I have seen a lot of misunderstanding out there about the faulty decision making by the Eagles at the end of their collapse to the Falcons. When I was in elementary school, I loved going to the High School football games and would always take my trusty notebook with me to keep track of run/pass splits and man/zone defense alignments. As soon as football season was over, it was on to basketball where I kept shot charts for every player on the team. I knew from an early age that I was a very (very) big nerd and sometimes I forget that everyone else isn’t. So what follows is me attempting to explain optimal decision making for football games, and since it is fresh I’ll use the Eagles game as an example, although this logic applies to any game.
Declining a Penalty on 3rd and 1
The first decision point that I’ve seen confusion about is the Eagles declining an encroachment penalty on 3rd and 1 at the ATL 18 with 2:00 left. This was the correct call. If the Eagles accept the penalty, it is now 1st down and the Falcons, assuming they get a stop, can use 3 timeouts on 1st, 2nd, and 3rd down to minimize clock runoff and get the ball back with about 1:45 on the clock. By declining the penalty, the Eagles force the Falcons to call a timeout after 3rd down which means that they will now be out of timeouts following 2nd down on the next set of downs and there will be an entire play clock the Eagles can run out following the next 3rd down. The probability of being stopped on 3rd and 1 is very small, especially with the Tush Push at their disposal, and if they are stopped they can still go for it on 4th down. The Falcons knew that, that’s why they intentionally committed a penalty. The Eagles knew that too, that’s why they declined the penalty.
Passing the Ball on 3rd & 3
Now we get to the exact situation that we are trying to set up when we declined the previous penalty. It is 3rd and 3 at the ATL 10. There is 1:46 left on the clock and the Falcons are out of time outs. If you pick up a 1st down the game is over. If you don’t, you have a decision to make on 4th down (which is a fairly easy one, but we will get to that momentarily). In the world where you don’t get the 1st down, it is vital that the clock continue to run. With the time to run the play and the ensuing 40 second play clock, you can take the clock below 60 seconds before a potential 4th down snap.
The one thing you can’t do, is let the clock stop. The Eagles dialed up a pass play and it resulted in a drop. The play call is a bad decision. Should it have worked? Yes. Saquon Barkley should have secured the catch and the game should be over. But he didn’t, and now it isn’t, and now the clock stops. If you call a run play in that situation, the clock will continue to roll, even if you don’t pick up the 1st down. Sure, the Falcons are expecting a run, but they have been expecting a run the entire drive and your last 3 handoffs have gone for 9, 4, and 3 yards. But the yardage isn’t even the most important thing, the time is.
Per ESPN analytics, the Eagles would have had a 96% chance to win the game if they had run the ball on 3rd down and had no gain. The decision to decline a penalty on 3rd and 1 so you could get to this exact situation to run the clock, and then not run the clock is inconsistent and the wrong decision.
Kicking a FG to Go Up 6
This brings us to the final poor game management decision: kicking a FG to turn a 1 score game into a 1 score game. It is 4th and 3 at the ATL 10-yard line with 1:39 on the clock. The Eagles make the decision to kick a FG, which not only eliminates the possibility of winning the game on 1 play, it also gives up 20 yards of field position (via the ensuing kickoff/touchback) with no time runoff in a situation where time is paramount, and the Falcons have no timeouts.
Per ESPN analytics, the probability of winning the game if you go for it on 4th down is 95% while the probability of winning if you kick a FG drops to 90%. But that is baking in the fact that you could convert on 4th down. What if we remove that from the equation entirely? What if we input the scenario of giving the Falcons the ball at the 10-yard line down 3 or the 30-yard line down 6? The result is a 7% win probability in the former and a 10% win probability in the latter. Simply put, the decision to kick a FG, even if you make it, hurts your win probability more than failing a 4th down attempt.
Why is that? It is for the simple reason that teams down 3 tend to score 3 points, while teams that are down 6 tend to score 7 points. Down 3, teams get conservative once they are in FG range, content to take the game to overtime. When they are down 6, they are forced to be aggressive and go for the win.
Kicking a FG in this situation costs you a chance to win the game with a conversion, 20 yards of field position, and the opponent’s natural tendency towards conservativism while really only gaining you the ability to not go to OT if they kick a FG. It is the quintessential “playing not to lose” strategy.
When Bad Decisions Add Up
Each decision taken in isolation is bad, but when you combine the cumulative effect, it is much worse. Had the Eagles run the ball twice, failing to convert both times, the Falcons would have had the ball with around 55 seconds left with no timeouts, 90 yards away from the endzone. Though they would have only needed 3 points to tie the game, the Eagles still would have had a win probability of over 98%.
The Core Tenant of Aggressiveness
This brings us to the core tenent of aggressiveness: be consistent. You might find yourself rolling your eyes as you read this, muttering to yourself, “Like those small edges really matter, nerd.” and you would be right! Any 1 decision in isolation can turn out poorly. A 5% win probability swing can still mean that you fail on 4th down or miss the 2 point conversion and lose the game. Any decision with a small edge can turn out poorly in isolation. That’s why a consistent approach is key. Over time, the numbers will bear out.
Think of it this way, if you go to the Casino and play Blackjack, you might win 1 hand. Heck, you might even have a great weekend. But if you continue to play over the course of time, you will lose. Why? Because the house has somewhere between a 0.5% and 2% edge. The hand you won doesn’t mean that edge wasn’t there, it means you got lucky. Likewise, 1 analytically sound decision can have an unlucky result. But over the course of time, you will get the edge.
This brings us back to the 1st quarter. On the Eagles 2nd drive they faced a 4th and 4 from the ATL 9-yard line and they chose to go for it. Analytically speaking, this was the correct decision, but it resulted on a turnover on downs. So why, if the Eagles were willing to be aggressive in the 1st quarter, were they unwilling to be in the 4th quarter? Had the Eagles been conservative in both situations, they would have had 6 total points from 2 FGs. Had they been aggressive in both situations we can’t say what would have happened at the end of the game, but regardless of the outcome it would have improved their chances to win. But instead, they waffled between being consistently aggressive and consistently conservative and they lost the game.
Winning games in the NFL is a razor thin margin and there is no room for riding the fence.
