Second-guessing an NFL coach on whether he should have gone for it or kicked on fourth down is one of the great pastimes of watching the sport. Now the league is turning to Amazon Web Services and Next Gen Stats to better inform the armchair quarterbacking by fans and broadcasters.
AWS and the NFL announced a new stat called NGS Decision Guide on Wednesday that will be used during the 2021-22 season.
Using a series of machine learning models, the equation focuses on win probability — how much the game will change in the hypothetical event of each outcome — and conversion probability — the likelihood of the offense converting a fourth down or 2-point conversion.
Broadcasters can show fans which call makes the most sense based on data before the play happens or break the decision down in replay analysis.
The NFL and AWS are planning more new advanced stats for the upcoming season including:
- Quarterback expected rushing yards: The Expected Rushing Yards model of 2020 was designed to estimate how many rushing yards a ball carrier will gain from the moment of handoff. This year’s QB metric analyzes plays without a handoff, using a snapshot when the quarterback makes clear his intention to run.
- Quarterback dropback type: Analyzing whether a quarterback does a straight drop, rolls out of the pocket, is forced to scramble or engages in a designed run will help to more accurately classify a play call’s intended play type and offer more robust analysis of run-pass play-calling tendencies.
- Next Gen Stats big play score: Watching the biggest plays of the week or season is the fun of watching football. This new model grades every play on a score from 0 to 100 based on three primary components: win probability effect; points added; and play improbability.