The game was tied at 24. The Broncos began a drive with 3:27 left to play. After a big Elway pass and several Terrell Davis runs, Denver put Green Bay in the Field Goal Choke Hold. Eventually, Denver fought its way to a 1st and goal from the Green Bay 8. A holding call on Shannon Sharpe moved Denver back to 1st and goal from the 18. Another Davis run set up 2nd and goal from the 1 with just 1:47 to play. Rather than allow Denver to run down the clock any further, head coach Mike Holmgren elected to allow the TD on the next play to give his offense a better chance to respond with a TD of their own.
In the wake of my previous five-part analysis of intentionally allowing a TD, I learned what the Internet jargon tl;dr stands for. I promise to make this one shorter. Previously, I looked at situations in which an offense that's trailing by 1 or 2 points could run out the clock before kicking a field goal to win. In many cases, depending on the time, score, field position, and number of timeouts remaining, it makes sense for the defense to allow a TD rather than try to force a stop and a FG attempt.
This time I'll examine similar situations where the score is tied. The considerations are a little different than when the defense has a 1 or 2 point lead. A tie score means that the defense can't be relatively assured of a win in the event of a miss. And given a successful FG to break a tie, a FG in response only re-ties the game.
*Remember that this analysis cuts both ways. Whenever it's a good idea for the defense to intentionally allow a TD, it's also a good idea for the offense to take a knee prior to scoring.*
I used the same basic methodology and assumptions as the previous analysis. Starting with a model of when the defense could get the ball back after forcing a stop and FG based on the number of timeouts on hand, I estimated the chances it could respond to a FG with a score of its own given the time remaining. The plots below show the raw and smoothed plot for the FG and TD rates for teams down by 3.
Next, I estimated the chances of the team on defense winning given a FG failure, which includes either scoring within the regulation time remaining or winning in overtime.
Based on standard FG probabilities, I computed the win probabilities for the force-the-FG strategy.
I also estimated the win probabilities for teams that need a TD to survive in the endgame based on time remaining. This estimate assumes the team in the choke hold will start with the ball near their 20-yard line. (Specifically, I used an average of starting field position between the 15 and 25 to incorporate a larger and more reliable range of data.) The chance of a TD in this situation is different than for when the team was down by only 3 following a successful FG, because the offense will obviously play differently needing nothing less than a TD to survive.
In a nutshell, the win probability of the team in the FG choke hold shakes out like this:
wp|offense makes FG = p(response TD) + 0.5 * p(response FG)
wp|offense misses FG = p(TD or FG) + 0.5 * p(no score)
wp(force FG) = p(FG success) * wp(offense makes FG) + (1 - p(FG success)) * wp(offense misses FG)
wp(allow TD) = 0.5 * p(response TD)
When all the numbers are crunched, the results can be compiled into a graph like the one graph below, which plots the win probability for a defense with two timeouts remaining. The colored lines are the wp for forcing the FG based on the field position and time remaining. The black line is the wp for allowing a TD. Whichever line is higher for the given combination of time and field position should typically be the recommended strategy.
The red diamonds mark the final few plays of Denver's winning drive. Point #1 was a 1st and 10 at the GB 32. This was well outside the choke hold zone. But 2 plays later, an Elway pass to fullback Howard Griffith gave DEN a 1st and 10 at the GB 8 (#2). At this point, GB was inside the choke hold zone, and would typically have been better off allowing a TD. But on the next play Shannon Sharpe was called for holding, putting DEN back at the GB 18 (#3), barely outside the envelope.
Denver gave the ball to Terrell Davis, who took the ball to the GB 1 but ran out of bounds to stop the clock. The next play was the allowed TD (#4). [Although this was a 2nd down and not a 1st down, the clock stoppage due to Davis running out of bounds means that the same analysis can be approximated by adding the time of one play (about 6 seconds) to the time remaining.] It looks like Mike Holmgren made the right decision, especially with Brett Favre, in mid-90s form, at quarterback.
How rare are these circumstances? Not that rare. Just in the tie scenario alone, since 2000 there were 256 1st downs that would qualify, or about 20 per season. Some of those may have been in the same game, so I'd guess it's about game per week where it's worth considering intentionally allowing an endgame TD when tied. There were another 166 1st downs in which the defense clung to a 1- or 2-point lead inside the choke hold zone. That makes about 10% of all games where it's at least worth considering, if not actually allowing, the intentional TD. There were 37 total situations in the 2012 season alone.
Remember the tearful PIT victory in BAL earlier this year? Charlie Batch led the Steelers to a dramatic win over their rivals with a late game FG drive. With the score tied at 20, PIT entered BAL territory at the 2-minute warning, with a 1-10 at the BAL 44 (#1 on the chart below). BAL was called for roughing the passer, which gave PIT a 1-10 at the BAL 19 (#2). The Ravens also had an injury on the play, which by rule forced them to forfeit their final timeout. With none remaining and 1:46 left to play, BAL was well inside the choke hold zone.
Allowing a TD at this point would have more than doubled their chances of winning. Instead, PIT was allowed to burn down the clock to 3 seconds remaining before nailing the game-winning FG as time expired. Even after a 5-yard false start penalty (#3) on the first play of the final series, PIT was still well within choke hold territory.
You might say that many situations like this don't really offer a big advantage. After all there's not much difference between a 6% and a 12% chance of winning. That's one way to look at it. The other way to look at it is that whatever chance you might have, you might be able to more than double it, and in some situations do much better than that. If the football gods offered you a 6% chance or a 12% chance of winning, you wouldn't just say, "Screw it. Let's just go with 6%."
And one more reminder here at the end of the article: If a defense does intentionally allow the TD, a choke hold-aware offense could always take a knee prior to getting to the end zone and set up for the chip shot winner, taking the defense's wp to near zero. That's going to be a risk, and there are a few wrinkles to note.
First, a defense should be aware of when it is about to enter the choke hold zone, so that they can say if there is a 1st down inside this yard line, just let the guy score. Don't even wait for the next play on 1st down --especially if there is no clock stoppage on the conversion because that just costs you another 45 seconds. The offense will not yet be prepared to talk about taking a knee, and can easily be suckered into the end zone. This would have been the case in point #1 above..."Ok men, if they get inside the 20, just take a bad angle and let the guy get to the end zone..."
Second, this approach might help defenders to make a big play. Not having to worry about making a tackle after the catch, they can play tighter coverage and shade underneath the routes, increasing their shot at defending or even intercepting a pass.
Lastly, this analysis is just as valuable to the offense as it is to a defense. These are the exact same situations in which it would be better not to score a touchdown. Teams should have a plan for when to take a knee. They can use these charts or do their own analysis. Either way, they should be prepared, because these situations happen more often than we realize.
Here are the generalized charts for when to intentionally allow a TD when tied. Each chart is for each possible number of timeouts remaining. Wherever possible, I've made assumptions that would favor the conservative approach of always trying to force a stop and FG attempt. Therefore, these charts should be interpreted as the minimum criteria for intentionally allowing the TD. In other words, in ambiguous circumstances, you should lean toward allowing the TD.
The envelope of the choke hold zone shrinks with one timeout remaining, and is very small outside the 2:00 warning.
With two timeouts in its pocket, a defense would only want to allow the TD inside the 20 with between 2:00 and about 1:00 remaining.
With all three timeouts left, it's never a good idea to intentionally allow the TD. And if the 1st down occurs prior to the 2:00 warning, there's the bonus of keeping one timeout for the offense.
One final note: By no means do I claim that this is the final word on the question. It's simply the first draft. It's a framework for approaching the question. Later, we might want to change our assumptions or refine the approach, but this can serve as a foundation.