READING THE BOARD
Basketball pace vs scoring efficiency: compare equal opportunities
More points do not necessarily mean better scoring efficiency. Compare how many scoring opportunities produced those points before ranking two basketball offenses. Even an equal-opportunity comparison describes the sample you measured; it does not, by itself, predict the next winner.
Start with the denominator
The NBA glossary defines pace as possessions per 48 minutes and offensive rating as team points per 100 possessions.
For the original arithmetic exercises below, imagine two fictional teams observed in separate samples. Every number is invented, and the possession counts are supplied assumptions rather than estimates from a box score. We are comparing each team's own scoring, not the two sides of one game.
A higher score can hide a lower rate
| Sample | Points | Possessions | Points ÷ possessions |
|---|---|---|---|
| Team A | 120 | 110 | 1.091 |
| Team B | 114 | 100 | 1.140 |
A has six more points, but B produces more points for each supplied opportunity: 114 ÷ 100 is greater than 120 ÷ 110. For an equal 100-opportunity arithmetic comparison, the rates correspond to about 109.09 for A and 114 for B. That is a rescaling of these fictional samples, not a predicted score for a future game.
Change the opportunities while holding the rate fixed
Now stipulate a completely separate toy scenario: a constant scoring rate of 1.10 points per opportunity. At 95 opportunities, multiplication gives 104.5 points; at 105, it gives 115.5. The 11-point difference comes entirely from the assumed opportunity count. Fractional values here are arithmetic scenario outputs, not possible final scores.
This exercise also exposes a forecasting gap. Choosing 105 opportunities instead of 95 requires evidence; writing a multiplication formula does not validate that choice. Neither scenario tells us what the opponent scores or how much results vary.
Make the comparison reproducible
- Retain the points and possession counts together, along with their source and observation date.
- State the games and time window included. Do not silently compare a short recent sample with an entire season.
- Record filters such as selected lineup or game segment, and use equivalent filters on both sides.
- Keep observed rates separate from assumptions about the next matchup. Label a rescaled historical number as a comparison, not a forecast.
What this means when reading a pick
A claim that one offense “scores more” is incomplete without the sample and denominator. Ask which quantity the claim actually supports: observed scoring volume, observed scoring per opportunity, or a separately tested prediction. One cannot be substituted for another just because all are expressed with basketball numbers.
Our daily board uses the experimental winner-selection method described on the site; this guide does not add a possession model to it. A projected score or scoring rate also cannot be relabeled as a spread-cover probability without a supported distribution of outcomes. Use the prediction-record guide to evaluate forecast claims.
NBA metric definitions checked October 10, 2026. All teams, counts and scenarios above are original hypothetical illustrations. No current team statistics, tested predictive advantage or NBA endorsement is claimed.