This season so far: 2026, through October 11, 2026

The same nine questions as the five-season sheet, asked of this season alone: weeks 1-5, through the games of October 11, 2026. The groups are small and the tool says so wherever it matters.

Too few games to rule

11 team-games in each group of games close at halftime so far; one game moves a win rate by 9.1 points. In those games the early gap is -9.1 points; across all games it is 15.8 early and 68.4 late. The sample cannot carry a ruling yet. Ask again in a few weeks.

15.8points early: the first-half gap

-9.1points early, games within 3 at halftime

68.4points late: the second-half gap

The verdict follows a rule written in season-sheet.mjs, so you can disagree with it. It reads the first half of games within 3 points at halftime: under 100 team-games per group is too few to rule; a gap under 5 points does not hold as stated; a gap of 10 or more holds in part; between the two is mixed. This is the page's rule, not the agent's. Every number is the tool's.

The whole game

Every rush and pass, start to finish, overtime included.

node ./query-plays.mjs --seasons 2026

13.2%

73.7%

73.7% ran most, 13.2% ran least, a gap of 60.5 points. 38 team-games in each group, 78 games.

First half only

"Establish" means early.

node ./query-plays.mjs --seasons 2026 --half 1

39.5%

55.3%

55.3% ran most, 39.5% ran least, a gap of 15.8 points. 38 team-games in each group, 78 games.

Second half only

Who is running now?

node ./query-plays.mjs --seasons 2026 --half 2

15.8%

84.2%

84.2% ran most, 15.8% ran least, a gap of 68.4 points. 38 team-games in each group, 78 games.

First half, games within 3 points at halftime

Take away who was ahead at the break.

node ./query-plays.mjs --seasons 2026 --half 1 --halftime-margin 3

54.5%

45.5%

45.5% ran most, 54.5% ran least, a gap of -9.1 points. 11 team-games in each group, 22 games.

Small sample: 11 team-games per group. One game moves a win rate by 9.1 points.

First half, score within 8 points

Early, and one score apart.

node ./query-plays.mjs --seasons 2026 --half 1 --margin 8

48.6%

54.1%

54.1% ran most, 48.6% ran least, a gap of 5.4 points. 37 team-games in each group, 78 games.

Second half, score within 8 points

Late, but still a game.

node ./query-plays.mjs --seasons 2026 --half 2 --margin 8

36.0%

68.0%

68.0% ran most, 36.0% ran least, a gap of 32.0 points. 25 team-games in each group, 66 games.

Small sample: 25 team-games per group. One game moves a win rate by 4.0 points.

Second half, score within 3 points

Late, and a field goal apart.

node ./query-plays.mjs --seasons 2026 --half 2 --margin 3

53.8%

61.5%

61.5% ran most, 53.8% ran least, a gap of 7.7 points. 13 team-games in each group, 50 games.

Small sample: 13 team-games per group. One game moves a win rate by 7.7 points.

Regular season only, first half

Leave the playoffs out.

node ./query-plays.mjs --seasons 2026 --type REG --half 1

39.5%

55.3%

55.3% ran most, 39.5% ran least, a gap of 15.8 points. 38 team-games in each group, 78 games.

Playoffs only, the whole game

Thirteen games a season, 65 in five.

node ./query-plays.mjs --seasons 2026 --type POST

The filters left fewer team-games than groups, so the tool gives no result. That is an answer too.

No result: fewer team-games than groups.

Not on this sheet: why. Every gap here is an association between running and winning in the plays kept. It does not say which one causes the other. Teams that are ahead run to use the clock, so winning can cause running.

What the agent ruled on air

The claim: In games tied or close at halftime (within 3 points), do teams that ran the most in the first half win clearly more often than teams that ran the least?

Five seasons (2021-2025, regular season and playoffs):

First half, games within 3 points at halftime: ran most 46.3% vs ran least 55.3%, gap -9.0 points, 188 team-games per group.

First half, all games: ran most 53.0% vs ran least 46.1%, gap 6.9 points, 710 team-games per group.

Second half, all games: ran most 87.0% vs ran least 11.7%, gap 75.4 points, 710 team-games per group.

The pattern goes the wrong way for the claim. When the halftime score is close, the teams that ran most early won less often. The small positive gap across all first halves fits teams running more because they were already ahead. The large second-half gap is what you would expect from teams with a lead running out the clock.

This season (2026, weeks 1-5, same filters):

First half, games within 3 points at halftime: ran most 45.5% vs ran least 54.5%, gap -9.1 points, 11 team-games per group.

Second half, all games: ran most 84.2% vs ran least 15.8%, gap 68.4 points, 38 team-games per group.

The tool flagged the halftime test as a small sample. With only 11 team-games per group, a single game moves a win rate by 9.1 points, so that gap of -9.1 is the size of one game.

The gap between the two answers: In close games, five seasons show -9.0 and this season shows -9.1. In the second half, five seasons show 75.4 and this season shows 68.4. Both pairs point the same way, and neither gap differs by more than this season's few games can support.

Ruling: Does not hold. Over five seasons, in games close at halftime, the teams that ran most in the first half won less often than the teams that ran least, not more. The strong link between running and winning shows up only in the second half, where the score at that point can explain it.

What these plays cannot tell you: These plays can't say whether early runs set up play-action, wore down the defense, or gained yards, and they can't say whether running causes winning.

Commands run:

node ./query-plays.mjs --seasons 2021-2025 --half 1 --halftime-margin 3

node ./query-plays.mjs --seasons 2021-2025 --half 1

node ./query-plays.mjs --seasons 2021-2025 --half 2

node ./query-plays.mjs --seasons 2026 --half 1 --halftime-margin 3

node ./query-plays.mjs --seasons 2026 --half 2

From examples/05-ruling.md.