The Arithmetic of 23-0: When the Box Score Argues With Itself

The first thing I did after reading about Detroit's 23-0 run was put the scoreboard on paper. 7:34 left, Detroit down 99-95. The next line has Detroit ahead 107-99. Then the text says the run eventually reached 23-0. And with 2:00 left the lead is 114-105. A genuine 23-0 run from 95 points puts Detroit at 118-99. The 107-99 and 114-105 figures sit somewhere else entirely. Those three numbers cannot all be true at once. And when the most dramatic sentence in a report does not reconcile with that report's own scoreboard, every other line has to be re-examined.

Opening-night recaps are the fastest-written and least-checked pieces of the calendar. The first game of a season arrives loaded with feeling, and arithmetic gets left behind in the feeling. At the 2026 NBA India Games in Mumbai I coded 392 team possessions across the two Pacers-Kings games — Indiana won 132-131 and 130-106 — and filed the Game 2 chart within six hours. The rule then was the same rule now: any number carrying the weight of the argument gets triangulated against a second source — video, a contemporaneous wire report, a second scoresheet. This report fails that test in three places.

The first failure is a name. The report states that Paul George scored a game-high 38 points in his Celtics debut and that it was the most ever scored by a Boston player in a first game with the team. In the established NBA record, Paul George has never been a Boston Celtic. This is not a small slip; it is the kind of error that arrives with full precision and full confidence and zero fact behind it. Duncan Robinson appearing as a Detroit Piston belongs in the same category — unverified. The second failure is the arithmetic this piece opened with: 95 to 107 to 114 does not accommodate a 23-0 run. The third is the heaviest — across all 55 information points in the document, not one source is recorded. No named reporter, no box-score citation, nothing.

Now the basketball that survives, rebuilt at the resolution the record can actually support. Detroit shot 38.6% from the field and 22.2% from three in the first half; Boston shot 48% and 40%. Detroit's third quarter was 37-30. Javonte Green did not play in the first half and finished with 19 points. Cade Cunningham had 22 points and 8 assists, Jalen Duren 22 and 9, Duncan Robinson 21 on five threes, Ausar Thompson 11 and 11. Four Detroit players landed between 19 and 22. On the other side, Jayson Tatum's line reads 13 points, 11 assists — and 6-for-23 from the field, 0-for-7 from three.

A portability note is required here. The report supplies no true-shooting percentage, no usage rate, no net rating for any player. When field-goal attempts and minutes are all you have, an efficiency verdict can only be directional, never declarative. Calling Tatum's 0-for-7 a bad night sits inside the record; calling it a decline sits outside it. Across the matches I have watched with a possession log over eleven years, one rule has held — a single game's efficiency never tells you who a player is, only what the weather was that night.

Now the part that deserves the most scrutiny. Assume all of it is true. Even then, this is the least repeatable category of win. A bench player suddenly running hot while the opposing team goes cold from three is variance, not scheme. Runs of 23-0 are built this way and they break this way; Detroit's three-point percentage will drift back toward its mean over the next ten games. Green going scoreless in the first half and then posting 19 is evidence of a coach's instinct and also evidence that the rotation was unsettled on opening night.

For Boston, the real question is not the fourth-quarter collapse. While charting 318 possessions through India's 2026 Asia Cup campaign I built the possession-tax argument, and the lesson there was that the final five minutes must be separated from the final five minutes that mattered. Boston's lead here was inflated from the start — George's three threes in under a minute pushed the score up, and when that shooting regressed there was no second route. When the primary engine stalls at 6-for-23 and no secondary creation appears, the loss is structural, not clutchtiming.

So what to watch. Detroit's three-point percentage over the next ten games — if it regresses, this night was an outlier. Tatum's field-goal percentage over the next five — sustained sub-40% becomes an injury or form question rather than one cold evening. And Boston's clutch net rating — final five minutes, margin five or fewer. But before any of that, one task remains: open the roster on NBA.com or Basketball-Reference and check whether Paul George is in Boston at all. I counted what the empty press box was still saying, and all it was saying was that the source was nowhere. The possession map did not lie; the crowd did. Until a scoreboard reconciles with its own arithmetic, a story stays a story — not a record.