Passing, rushing, and field effects
Rushing versus expectation
How game rushing output and provider week RYOE or box signals may be compared only at their stated scopes.
Precise definition
Rushing versus expectation compares game rushing output with provider week expected rushing yards or related box signals only at their stated scopes. Model inputs and week aggregation prevent assigning causal credit to the runner or blockers alone.
Why it matters
Rushing yards and attempts describe what happened in the game; week expectation or box metrics are separate signals and do not authorize blocking-scheme or front claims.
How it works
Game attempts and yards describe what happened
Official rushing attempts, yards, and success-style rates when sourced are game-level outputs. They do not authorize blocking-scheme or front claims.
Expected rushing yards are a separate model signal
NFL Next Gen Stats expected rushing yards and rushing yards over expected compare carries to a model baseline. Those are model-derived and often week-aggregated publicly.
Box signals stay inside provider scope
When a provider publishes defenders-in-the-box style signals, they remain scoped tracking or charting inputs—not proof of the front called or the blocking scheme used.
Observable signals
Game rushing attempts and yards
What it can mean: The contest produced a countable rushing output.
Caution: Yards do not assign credit between runner, blockers, and defensive fit.
Week RYOE or boxed-run signals at the provider's stated scope
What it can mean: Across the week sample, carries gained more or less than the model expected under its inputs.
Caution: Model inputs and week aggregation prevent causal credit assignment to runner or blockers.
Two carries with similar stated conditions and different gains versus expectation
What it can mean: Even under similar labeled conditions, results can diverge from the model baseline.
Caution: Similar stated conditions are not identical true situations; missing inputs remain unknown.
Illustrative example
Illustrative two carries versus provider expectation
Illustrative teaching sequence, not a real game. Neutral offense; two early-down carries with similar stated down, distance, and box labels in a provider view.
Step 1
Carry one exceeds expectation
Observation: A four-yard gain is recorded on first-and-10; a provider expected-rushing-yards figure for similar conditions sits lower than four.
Interpretation: The residual is model context for that carry class—not proof the runner or blockers 'won.'
Step 2
Carry two falls short
Observation: A one-yard gain on a similarly labeled second-and-6 sits below the provider expectation for that class.
Interpretation: The same comparison framework now shows a negative residual without explaining the missing inputs.
Step 3
Scope separation
Observation: Game rushing totals add both carries; any week RYOE table covering the runner remains a weekly aggregate.
Interpretation: Useful comparison keeps game outputs and week model signals explicitly separated.
Takeaway: Compare rushing to expectation only inside the provider's stated scope; do not assign the residual to one causal actor.
Limitation: This invented pair cannot recover true box, blocking, or front details, and week aggregation still limits single-game claims.
Common misinterpretations
Mistake: Assigning the entire RYOE result to the runner or to the blockers.
Better reading: Treat RYOE as a model comparison at stated scope; do not hand causal credit to one side from the residual alone.
Mistake: Mixing week expectation metrics into a single-game rushing grade without relabeling scope.
Better reading: Keep game yards and week RYOE on their own denominators; compare only within each stated window.
Metrics and limitations
- Attempts, yards, and success rate
Measures: Game-level rushing attempts and yards, plus definition-based success rate when sourced.
Useful for: Describing rushing output inside the contest.
What it misses: Does not authorize blocking-scheme or front claims.
- Expected rushing yards and RYOE
Measures: Model-derived expected rushing yards and yards over expected, often published as week aggregates by Next Gen Stats.
Useful for: Comparing carries to a provider baseline at the stated scope.
What it misses: Model inputs, formula versions, qualifying samples, and week aggregation prevent causal credit assignment.
- Box signal if provider-scoped
Measures: Defenders-in-the-box or similar charting/tracking signals when the provider states their scope.
Useful for: Adding a labeled situational input beside the rush result.
What it misses: Provider-scoped only; not proof of the front called or the blocking scheme used.