A
dissection of John Fox's series of conservative decisions during Saturday's playoff game between the Broncos and Ravens.Did his decisions make the decisive difference in the game?
NFL coaches will often refer to “playing the percentages.” But if there's one thing I've learned by studying strategic decisions, it's that coaches don't have a firm grasp of those percentages. And when anyone is uncertain of the odds, he'll fall back on the sure thing. That was the case with Broncos coach John Fox, who opted for the conservative approach at almost every opportunity in Saturday’s playoff game against the Ravens. Even so, Fox is getting more criticism than he deserves...
...There was also the matter of Peyton Manning's arm. For whatever reason—possibly the cold weather having some effect on his grip—Manning did not appear to have the velocity needed for deep passes. Only 2 of his 43 attempts went more than 15 yards downfield. (Quarterbacks typically throw about 20 percent of their passes deep downfield, and Manning averaged 19 percent in the regular season.)
Here's the
Deadspin link.
The relative lack of wind and the high altitude more than make up for the (supposed) effect of the cold on FG accuracy.
Who the hell was throwing F-Bombs over at Slate? How do you put up with it?
Anyway, I agree with your assessment of the 4th and 1 in overtime. That was one of the few plays I saw, and I yelled at Fox to go for it.
One factor that I wish got more coverage in this article is the fact that Fox called lots of limited upside runs up the middle on 2nd and short and 3rd and short in the second half and OT. Are not those situations in which throwing (ideally a longer pattern, but if Manning wasn't up to it, at least going for more yardage) is optimal? Perhaps I am wrong in applying the logic I've seen in other places on this site to this game and those plays in particular, but it sure seemed to me he was very conservative in his play calling.
If you intend to make a chart like last week of WPA sacrificed by coaches on suboptimal 4th down decisions, it might not include these play calls but I'd think Fox will have left a lot of chances to win on the table with all these factors together.
On the other hand, maybe those are plays in which there are two called plays (pass and run) and Manning decided to do the run at the line of scrimmage. If so, they are less on Fox.
It also seems pretty clear to me that IF you are going to punt late in regulation, then it is a no-brainer to make sure that you have a safety deeper than the deepest receiver. That didn't happen, and the Denver defender seems to have owned up to not being where he needed to be on that tying TD play.
No matter what that was a great game from the comeback and excitement perspectives and it is wonderful that this site provides a way to quantify that. That is only one of many things that this site does well. I'm so glad I found it during the last month of this regular season; seeing the EPA, WPA and other analytic angles truly change the way one watches the sport. Thanks to Brian and others for opening that up for me!
Tyson
Now that you mention it, I felt like (no data to back up) that teams were a lot more predictable in the playoff games I watched. Seemed like after a successful run, the coach would call another run right after it, when you really should throw in some passes. Curious to see if it was true, but overall coaches went wimpy in the playoff games.
Manning is a proven choker. This isn't news. Looked like a dear in headlights in the 4th Quarter and overtime. Classic playoff Manning. Look who saved $40 million
I said this in the Slate comment sections, but since you would be wise to avoid that comment section like the plague, I'll say it again here.
Your argument that Fox was right to commit to the run on 3rd and 7 ignores a third possible outcome of a passing play - that is, that it is a completed pass that fails to convert the first down.
Even if I grant the league-average rates on conversions, it's still better to use the full playbook than to run into the line if the chance of a completed slant/screen/dig route for less than 7 yards is greater than 16%. Given the QB and the offense up to then, that seems very realistic to me.
I wish someone could easily break down the first down running plays from the second and short and third and short running plays. I am a biased depressed fan but I think there were about 8 1st down runs for 9 yards and then Fox just tried to pound it. He we had done a more creative run or a pass some of those times, we would have dominated the game. This is not a low variance strategy, it is a stupid strategy.
Tarr-I thought of that. There were a number of other considerations that outweighed that possibility, the biggest being the possibility of a successful conversion on the run.
