The chances of a game-winning drive: How likely are teams to score in the final moments, and how should that affect coaching decisions?

The divisional round of the 2021 NFL playoffs gave us the best weekend of playoff football in a very long time.

The offensive masterclass on show in the Kansas City Chiefs–Buffalo Bills game served as the highlight, but the weekend as a whole featured three extremely close games and another game that somehow became close in the final moments.

Every single divisional-round game came down to the wire, with the winners all taking the lead on the very last play of the game. This also means that teams had tremendous success on late, high-leverage drives.

To summarize, here is what happened this past weekend:

– The Cincinnati Bengals got the ball with 20 seconds left at midfield and advanced 20 yards for a game-winning field goal.

– The San Francisco 49ers got the ball with 3:20 left, gained enough yards and first downs to milk the clock and score a walk-off field goal.

– The Tampa Bay Buccaneers got the ball with 2:25 left at the Los Angeles Rams’ 30-yard line and tied the game with a touchdown. However…

– The Rams got the ball with 42 seconds left and advanced to the 12-yard line to win the game with a last-second field goal.

Click here for more PFF tools:

Rankings & Projections | WR/CB Matchup Chart | NFL & NCAA Betting Dashboards | NFL Player Props Tool | NFL & NCAA Power Rankings
Best Bets Tool

And we all know what happened in the Bills-Chiefs classic that featured five scoring plays that either secured a lead or tied the score after the fourth-quarter two-minute warning.

Yes, that’s five scores in a row on what were essentially do-or-die drives. The Ringer's Benjamin Solak probably summarized it best.

In review pic.twitter.com/UeTZ0pmOif

— Benjamin Solak (@BenjaminSolak) January 24, 2022

But how much substance does the classic saying, “Team X left too much time” really have?

Are scoring drives in one-score games inevitable, especially for good offenses? And what does this mean for the teams that have the ball late in a game?

Just last weekend, teams either tied the game or took the lead a combined nine times inside the two-minute warning or overtime. Some of those came in games where the offense didn’t get much going overall (Bengals, Buccaneers and 49ers) and others came from offenses that were essentially cooking the whole game (Chiefs, Bills and Rams).

We have to note that we also saw at least two failed attempts at game-winning drives, as both the Tennessee Titans (2:43 left) and Green Bay Packers (4:41 left) had the opportunity to win the game with a sustained drive and a last-second field goal. Still, 9-of-11 is an insanely good rate in a league in which scores happen on less than 40% of drives.

The sample here is relatively small, but there are other examples of the “left too much time” theory from this season. For example, there are the Detroit Lions‘ early-season heartbreak losses against the Baltimore Ravens and Minnesota Vikings, or even the regular-season matchup between the Packers and 49ers when Aaron Rodgers led his squad to a game-winning field goal within the final 37 seconds.

To widen the sample, we will look at all “do-or-die” and “do-or-tie” (i.e., potential game-winning drives in a tied game) situations in recent years and see how offenses fared in these compared to “usual” drives.

Are teams more likely to score when they need to score?

First of all, we look at four different drive situations.

The first is a “usual drive,” which is basically any drive, but we only consider plays where the offense is actively trying to score. We remove all drives that contain a kneel-down.

The second situation is a drive late in the second quarter, starting with 1-3 minutes left on the clock.

The third situation is a late fourth-quarter drive when the team needs a field goal (trailing by 3, 2, 1 or 0 points). The fourth situation is the same only the offense needs a touchdown (trailing by 4-8 points). In both cases, we filter to drives that started with 1-3 minutes left on the clock.

In all four situations, we filter to drives that start in the offense's own territory.

The following table shows how often teams scored since 2018: