## What if KC Had Just Kneeled Out the 2nd Half?

With 13:39 left in the 3rd quarter, KC led IND 38-10. Obviously, 3rd grade math tells us IND needed at least 4 touchdowns just to tie the game at this point. A little more arithmetic might illustrate just how badly KC bungled this game.

There were a total of 1,519 seconds left in the game. KC can burn 40 seconds between plays and 6 seconds during a typical play just by calling a super safe run (that stays in bounds) or even a kneel. Even if KC doesn't try to convert a single first down, they can burn 144 seconds on a series. However, IND can use its 3 timeouts to make one series only take 24 seconds off the game clock.

Because IND is due to receive the kickoff at 13:39 in the 3rd, KC was guaranteed to have at least three possessions--one between each theoretical IND TD. That means that just by kneeling, KC can burn a total of 456 seconds (7:36) off the game clock, which leaves a total of 1,207 seconds of game time (20:07) and no timeouts for IND to score 4 touchdowns.

But if we assume for one reason or another--a long kick return, a turnover, or perhaps a fortunate run--that KC found itself in easy FG range, then IND would be forced to score 5 times. This would give KC a minimum of 4 possessions rather than three, and they would be able to burn 456 seconds just by run/kneeling their way through the 2nd half.

That gives IND 1,063 seconds (17:43) to score 5 times, 4 of which must be TDs. That limits IND 3:33 per scoring drive with no timeouts!

It turns out IND had 6 possessions following the KC TD to go up by 28--5 TDs and an interception that KC in fact turned into a FG after gaining only 4 yards. IND also wasted a timeout following an incomplete pass prior to a KC FG. That means KC gets 5 opportunities to run/kneel and an extra 40 seconds between plays. They could have burned 640 seconds (10:40) without even trying to move the ball. The balance of time would have allowed IND only 2:55 per scoring drive, plus whatever time they burn on the non-scoring drive.

It's easy to criticize Andy Reid for not burning enough clock on offense in the second half, but the real blame goes to the Reid's defense. IND's 5 TD drives took 1:52, 1:28, 1:41, 4:02, and 1:15. "Prevent defense" is the butt of a lot of jokes, but there's a reason it exists--to stop teams from averaging 2:03 on 5 2nd-half TD drives. If there was ever a time to take away the sidelines and big passes over the top, that was it.

Admittedly, there is some hindsight in this analysis. But the point is to illustrate just how absurd the situation was. Keep in mind this scenario assumes zero first downs by KC, even due to penalty, and no time burned on kickoffs. Kneeling out virtually an entire second half is, of course, ludicrous. But doing so would still have forced IND to do the (nearly) impossible.

### 13 Responses to “What if KC Had Just Kneeled Out the 2nd Half?”

1. Anonymous says:

But then the Luck legend would not have been embellished and Pagano would not have had the gratuitous comment that Colt's, down 28, had Chiefs right where they wanted them.
Maybe the scrubs that played exceptionally well in SD should have been brought in.
Brian, this is the NFL. Testosterone rules.

2. Brandon says:

^How bout you put a name with this dumb comment sissy boy!

3. Chris says:

So I think what you're saying, is that NFL coaches should be more conservative? Yes, pretty sure that's it! :P

4. Anonymous says:

The better strategy is to continual delay of games. The clock still runs after a delay of game.

5. Anonymous says:

So, what is the answer to the question?

It seems like KC would be much worse off if they kneeled, since they were able to get several first downs and score twice.

Not to mention, that the kneel strategy would require some very silly kneels when the game got close. You don't see teams up 3 points with 10 minutes left and kneeling.

6. pudds says:

@Anonymous2 - the point isn't to suggest that kneeling the game out would have been the superior strategy, clearly it wasn't. The point is to show how far behind Indy really was - that even if KC had kneeled out the game and stopped trying to score, Indy needed to average less than 3 minutes per touchdown to have a chance - numbers that sound insane, and most of the time, result in an easy win for the team in KC's position.

7. George Frisvold says:

I heard commentators questioning if this was perhaps the greatest collapse of all time. One way to quantify this would be the ratio of the area under the WP curve to the area of the total rectangle (60 min x 1.0 probability). You could use the WP of the losing team (as shown for KC) as the y axis and the area under that curve as the numerator.

8. Jeff Clarke says:

Its unsual for me to disagree with you but I couldn't disagree more strongly with the premise of the discussion. Indianapolis did do the highly unlikely. I don't think they did the nearly impossible. The advancednflstats model had it KC's WP topping out at 97%. I built my own model a couple of years ago. Mine had it a little higher. I had KC as a 160-1 favorite. Either way, 160-1 or 30-1 are both unlikely but hardly unthinkable comebacks.

I think the real problem is the kneeldown mentality at the end of games. Teams are routinely thinking that all they need to do is waste a little clock and play prevent defense. I'd argue that KC's problem was that they went into prevent too quickly not too slowly. I'd also argue that they did the equivalent of an early kneeldown strategy.

Prevent is a strategy that might work if the other team has more than twice as many yards to go than seconds remaining. Using it any earlier than that is not smart. It doesn't stop 2:00 touchdown drives. It enables them.

This is one area I'd like to see a lot more research from the stats community because I feel like the conventional wisdom is just wrong.

You don't need a lot of bombs and sideline routes in order to go down the field in 2:00. Imagine a team has to go 75 yards. They have no timeouts. They don't ever pass to the sidelines. They pass for 15 yards each time. They only complete half of their passes. How long does a touchdown take?

Its a pretty simple math question. They will need 5 completions to get there. The last one will stop the clock since it occurs in the end zone. They will burn about 13 seconds rushing to spike the ball after each of the other four completions. That is 52 seconds total. They will throw the ball 10 times at 6 seconds a play. That is 60 seconds. So the whole thing takes only 1:52.

I know people are thinking that this assumes everything goes right. However, it doesn't. In fact, its actually kind of pessimistic because its assuming that the offense is only successful at taking the play the prevent is designed to allow 50% of the time. It also assumes no timeouts and a relatively long 75 yards to go. I've seen defensive teams go into the prevent with a lot worse circumstances.

This happens over and over. Teams routinely drive the distance of the field in what appears to be a short time. The only times when the prevent seems to actually work is when there is a complete failure of the offense. They take a sack against 3 rushers. They refuse to accept the open 15 yarder in the middle of the field.

Its not really applicable in this game, but I see a lot of situations. The leading team has 3rd and long with 2:40 left. They can run and guarantee the other team 2:00 to score or they can try for a game winning first down pass. Failure gives the other team 2:30 instead of 2:00. The vast majority of the time, they run. I believe its the wrong call. Contrary to popular belief, I don't think the extra 30 seconds materially changes the trailing team's chances. I note how few times teams actually run out of time when they start with more than 1:20 or so to go. If they lose, they lose because they run out of downs or turn the ball over, not because they run out of time.

Prevent loses games

9. Anonymous says:

I've been a longtime reader but never commented; however, something puzzled me about the final 2:10 of this game. The Chiefs had a timeout remaining and had 4th and 11 from approximately mid-field. The clock was running. Reid elected to burn a timeout before the two minute warning to setup a 4th down play to hopefully get a first down. Why not run a quick play without taking time timeout? If you convert, great. If you don't, then you still have a timeout AFTER the two minute warning to use to force a punt very late in the game giving you a slim chance to score on a punt return (remember Eagles vs. Giants a few years back). I wasn't paying too much attention to the game and I could be wrong about the scenario. So why didn't they elect to go this route?

10. Brian Burke says:

Note: the model doesn't go past a 21 point lead, so the 97% WP for KC is low. Was certainly 99% at the 38-10 point.

11. Eric Valpey says:

If we are talking about offensive strategy, we need to set up the problem with a point in time when the KC offense has the ball. KC's offense never has the ball when they are up 4 touchdowns so we are overstating the magnitude of offensive strategy failure. We can pick KC getting the ball up 21 with 11:47 in the 3rd. Or maybe pick 3rd and 2 at the Colts' 10 with 13:42 in the 3rd up by 21 and start a conservative run mostly strategy. Either way, it is pretty clear that a strict kneel only or even run safe only strategy is not optimal to maximize win probability. No, this was a defensive failure.

Prevent defense is the worst.