You may be right and you may be wrong, Brian. Given the way the Ravens defense was stacked up after 5 straight runs, I tend to believe that the tackled receiver in the flat was considerably more likely than a 7 yard run up the middle.
The point, I guess, is that unless we're actually pulling data on these two outcomes, it's silly to assert that there really was a 15% versus 13% advantage to Fox's decision. Give him the benefit of the doubt if you like (personally, I don't), but don't pretend that it's supported by a data-driven analysis. We're guessing.
I agree with Tarr and Andy...
To Tarr's point, the third conversion rate is the NFL average, not what PM and the Broncos conversion rate is. I'm sure the Broncos would have an above average conversion rate. Secondly I've gotta imagine that Manning would've only thrown the ball if he was fairly confident he could get a completion, even if its not for a first down. If not he could've always taken a sack. IMO there was a very low probability that a pass play would've ended with the clock stopped.
To Andy's point, Fox's low variance strategy also made the team worse. No creative in the gameplan makes a team more predictable and easier to stop. Running the ball repeatedly when you can't run the ball effectively is a poor and conservative gameplan. The Broncos are a much better passing team and should've been passing more.
"There was also the matter of Peyton Manning's arm. For whatever reason—possibly the cold weather having some effect on his grip—Manning did not appear to have the velocity needed for deep passes."
Maybe it was the gloves? Anyway, the so-called best QB ever can´t be excused for lack of arm strength. If you can´t play, don´t suit up. The weather was the same for Flacco, who threw tight spirals without gloves, climaxing in a perfect 60-yds-in-the-air-throw on the money at a tight sideline window, bad defense or not.
After all, sample size or not, Manning is, was, and will be a overpaid playoff choker who is wayyy overpaid, who can´t even lead his team to a win with the help of TWO return touchdowns. Period.
Karl, Germany
Any thoughts on the Packers' punt down 2tds (4th and 5 on GB 49) in the fourth? Given the dominance of SF's O, it probably wouldn't have mattered. However, that very fact seems to be an additional reason not to give the ball back to SF.
Although it appears Fox made some poor descisions, I don't understand why Atlantas Mike Smith's coaching isn't recieving more press.
Why in the world would you call the timeout for your teams game winning field goal with 14 seconds left. There was zero reason not to ensure that field goal kick was the last play of the game. They managed to hang on, but but everyone seems to be ignoring that there shouldn't have been any potential music city miracles and a needed Julio Jones intercepted hail mary to win the game.
You're so right, Karl. Peyton Manning allowing the Ravens to score FIVE touchdowns is simply inexcusable!
Had to be the gloves. Haven't seen that since Jim McMahon.
I read today (I think PK's MMQB) that Manning said he audibled into the the 3rd and 7 run. That's pretty telling on all accounts. 1. Manning was not confident in the pass, and 2. It wasn't Fox's call.
maybe he audibled, but if i remember correctly they had two backs in so manning didn't enough passing options given the defense. and also it didn't look like he audibled (no omaha omaha) so i think he is just falling on the sword.
At least Jim McMahon didn´t choke with gloves on :-)
He did go deep and connected on them, something totally missing with "Nervous-Playoff-Peyton"
Karl, Germany
"Peyton Manning allowing the Ravens to score FIVE touchdowns is simply inexcusable"
I didn´t know that PM interception returns are counted against the DEN defense. Thanks for enlightening me...
All he had to do was playing to his normal (regular season) form. That is not too much to ask for against the 16th ranked pass defense.
Admit it, trust your eyes: Peyton is overpaid, b/c he is a playoff choker.He underlines that year in, year out.
Karl, Germany
Tarr--It's just as silly to say 'we're guessing' as it would be to claim that the .13 or .15 numbers are perfectly definitive.
The numbers, without question, say the decision was at the very least defensible, and I (think I) explained why. The burden should now be on the other side to mount a case. If anyone is 'pretending' here, it's not me.
