End of Game Scenario

Kick It or Go For It? Is There a Door Number 3?

Football coaching decisions, specifically when trailing at the end of the game, can be similar to a chess match. It is a pick your poison type of move where none of the options are desirable. With the rise of analytics, a coach's 4th down decision making has been a been a hot topic of decision between NFL experts, talk show hosts, and the general public alike. However, all three debate the same options: kick/punt it or go for it. Why not think outside the box and think about all the possibilities available to maximize the probability of your team winning. The first game of the 2023 NFL season provided an excellent opportunity to do just that.

...

The scenario, as seen above, was the Chiefs were trailing the Lions by one point facing a 4th and 25 from their own 30 yard line with 2:09 remaining and all of their timeouts left. This left the Chiefs in a very peculiar spot. Obviously, a FG here was not an option so they had to decide whether to punt the ball back the the Lions or convert on a very long 4th down, right? What about a 3rd option? Patrick Mahomes runs back to his own 9 yard line and kneels the ball. Sure this would give the Lions the ball with a great scoring opportunity, but it also may be the Chiefs best decision. Let's take a look at each decision: punt, go for it, and take the sack.

Scenario 1: Punt

By punting the ball back to the Lions, the Chiefs can most likely safely punt the ball 40 yards out of bounds to stop the clock prior to the 2 minute warning. This would result in the Lions having the ball on their own about 30 yard line with about 2:02 to go and, remember, the Chiefs had all 3 timeouts. Now the average punt play in the NFL in the 2023 season with no return averaged about 7 seconds, so this is a safe assumption. How do their chances look? With the 2 minute warning and 3 timeouts, they theoretically could allow a first down immediately and still have a chance to get the ball back. Best case scenario would be stopping the Lions and receiving the ensuing punt around 1:45 left and a timeout. Of course, if the Lions do get a first down (which they did) the game is essentially all but over.

I looked at the historical rate of teams coming back in the last 10 years given these parameters:

    - Trailing by 1
    - Opposing team possesses the ball
    - All 3 timeouts left
    - Less than 2:40 but more than 2:00 minutes left

To be honest, the numbers were a little surprising. Of the 10 occurrences I found, seven of the trailing teams got the ball back with an opportunity to tie or win the game. Although, with a sample size of only 10 games, this should not be considered the expected rate to get the ball back.

...

For what it's worth, the average win probability for the trailing teams was about 24%, according to NFLreadR. While only one team did complete the comeback, there were a few close calls. The Vikings infamously missed a game winning field goal vs the Seahawks in the wild card round of the 2015 playoffs.

Taking into account the above information, the 24% win probability does seem to carry some weight here, and the Chiefs do have the advantage of having Patrick Mahomes as their quarterback. However, given the Lions' excellent O-line and Dan Campbell's gutsy 4th down decision making, the Chiefs could not feel great about getting the ball back. In reality, the Lions did in fact run right through them for a first down to ice the game. Yet, overall, punting does not seem .....that.... bad given the circumstances.

Scenario 2: Go For It

In scenario 2, NFL teams converted roughly 6% of 4th and 21 or more from 2018 to 2022 (arbitrary date range but I had the data handy). If they convert, we'll assume they have the ball around the Lions' 42 with 2:00 left. If they fail, we’ll say the Lions have the ball at the Chiefs 30 with 2:02 left. Here are the analytical numbers according to Ben Baldwin's 4th down bot:

...

While conventional wisdom and win probability models agree, if the Chiefs were to convert, they'd be heavily favored to win the game, the bot agrees the likelihood of converting is only about 6% for the Chiefs. Overall, this is a fairly strong recommendation to punt the ball. It is not often a coach makes an error according to the analytical models when punt is the recommended decision, especially one this strong. For what it's worth, ESPN also disagreed with the Chiefs decision to go for it.

Going for it in this scenario does seem to be a little bit of wishful thinking.

Scenario 3: Take The Sack

Now what if instead of going for it or punting it away, Patrick Mahomes runs back to the Chiefs' own 9 yard line and takes a knee? He could easily accomplish this in 8 seconds or less to give the ball to the Lions with a first and goal on the 9 yard line with 2:01 left in the game. Armed with 3 timeouts and the two-minute warning, only two things could stop the Chiefs from getting the ball back with a chance to win or tie the game:

    - An automatic 1st down penalty allowing the Lions to run out the clock
    - The Lions scoring a TD and converting the 2-point conversion to ice the game
In the last 10 years I could only find one similar scenario. Back in week 6 of the 2020 season, the Texans had the ball with a 1st and goal on the Titans 9 yard line. NFLreadR gave the Titans a 16% win probability. The following ensued:

...

The worst case scenario almost happened. The Texans ran the ball three straight times to burn through the 2-minute warning and the Titans remaining timeouts. Then they elected to go for the TD and converted. Finally, they even tried to ice the game by going for the two-point conversion to take a two possession lead, but they failed. This gave the Titans 1:45 left to drive down the field to scoring the game-tying touchdown, and well, I think you know where this is going. With 4 seconds left, this happened.

After tying the game with the extra point, the Titans received the ball first. And of course the only example of this scenario has the perfect story book ending. The Titans immediately drove right down the field to win the game with a Derrick Henry 5-yard walk-off touchdown.

Its important to mention the Chiefs had all three timeouts as opposed to only two like the Titans. And despite the Texans taking all four downs to score, they still had plenty of time remaining for one final drive. Maybe the most important factor, though, is the Chiefs have the greatest quarterback on the planet. This scenario provides the best possibility of just giving him one last chance, and a chance is all he needs.

What Would You Do?

Unique situations like this don't always have a perfect answer. The analytical models say to punt, but models can admittedly be a bit shaky with end of game clock scenarios. I'm going with scenario three with the hopes of getting the ball back into Mahomes' hands.

Do you agree with me, or even better, disagree with me? Let me know why on twitter @ Timboslice003. Link to my bio below!