Thank You for Not Coaching, Week 13
You demanded it! Thank You for Not Coaching is back in its usual Tuesday time slot to review the sprawling action of Week 13. Of course, while the NFL stretched out and played this week’s games over a five-day stretch, the most-discussed decisions of the past seven days both took place on Saturday, when Michigan and Alabama made calls that had an enormous impact on the college football season. There wasn’t a coaching decision quite as meaningful in the pro ranks this past week, but one team did critically injure its playoff hopes with a surprising misstep. As you might suspect, they’re at the very end of this week’s column, and as always, we’ll start on the positive side of things.
The Three Best Decisions of Week 13
3. The Vikings get aggressive on fourth down. Minnesota ticks off just about every box in terms of teams that should be hyperaggressive on fourth down. They have a great running back in Adrian Peterson, who is capable of pushing piles forward and finding the extra yard with his effort and vision. They have nothing to lose, as they’re a struggling football team with no reason to stay tied to conventional wisdom. On Sunday, they also had the benefit of a great matchup, since Chicago’s run defense is putrid, if only because its front seven is riddled with injuries.
After halftime, they got aggressive on fourth down. They converted a fourth-and-1 from the Chicago 35-yard line in the third quarter with a Peterson run for five yards on a drive that eventually produced a field goal. Two drives later, Peterson ran for 19 yards on fourth-and-1 from Chicago’s 31-yard line. That conversion eventually led to a touchdown. Minnesota needed those 10 points to get to overtime, where they eked out a win. But more on that later.
2. Jack Del Rio gets off a pair of low-risk, high-reward challenges. In his final game as interim Broncos head coach, Jack Del Rio actually did an excellent job of using the challenges he had left in situations with significant upsides. Nursing a lead in the fourth quarter, Del Rio threw his challenge flag out for two would-be Chiefs fumbles that were close enough to merit a replay. In both cases, winning the challenge would have essentially ended the game. The first would have given Denver the ball deep in Kansas City territory with a 35-21 lead and 12 minutes to go, while the second would have given Denver the ball inside the Kansas City 5-yard line with a seven-point lead and 3:27 left to go. With a lead and Peyton Manning in the fourth quarter, Del Rio was right to try to use those challenges (and the attached timeouts) to end the game on close calls.
1. The Dolphins go for it, fail, and end up taking the points anyway. Just as the Vikings and Broncos adapted their decision making to the situation at hand, Miami coach Joe Philbin made the right call by being aggressive at the end of the first half in his game against the Jets. With two timeouts and two minutes to go, the Dolphins faced a fourth-and-goal from the 1-yard line and chose to go for it, with Ryan Tannehill throwing an incomplete pass to Charles Clay. That’s the right decision in terms of your average point expectation, and given that the Jets have a very good run defense, throwing for the yard is probably the right play call. (Miami failed on a previous fourth-and-1 in no-man’s-land in the first quarter when Tannehill overthrew an open Brian Hartline for a would-be touchdown.)
What makes the call so great is that the Dolphins were in a 0-0 game against Geno Smith. A touchdown in that spot might be enough to win the game outright. Then, if the Dolphins fail, Smith is a turnover machine who is entirely capable of giving the ball right back if the Jets try anything remotely aggressive, meaning that Miami’s next drive is likely to begin with excellent field position. As it turned out, the Dolphins used their two timeouts after two Jets runs, and when a third-down pass fell incomplete, the ensuing punt saw Miami take over on the Jets’ 46-yard line with 1:35 left. Not only did the Dolphins have enough time to drive back downfield and get in range for another field goal, they even picked off Smith on the subsequent drive and kicked a second field goal before halftime. Miami was right to be aggressive and ended up with two field goals to show for its efforts, even if the touchdown attempt failed.
OK. For one week only, I’m sneaking a special college-football edition into TYFNC. The decisions were just too big to ignore. Don’t think this means that I’ll be analyzing every Hawaii fourth-down decision next year; there’s just too much football during the week to cover. But with the weird NFL week, there’s space to look at the big decisions from Saturday football.
Let’s start with the obvious one: Michigan–Ohio State. There, a Michigan touchdown brought the Wolverines within one point of their hated brethren, 42-41. With just 32 seconds left on the clock, Michigan head coach Brady Hoke reportedly conducted a quick poll of his seniors and asked them whether he should go for two to win the game. They all obliged, so Hoke sent out his offense, only for a Devin Gardner pass to be intercepted, ending the game.
Was Hoke’s decision the correct one? Absolutely, and it’s not even close. You’re weighing the probability of winning by converting from two yards out versus the probability of winning in one or more overtimes, and the former is clearly greater than the latter. Michigan were 17-point underdogs heading into the game, and if you’re a massive favorite, your advantage is more likely to show over a larger sample than it is over a very small one. Bringing the game down to a single play increases the variance of outcomes, which is of huge value to the underdog. Even if the pregame spread overstated the difference in play between the two teams, Michigan would have unquestionably been comfortable underdogs heading into overtime, which would likely take place over 15 plays or more.
