Ask the Stat Geek: Should you ever just let a team score?

Do you have a statistics question (either large or small) about the Steelers or NFL football? Send them in to BTSC’s stat geek at STLRSuperFanDad@gmail.com and I’ll see what I can find.

In this installment, a pretty big question (which took a long time to compile) is the lone stats question.

Geoff Harman from Williamsport, PA.

“As I was watching the KC defense get shredded inside of 2 minutes and NE stopping at the ~5 yard line with ~50 seconds left, I wondered ‘why not just let them score and get the ball back?’ Do you have any idea how many times this season in a one score game, the team needing the comeback scores last and then wins the game (versus having the ball and losing because of running out of time or turning the ball over by downs, interception, or fumble)? I guess what I’m asking is to look at whether it’s worth letting your opponent score so that you can have the ball to control your destiny.”

There are two parts to this question, one of which leads to the understanding of the other. So to start off, I searched through every NFL game from the 2018 season (including playoffs) to see if they met a very specific end-of-game scenario. I was looking for games where a team has the ball in the final two minutes where the game is tied or the team is trailing by one score (8 points) or less and is in field goal range. Since the definition of “field goal range” varies from team to team, I set the standard as the opponent’s 35 yard line. Using these standards, they were 49 of such games in the NFL this past season.

When it comes to which team is the focus of these games, I concentrated on the team who was on defense first due to the second part of the question. For example, when Pittsburgh defeated Cincinnati in Week 6, they were the team who was on defense first inside of two minutes. After the Bengals scored a touchdown, they went on defense to protect the lead. Because the Steelers were the first team to defend, they are the team in which the statistics are seeing if the defense held on to win.

Of the 49 games in the NFL this past season which qualify, 19 of them were won by the team on defense, three of which were went to overtime. Out of those 19 wins by the defense, only 10 games the defense held the offense from making any kind of score. In the three games which went to overtime, the defense gave up some kind of points. In three other games, the offense scored but missed the extra point or two point conversion. In the three remaining games, the defense gave up points but their offense scored in response to win the game in regulation. Of those 10 games where the defense held, they had a lead of four points or more in nine of them, so a field goal was not beneficial to the offense.

So to answer the first question, a team scored in 39 out of 49 games. Of those 39 games, the team won 29 times and tied once.

I realize the first part of the question was not nearly as important as the second. Is there a point where it is beneficial for the defense to just allow the offense to score? The answer is yes, but in very specific situations based on the statistics. It also, by no means, gives the team a probability of winning, it’s just a greater chance statistically than getting a stop.

Before throwing out a bunch of numbers, I wanted to explain that I needed to find the proper sample size in order to calculate the statistics. At first I calculated several stats using just the 2018 season. But one single anomaly could throw off the statistics, which it did in one category (Miami scoring a touchdown to beat New England after receiving the kickoff with less than 30 seconds remaining). After also gathering statistics over the past five seasons, 10 seasons, and 20 seasons, it was determined that the best stats came from averaging the last five NFL seasons. The reason going back farther would not give the best statistical representation is because the way both passing yards and attempts have increased in recent years. The last five years is the best sample size which best represents the current NFL landscape.

There are three determining factors as to the best choice statistically in playing out the end of the game: the score differential, the line of scrimmage, and the time remaining. The score differential is key because of determining if a field goal will tie the game, win the game, or if a touchdown is needed. Additionally, all of these situations are assuming the team does not have the timeouts in order to preserve time while also attempting to make a stop.

Statistically, if a team is leading by seven or eight points they should NOT let their opponent score. Of drives which reach the red zone in the final two minutes of the game and the team needs a touchdown to tie or take the lead, teams score a touchdown 67.2% of the time. Based on that number, the defense gets the stop 32.8% of the time. Although these odds don’t seem great, the odds of the team scoring after getting the ball back are no better in the best case scenario. This is because of the statistics for a “response field goal.”

Response Field Goal or Touchdown:

In games where a team receives a kickoff and they are either tied or down by as many as three points, teams score a field goal 32.9% of the time and a touchdown 6.6% of the time. When there is under a minute remaining, the statistics plummet as a field goal is only scored 12.3% of the time with a touchdown 2.5% of the time. If a team is down by more than three points and needs a touchdown, there is a 17.5% chance if there is more than a minute remaining while there is only a 6.7% chance of scoring a touchdown if receiving the ball with less than a minute remaining.

So even though the odds of getting a score in response are about the same as giving up the score, the odds of winning in overtime are 50% which are much more favorable. If the team goes for a two-point conversion to tie or win, the two-point conversion success rate over the last five years is 49.9%. So the odds are much better in getting the stop on the conversion than stopping the touchdown.

In 2018, there were 10 NFL games in which the team was leading by seven or eight points with their opponent in field-goal range with less than two minutes remaining. Two of the games ended up going to overtime while seven games the defense still won the game either by holding their opponent or by having a response score. The only game where the defensive team lost in regulation was when the Chargers scored a two-point conversion to beat Kansas City rather than kick the extra point to send the game to OT.

