Following a TD to pull within an extra point of tying the game with just seconds left to play, WAS head coach Mike Shanahan elected to attempt a two-point conversion to win rather than enter overtime. It’s not fashionable to defend Shanahan around DC these days, but I think this was the right decision for a couple reasons.
First, WAS was the underdog in the game. They were probably lucky to get to where they were, just down a point at the end of regulation. Their chances in overtime were not much better than the baseline 47% success rate of two-point conversions. Better teams win more often because the very tiny advantage they have on any one play adds up over the course of a game. An underdog should prefer to bet the game on a single play when the opponent’s advantage is slight, rather than bet the game over the course of many plays in overtime as the opponent’s advantage accumulates.
Second, both WAS and ATL have solid offenses and weak defenses. The Redskins’ best matchup in any phase was when its defense and special teams were not on the field. It’s to their advantage to settle the game with their offense vs the ATL defense. Shanahan bet the game on a situation in which they would have his best squad facing ATL’s weakest squad.
The conversion attempt failed, causing the Redskins to lose, but that does not make the decision wrong. It gave WAS its best chance of winning.
- Home
- game analysis
- Did Shanahan Make the Right Decision to Go for Two?
Did Shanahan Make the Right Decision to Go for Two?
By
Brian Burke
published on 12/16/2013
in
analysis,
game analysis
Subscribe to:
Post Comments (Atom)
With the Titan, you are looking at the reverse situation. They have a below average offense going against a top 5 Defense. They are probably well below 50% to convert and are better off taking their chances in OT.
> How about the Tennessee Titans decision NOT to go for 2 at the end of
> regulation?
You're probably right, but it seems like a very close call to me. The Titans were small underdogs, and Arizona has a solid defense.
I expect the decision to take a field goal on 4th and 1 at the Arizona 27 would come up as a much bigger factor.
In practice, the chance to convert is close to 50%, and, thanks to the influence of the coin flip, overtime is close to a 50% proposition. That means that going for 2 is - baseline - a pretty marginal decision.
Since the risk of injury in OT is non-zero and injuries can have an exceedingly detrimental impact on one's season, it seems odd that more coaches don't attempt a 2 pt conversion rather than tie, especially since both have nearly the same odds of success.
I agree with bluto. Fewer overtimes means less injuries.
And beyond just injuries, there's also the fatigue factor. If you wear out your team, they might have less chance of winning the following week. Though I'm not aware of any studies seeing whether a team who was in overtime the previous week performs worse than a team who wasn't.
In the Titans' situation, there was an offsides penalty called on the extra point. Does the 1 yard make a difference?
Totally agree...in overtime, they may never get that close.
Washington didn't call this play but I'd like to see stats on the most common 2 pt conversion attempt. They snap and immediately throw the ball to the wideout. I feel like I see this play on at least two thirds of 2 pt attempts, including at the end of GB-Dal. I also feel like it almost never works.
I'm curious why coaches are so in love with this play. I feel like the stats for all other 2 point conversions are considerably higher than 47%.
One of the reasons this play sucks so badly is that it fails to take advantage of the defining feature of a 2 point play. The defining feature is that there is no worse outcome than a failure. That isn't true normally. On a regular play, a sack is worse than an incompletion and the worst possible outcome is a pick-six. The quick out has a very low probability of an interception and almost zero probability of a sack. Why do they use it here more than other places.
Coaches should be more creative and call plays that recognize the inherent nature of the play. How about a running play with a lateral if the back realizes the hole is full? How about the quarterback stays in the pocket virtually forever and then just throws the ball 50 feet in the air up for grabs as a complete last ditch maneuver? There are several other things that would be far too risky to attempt on a normal play that would make perfect sense on a conversion.
I ran the numbers for how underdogs do in overtime for a recent post (link).
Underdogs of 1 to 5 points have won 46% of games.
Underdogs of 6 to 10 points have won 42% of games.
Washington was a 5.5 point underdog, so that would put their baseline win probability in overtime at around 44%, making the conversion a positive WPA play, assuming a 47% chance of success. I suspect it may be even lower than 44% since the numbers above come mostly from the old overtime rules. The new rules, for all their flaws, probably do a better job of ensuring that the best team wins (there are now more plays over which team differences can accumulate).
And I agree that the back shoulder fade route has to be the worst play in football.
Titans definitely should have gone for 2 after the offsides. As Mike Tanier points out, it's pretty sad to spend megabucks and 1st rounder on offensive line and RB and then not like your chances of getting the ball in from one yard out.
Jeff Clarke--great point.
Michael Beuoy--thanks for doing the homework and getting those numbers.
Normally coaches make the more conservative but incorrect play so that they won't have to face controversy.
But here Shanahan is, rumor has it he's trying to get fired, and in a case of reverse-conservatism he actually makes the right play. I love the irony.
Jeff Clarke, great point. And the same principle applies when a team is going for it on 4th down, in a situation where failing to convert leaves the team doomed to near-certain defeat. Those situations may be less common than 2-point plays, but they happen reasonably often (see last year's Super Bowl and NFC championship), and the stakes are usually higher, which makes in even more imperative to maximize the chances of success.
Jeff Clarke - great point. I think the reason the kind of play calling you mention is not common is because slow developing passes are not likely to have an increased success rate when there is only 10 vertical yards of field to attack. Giving your receivers 5 seconds to get open sounds great, but you've also given the linebackers time to drop into their zones, all while the defensive front is working to get to the quarterback and the receivers don't have the chance to stretch the defense vertically.
Your point probably makes more sense as an argument for more creative running plays to stretch the field horizontally, and realizing that a reverse getting blown up in the backfield is no worse than an inside run getting stopped at the LOS. When they met in the Super Bowl, Pittsburgh beat Green Bay on a 2-pointer with a classic college triple option play with Antwaan Randle-El as the pitch man. That counts as creative in the NFL when your quarterback is Roethlisberger. Boom or bust running plays are so rarely useful that it's hard to justify spending time practising your double reverse and Big Ben triple option so maybe that's a stumbling block.