Your 5x5 Basketball Stats Lie in a 3x3 Game — Here's the Maths
August 2, 2026
A coach I know finished his first 3x3 tournament, opened the stats app he had used all winter, and concluded that his team had shot appallingly. Points per shot: 0.58. In the 5x5 season just gone, anything under 0.90 had meant a bad night.
His team had actually shot fine. The app was measuring a game it did not know it was in.
This is the quiet problem with 3x3 analytics. The rules are easy to learn. The numbers built on top of the rules are the part that breaks, and they break silently — no error message, just figures that look plausible and point the wrong way.
The one change that moves everything
In 3x3, a basket from inside the arc is worth one point and a basket from beyond it is worth two. That is it. That is the whole change.
But almost every efficiency statistic in basketball is a ratio with points on top. Change what a basket is worth and you have changed the numerator of nearly every number you rely on.
Here is what actually shifts.
Points per shot: the benchmark halves
Points per shot is exactly what it sounds like — points from the field divided by field goal attempts. It is the most useful single number in shooting analysis because it folds accuracy and shot value into one figure.
In 5x5, roughly 1.00 is average. Above 1.15 is a shot worth hunting.
In 3x3, the average is roughly 0.55, because the maximum any shot can return has halved. Above 0.75 is the equivalent of “worth hunting.”
Nothing is wrong with a 3x3 team at 0.58 points per shot. They are slightly above average. Read on a 5x5 scale, the same number looks like a catastrophe.
Effective field goal percentage: the weighting changes
eFG% exists to fix the unfairness of treating a three like a two. The 5x5 formula is:
eFG% = (FGM + 0.5 × 3PM) ÷ FGAThe 0.5 is there because a three is worth 1.5× a two — half a basket more.
In 3x3, a basket from the arc is worth 2× an inside basket, not 1.5×. The bonus is a full extra basket, not half of one:
eFG% (3x3) = (FGM + arc makes) ÷ FGAWhich, conveniently, is just points from the field per attempt expressed as a percentage. A 3x3 eFG% of 60% means 0.60 points every time that player shoots.
And note something that trips people up: in 3x3 this figure can exceed 100%, and it is correct when it does. A player shooting well from the arc genuinely produces more than one point per attempt. An app that caps it at 100 or flags it as an error is wrong.
True shooting: the factor of two disappears
True shooting percentage includes free throws. The 5x5 formula is:
TS% = PTS ÷ (2 × (FGA + 0.44 × FTA))That leading 2 is there because a field goal is worth two points — it converts “points” into “shot equivalents.”
In 3x3 the baseline basket is worth one point. The 2 has no business being there:
TS% (3x3) = PTS ÷ (FGA + 0.44 × FTA)Leave the 2 in and every player’s true shooting reads at half its real value. That is the single most common error I have seen in spreadsheets people have built themselves for 3x3.
(The 0.44 free-throw coefficient is an empirical 5x5 estimate of how many free throw attempts end a possession. In 3x3, where a shooting foul inside the arc gives one shot and beyond it gives two, it is a rougher approximation. It is close enough to be useful and worth knowing is approximate.)
Double-doubles do not exist
A 5x5 double-double means ten or more in two categories. In a game that ends at 21 points total for the whole team, a single player reaching ten points and ten rebounds is essentially impossible.
If you want the same idea — a player who contributed meaningfully in two different ways — the line has to move. Five is the honest equivalent: five or more in two of points, rebounds, assists, steals and blocks. It happens often enough to mean something and rarely enough to be worth noticing.
Any app that still highlights ten-and-ten in a 3x3 game will simply never highlight anything.
Per-game averages need a different mental model
A 5x5 starter scoring 11 points a game is solid. A 3x3 starter scoring 11 points a game has scored over half his team’s total.
Rough anchors from real 3x3 games:
| Stat | Typical 3x3 starter, per game |
|---|---|
| Points | 4–7 |
| Rebounds | 2–4 |
| Assists | 1–3 |
| Minutes | 7–9 |
| Team total | 15–21 |
If you are used to reading a 5x5 box score, halve everything in your head before judging it. A player at 6 points and 4 rebounds in a ten-minute game has had an excellent night.
What actually predicts winning in 3x3
Shorter game, fewer possessions, more randomness. That has two consequences worth planning around.
Single games tell you very little. A ten-minute game with maybe forty possessions has enormous variance. One hot stretch from the arc decides it. Judging a player on one 3x3 game is like judging a 5x5 player on one quarter — which nobody would do, but people do it constantly in 3x3 because the game feels complete.
You need volume. A tournament day of five games is roughly one 5x5 game’s worth of possessions. That is the unit worth analysing, not the individual result.
Efficiency beats volume, harder than in 5x5. With so few possessions, a turnover is proportionally more expensive and a wasted possession on a low-value shot costs more. The teams that win consistently are the ones taking arc shots at a decent rate and not giving the ball away — not the ones with the most spectacular individual scorer.
Track the offensive glass. Because an offensive rebound resets the shot clock to a full twelve seconds, it is worth more in 3x3 than the 5x5 equivalent. It is a full new possession, not a scrap of one.
The practical version
If you keep stats on a spreadsheet, three fixes get you most of the way:
- Make sure points = inside makes × 1 + arc makes × 2 + free throws.
- Drop the ×2 from true shooting, and change the 0.5 in eFG% to 1.
- Rewrite your benchmarks: 0.55 points per shot is average, not 1.00.
If you use an app, check one thing before you trust anything it tells you: tap in a made shot from beyond the arc and see whether the score moves by two or by three. If it moves by three, every efficiency number that app gives you for the rest of the season is wrong.
HoopNest 3x3 has all of the above rebuilt rather than carried over — eFG% counts an arc basket double, true shooting drops the factor of two, the shot chart benchmarks against 0.55, and the double-five replaces the double-double. It is an offline single file with no account and no subscription. Free 15-game trial. If you coach both formats, there is a bundle with the 5x5 version.