I'm not sure why you're bringing up a "burden of proof" here, but I've been posting on this blog for only a few months less than you've been publishing it; I think you know that I'm not one of these folks who rejects the concept of using statistics to drive these decisions.
When I say "we're guessing", I mean that we are guessing about two critical variables in this case:
1) What is the likelihood of the Broncos converting a 7-yard run for a first down against a heavily stacked Ravens front?
2) What is the likelihood that the Broncos complete a pass, but the Ravens defenders tackle the receiver short of a first down?
Without some data-driven estimate of how often those two things might happen, we really are just guessing at what those two variables are. When you say it's 15% versus 13%, you are implicitly arguing that the likelihood of those two events are both zero, which is obviously false.
It's possible that the likelihood of the running first down is high enough that it outweighs the probability of a pass completed short of a first down. Personally I doubt that, but it's certainly possible. However, even in that case, both of the probabilities should be lower. It's not 15% to 13%.
Karl, I don't know how you can be a long time reader of ANS and still think that about Manning in the playoffs. The answers are right there at your fingertips.
1. Peyton, even if he wasn't at 100%, was almost certainly still a better option at QB than Brock Osweiler, so him playing was the right choice.
2. Peyton has 129 EPA in 19 playoff games at 0.16 EPA/play. Brady has 114 EPA in 23 playoff games with 0.12 EPA/play. That's 33% more EPA per game and per play than Brady.
3. Want to measure clutch specifically? Peyton has 4.1 WPA in the playoffs at 0.22 WPA/game, while Brady has 4.23 at 0.19. That's 16% more WPA/game for Peyton than Brady. If Peyton is a playoff choker than Brady is even worse, and that's simply an absurd premise.
4. How to account for their different playoff records then? Just look at Peyton's teammates and their contributions. In the playoffs the non-passing aspects of Peyton's teams have added *negative* 6.1 EPA/game while Brady's team have added +0.2, according to Jason Lisk and PFR
http://www.thebiglead.com/index.php/2013/01/14/peyton-manning-had-one-bad-throw-that-will-define-loss-to-baltimore-but-bad-pass-defense-is-why-he-has-a-losing-playoff-record/
I am waiting for Karl to tell us what a choker Tom Brady is. After all, he only managed 13 points instead of 21 against the same defense, in better conditions, with more opportunities, and he had two picks as well.
Not only that, but Peyton had more WPA, EPA, and a higher success rate too.
Peyton's interception was more or less guaranteed once he made the poor decision to throw across his body. Reminiscent of Farve and many others. It's not a question of arm strength but of bio-mechanics. Anyone running to his right and throwing to his left will have a relatively weak throw. Even weaker if he doesn't open up his hips. For zip on the ball the QB needs to set his feet and step into it. Otherwise the throw will be lacking. You'd think any QB would know this after throwing the gagillion passes he has thrown over the years. Just goes to show you that these guys think more highly of themselves and their skills than they ought to. Bet he does it again next year trying to prove how strong his arm is. Or maybe it's just one hit in the head too many.
Got a question for anybody to answer. Brady was number 1 in EVERY Advanced Analytic, yet it was Manning who had all the sports writers thinking he deserved to be in the MVP race. Are the writers as stupid as they seem or do they just have a man crush on Manning or a bias against Brady. Inquiring minds really want to know.
Can't blame Fox for any of the conservative play calling because Peyton reads the defense and audibles as HE sees fit. If he made a bad read or got timid, then that's on Manning not Fox. Amazing how Manning is the Teflon king. Nothing ever sticks to him, screwup after screwup is swept under the rug or blamed on someone else, while his positives are over stated in a big way. Always seems to throw the big interception at the worst possible time, only to have the writers make excuses for him. Sports writers even tried to get him the MVP last year when he didn't play, and again this year although advanced stats (far superior to QBR) had another QB ahead in EVERY category. And Manning lacks the integrity to turn down awards not truly deserved. Two of his MVPs had stats that were clearly undeserving. Like to know why the writers are so bias for Manning. Anybody got the courage to out the writers?