We built a field-position stat. A simpler one beat it
We measured the yard-line head start each team gives itself, then tested it against seasons it had never seen. It adds a sliver to our read on a team — but plain win rate and moving the ball on schedule both call next season better, and most of what makes our stat unique is just luck.
NR
The Numbers Room
Ratings & power ·
6 min read
Here is a stat that does something most of our numbers can't: it tells you not who is good right now, but who is about to get better. The field-position gap — the difference between where your own offense keeps starting its drives and where you pin the other team to start theirs — held up when we tested it against seasons it had never seen, as an early tip on where a team is heading. Take two teams that look equally strong today: the one winning the field-position battle is the one we'd bet climbs. That's the real find. The flashier story — that this stat also fingers lucky teams about to fall back to earth — is one our numbers can't actually back yet, and we'll say so plainly.
decent, not great
Field-position gap, on its own
as a tip on next year's record
win rate, and moving the ball on schedule
But these tell you more
both call next season better
a real sliver
What it adds to our team read
info our power read didn't already have
earned part sticks, lucky part doesn't
Its unique part is mostly luck
the bounces don't carry into next year
It adds something, but it's the junior partner
Team
Field-position gap (yds)
Did they earn it on the field?
Miami (OH)
+14.6
No — slightly below water moving the ball
Indiana
+7.5
Yes — strongly on schedule
Notre Dame
+7.3
Yes — clearly on schedule
Tulane
+6.7
Yes — on schedule
UNLV
+10.3
Some — modestly on schedule
Iowa
+8.7
Barely — about break-even
2024 leaders in the field-position gap, and how much of it they actually earned by moving the ball.
What we measured
How well it tipped next year's record
Beats win rate?
Moving the ball on schedule (vs. the other guy)
Best of the bunch
Yes
Plain win rate
Nearly as good
—
Our read on how good they really are
About the same as win rate
No
Field-position gap
Decent, but a clear notch below
No
How well each number, on its own, tipped next season's win rate when tested against seasons it had never seen (more than 1,200 team-seasons). The simpler stat — moving the ball on schedule — beat our shiny new one.
The honest framing matters, because the field-position gap is not the best crystal ball on the board — it's a useful supplement. Standing alone, it does a decent job of tipping next season's win rate, but a team's plain record tells you more, and so does our own read on how good a team really is. More pointed still: the single best tip comes from how often a team stayed on schedule, snap after snap, moving the ball better than the other guy — that one actually edges out the record itself. The field-position gap trails all three. Its claim to a roster spot is narrower, but real: bolt it onto our power read and it nudges our handle on next year's record up by a sliver. It's information our power read didn't already hold — but it's a trim, not a transformation.
Where it earns its keep: spotting the teams on the rise
This stat is at its best not as a ranking but as an early-warning light for which way a team is trending. Hold our current read on a team fixed, and the field-position gap still points the way toward next year's quality. In plain terms: take two teams we grade dead even today, and the one ahead in the field-position battle is the one trending up. It's a gentle pull — one yard-line stat is not a whole season's fortune — but it's a signal you can't get from the standings, because the standings already tell you how good a team is now, not which direction it's pointed.
Earned vs. lucky — read it as bookkeeping, not a verdict
Pull the stat apart and a clean pattern shows up. The chunk you earned — field position you bought by moving the ball and blowing up the other guy's plays — both predicts next year's quality and shows up again season after season; more than half of it carries over. The leftover chunk — lucky bounces, kick-return touchdowns, short fields handed to you by turnovers — barely predicts anything and barely repeats; only about a fifth of it sticks from one year to the next. That's why the field-position gap tracks so closely with moving the ball on schedule, and leans on a defense that lives in the backfield: it's mostly downstream of the parts of a team that actually carry over.
But here is the correction the numbers force on us. That earned-vs-lucky split is a big-picture observation about what the stat is made of — it is NOT a proven way to call any one team's next season. When we sorted the high-field-position teams into those who earned it and those who didn't, and then checked what happened next, the two groups looked statistically the same once we tested it honestly: both gave back almost exactly the same amount of field position the following year. The mechanism is real in the aggregate; the verdict on any single team is not.
Which means the tempting headline — that Miami (OH) led the country in 2024 with a whopping +14.6-yard field-position edge despite being slightly below water at actually moving the ball, so it's due to fall back — is a hunch our numbers can't check. The data stops at 2024; there's no 2025 to test it against. Indiana (+7.5) and Notre Dame (+7.3) plainly earned their edges by staying on schedule, where Miami (OH) borrowed most of its from special teams, and that contrast is worth pointing out. But calling it a sure regression is a story we'd be telling, not one we've proven.
One hard limit, stated straight instead of hedged: we only have these field-position numbers as season-long totals, so we can't account for who a team faced. A team that piled up its edge against weak special-teams units can't be told apart from one that earned it against good ones — and we have no number to put a fence around that gap. So treat the field-position gap as a genuine but junior tool: it adds something our power read doesn't have, it's sharpest as an early read on who's rising, and it should always be read next to the stronger number — how often a team stayed on schedule, which is doing most of the heavy lifting here.
Gridpex's desks are model-driven, AI-assisted columns. Every figure is generated from our own data and ratings — not invented. We don't fabricate reporters, quotes, or sources. Published Wed, Jun 24, 2026 · research-lab:claude+dejargon.
Discussion
Weigh in on the analysis — the best takes rise to the top.
Discussion
Weigh in on the analysis — the best takes rise to the top.
0 Replies
Sign in to join the discussion.