Giants fans are in a weird spot right now. They're out there on social media willfully ignoring their franchise's decision to draft a running back No. 2 overall in what might have been the most loaded quarterback class in the last 20 years -- there will always be quarterbacks in the draft, they argue! -- and blaming Pat Shurmur for their problems despite the first-year Giants coach actually making smart decisions.
Specifically, Giants fans are IRATE at Shurmur after he went for a two-point conversion while trailing by 14 points in the fourth quarter to cut the lead to six. It's obvious, they scream, you just try to get two touchdowns and take the game to overtime. DON'T YOU DO MATH?
But here's the thing: Shurmur did the right thing in that situation.
Allow me to try and explain, while noting we discussed this previously in my Sorting the Sunday Pile column (every Sunday night!) after Doug Pederson did it against the Vikings several weeks ago.
The gist is game-theory based and involves win probability. You have to be willing to accept a couple of premises. First of all, that going to overtime is a 50/50 proposition. Basically -- literally -- a coin flip. And second of all, that converting a two-point conversion is roughly a 50 percent chance.
If you buy into those, the math is pretty simple. The Giants, trailing by two touchdowns, need to score 14 points to tie and force overtime and 15 points to win. Assuming they got the two touchdowns, they had a 97 percent chance of taking the game to overtime by kicking twice.
In other words, they had a 97 percent chance of giving themselves a 50 percent chance of winning.
Think about it. Having a 97 percent chance of forcing overtime doesn't give you a 97 percent chance of winning. It gives you a 97 percent chance of getting to participate in a coin flip.
Conversely, if you get two touchdowns and go for a pair of two-point conversions, the rate at which you'll succeed on one of those obviously jumps. (The rate of success on a pair of conversions will always be higher than on a single conversion. That's obvious right?)
If you get the first two point conversion -- also a 50-50 proposition -- you have just given yourself a 97 percent chance of WINNING THE GAME in regulation should you score the second touchdown.
If you miss the first-two point conversion, you still have a 50 percent chance of taking the game to overtime.
Smarter people than me (those exist!) have written about this elsewhere. Chase Stuart wrote about it at Football Perspective back in 2015, noting how rare it is to actually see this happen. Even if you assume only a 38 percent chance of success on the two-point conversion, it gives you a 38 percent chance of winning, a 38 percent chance of losing and a 24 percent chance of being tied.
Kevin Cole of RotoGrinders also wrote about it here, and included a handy chart to show you the rationale here.
It's simple math guys, but still shocked Shumur did it. Other teams must have studied it after Pederson finally did it a couple weeks ago https://t.co/JsdzJumHdq pic.twitter.com/nVTj6Fy0eL
— Kevin Cole (@Cole_Kev) October 23, 2018
Given the Falcons struggles at stopping two-point conversions, this was an even better gambit for Shurmur.
It's important to use league averages as your base rate when doing things like going for 2, but in case you're curious:
— Football Perspective (@fbgchase) October 23, 2018
Teams converted 11 of 19 two-point attempts against the Falcons defense since 2016.
I still say a ~50% rate is the right rate to use.
Brady Quinn and I actually ended up arguing a bit about this on Tuesday's edition of the Pick Six Podcast and Brady made some fair points. Not every single two-point conversion attempt has a 50 percent chance of success. It's important for these teams to understand who they are playing against, what works in terms of their personnel and other situational aspects. Following the chart and maximizing analytics is important, but there needs to be a plan in place to execute the two-point conversions.
When Doug Pederson went for 2 down 8 a couple weeks ago, analytics people inside the NFL rejoiced. I was told that multiple other teams' charts said to do the same thing. Now we know #Giants among them. Worst case, try for 2 again to tie. Best case, TD + XP wins on next drive.
— Tom Pelissero (@TomPelissero) October 23, 2018
For the record, with Shurmur there clearly was a plan. Eli Manning hit Odell Beckham in the hands for what should have been a conversion on the first one. And then on the second conversion -- after the Falcons hit a field goal to go up 11 -- Saquon Barkley sauntered into the end zone. You can argue Atlanta wasn't as worried at the time, knowing it simply needed to recover an onside kick, but the funny reality is in the loss, Shurmur's move actually paid off.
The Giants would have tied the game if Dan Quinn's aggressive late-game field goal call resulted in a miss and it would have been an analytically-inclined move that got it done while giving them a greater shot at winning the football game.