Art. 1 (“Use a 30-second shot clock.”); Rule 2-11.6.d.1 and d.3 (reset to 20 seconds). The fetched text truncated in Rule 4, so Rule 5 — the halves — was not read and is printed on no page here.