If a team is leading by four to six points, they should NOT let the team score UNLESS the ball is inside the 20 yard line AND there is more than one minute left. Once a team makes it inside the 20 yard line inside of two minutes when they need a touchdown, they have a 67.2% chance of getting into the end zone. This means the defense has a 32.8% chance of making the stop. If there is more than a minute left, the team has a 39.5% (32.9% + 6.6%) chance of coming back with a response score to tie or take the lead back. So although it is not a large margin, it is more statistically probable to get the response score than it is to stop the other team. Once the ball is moved inside the 10 yard line, the odds of stopping the other team drops to 23.4%, while going even lower to 19.6% of making a stop if the ball is moved inside the five. Regardless of the line of scrimmage, if there is under a minute remaining in the game the odds of getting the response score goes down to 14.8% (12.3% + 2.5%), so allowing the team to score does not give a statistical advantage.

A similar example of this from the 2018 NFL season was in Week 16 when the New York Jets were up five points on the Green Bay Packers. With over a minute and a half remaining, Green Bay had the ball on the Jets 1 yard line. While the Jets did not purposefully allow Green Bay to score because they had timeouts remaining, they used their timeouts properly in order to preserve the clock. After the inevitable touchdown, the Jets had enough time remaining to kick a game-tying field goal and send the game to overtime. Unfortunately, the Jets fell short in the extra period without ever seeing the ball, but at least they gave themselves a chance to win.

If a team is leading by three points, they should NOT let the team score. Even though the odds of a team getting a go ahead touchdown is more than half once they reach the 10 yard line (51.2% inside the 10 and 64.7% when inside the five), the odds of getting the needed response touchdown is only 17.5% when there is more than a minute remaining and 6.7% when less than a minute left in the game. With these odds, a team is better to try to dig in and force the field goal in order to make it to overtime.

Of the eight games in the NFL in 2018 where a team was holding a three point lead when the other team moved into field goal range, the game went to overtime in two of the games. In one other game, the defense forced a turnover in order to win. For the other five games, the defensive team lost while in two of the games the team gave up a touchdown with more than a minute remaining but we’re unable to respond with a touchdown.

If a team is tied or leading by two points or less, they SHOULD let the team score once they reach the 25 yard line if there is more than one minute left, or once they reach the 15 yard line if there is less than a minute left. In the given situation, the odds of the team on defense winning the game is very low. Once a team reaches the 25 yard line, their odds of making a field goal improves as they move closer. From the 25 yard line to the 21 yard line, field goals are successful at 84.6%, while from the 20 to the 16 yard line the success rate is 88.6%. When a team gets to the 15 to the 11 yard line, the odds of making the field goal jump to 95.5% while any time they get inside of the 10 yard line the odds go up to 97.7%.

If the decision is made to allow the team to score, the team will then need a response touchdown in order to tie or take the lead. Remember the odds of a response touchdown are 17.5% when receiving the kick off with more than a minute remaining and 6.7% when there is less than a minute remaining. Once a team reaches the 25 yard line, there is only a 15.4% chance they will not have a field goal, so the statistics are fairly close from this distance. But once the team reaches the 20 yard line, there is only a 11.4% chance the field goal will not be made. With there only being a 6.7% chance of scoring a touchdown if there is less than a minute remaining, the defense should only consider allowing the team to score once they breach the 15 yard line as a missed kick is less probable then a quick response.

There are several examples from the 2018 NFL season where teams should have allowed the other team to score in order to at least have a chance rather than face a chip-shot field goal as time expired. One example was the Steelers’ Week 13 game against the Chargers. When Keenan Allen caught the pass on 3rd & 4 with just over a minute remaining in a tie game, the Steelers would have had a greater statistical chance of winning the game if they had not bothered to tackle him and let him score the touchdown in order to respond. The Steelers would’ve had over a minute to answer with a touchdown. There was also another game with a very similar situation, and two other games when the score was tied but the team would’ve had just under a minute to respond.

In Week 17, Tampa Bay was leading Atlanta by one point with 1:47 left in the game when Atlanta had the ball at the Tampa 16 yard line. If they had just allowed the Falcons to score the touchdown, they would’ve had a chance to tie or win the game (depending on the results of the two point conversion), rather than give up a field goal to lose as time expired. This was one of four NFL games in 2018 where the team was up by one or two points and lost on a field goal as time expired when they could have let the team score in order to at least have a shot. In all four games, the team could have got the ball back with over a minute remaining.

In conclusion, if the situation is where a score would tie the game, the defense should not allow the offense to score. If a score would give the offense the lead, there are specific examples when allowing the team to score would be more beneficial based on time remaining and how close the team is to the end zone. In the specific game mentioned in the original question (AFC Championship between Kansas City and New England), Kansas City was better off allowing New England to score the touchdown once they reached the 4 yard line. It was important to have as much time remaining for Kansas City to respond, which they did with a field goal in order to force overtime.