When you compare that to the likelihood that Michigan would convert a two-pointer, it’s a no-brainer. Their chances of converting from two yards out (if we assume they don’t allow Ohio State to score in the subsequent 32 seconds) are far higher than their chances of winning an even game in overtime, especially given the relative strengths of their team. Michigan’s defense had basically been run over by the Ohio State rushing attack all day, with Carlos Hyde & Co. combining for 393 yards on 46 carries. Ohio State punted just three times all day. And Michigan’s offense had been brilliant, too, with Gardner going 32-of-45 for a staggering 451 yards with four touchdowns and no interceptions (before the two-point play). The Michigan running game had even averaged 4.3 yards per carry, and the Wolverines had scored on each of their three final possessions of the game. Going for two played to Michigan’s strengths as a team, eliminated their weakness, and created a higher-variance opportunity for the team to win. It’s the right call in that situation every time. You can argue about the play call all you want, but the decision to send the offense out there was correct.
As for Alabama’s kick and the stunning touchdown return that ensued at the end of a classic matchup with Auburn in the Iron Bowl, well, that’s trickier. Alabama were 9.5-point favorites, so it doesn’t make quite as much sense that they would opt for the high-variance opportunity of a long field goal at the end of regulation. That being said, the downside was much lower, despite what this specific outcome might evince.
Start with the upside. If Alabama backup kicker Adam Griffith hits the 57-yard field goal, Alabama wins without having to go to overtime. The odds of that kick going through the uprights aren’t great at any level. Since 1999, pro kickers are 31-for-87 (35.6 percent) on field goals from between 56 and 58 yards out, and they’re obviously much better than your typical college kicker. The only argument in Griffith’s favor is that he, unlike some pro kickers in this situation, is not stuck kicking in a situation for which he lacks the leg. Nick Saban didn’t need to stick Griffith out there to try to win the game; he would only send out his backup kicker if he thought Griffith had a legitimate shot at making the kick. Given that Griffith nearly put the bomb through the uprights, Saban wasn’t wrong to believe in Griffith. Let’s throw a wild guess out there and suggest Griffith’s odds of making the kick were right around 18 percent.
Now, the downside. What if Griffith misses? Well, in most cases, the game goes to overtime anyway. Since we saw what happened, it’s been assumed that any failure on Griffith’s kick was going to result in a Auburn touchdown, which just isn’t remotely true. Think about everything that has to go right for Auburn to return that kick for a touchdown. Griffith has to miss the kick, but that’s not enough. He has to miss it short; he can’t push it to either side of the uprights and into the net. He can’t miss it too far wide, because if the kick is wildly to the left or the right, Auburn return man Chris Davis isn’t going to be able to field it cleanly (or at all) and won’t have a return play. So, Griffith has to miss it straight-on and come up just short of the uprights to set up a possible return. The odds of all of that in itself happening are already likely below 18 percent.
Then, of course, Davis has to return the kick 100-plus yards for a touchdown, as if that were some sort of assured likelihood from the moment he touched the football. The common logic has become that teams are naive for trying those long field goals because, in the case of a return, the big guys they have blocking are stuck trying to tackle a speedy return man. You know, like we’re playing NES Ice Hockey here or something. Adding a return man to the equation does put a smaller, faster player on the field, but the defense that lines up to block field goals and extra points has plenty of slow guys lining up across from the big offensive linemen, too. Watch the play again and you’ll notice that Alabama has several smaller players in pursuit of Davis, while Auburn has a number of enormous players blocking on Davis’s behalf. The problem was that Auburn’s players out-executed Alabama’s guys: Note 194-pound Ryan Smith pancaking star Alabama lineman Arie Kouandjio, who’s listed at 315 pounds. I don’t doubt that Auburn’s guys on the field were faster than Alabama’s group, but it’s not as cut-and-dry as has been reported.
The truth is that returning the kick for a touchdown is far from a sure thing, despite what selective memory tells us. From my count, there have been four such returns in the NFL since 2002, each of which came from a kick from a minimum of 52 yards. Even if we don’t consider the made field goals, that’s four touchdowns amid 389 missed field goals from that distance — a mere 1 percent rate of kicks returned for touchdowns. Even if you assume the odds are greater just by having a guy back there to return and you throw in the odds of a blocked field goal being returned for a touchdown, you’re never going to come to a number that’s higher than the odds of Alabama actually making the kick. Saban was right to try for the game winner. He played to win the game.
The Week in Riverboat Ron
He might want to be called Analytical Ron, but here, he’ll always be Riverboat Ron Rivera. And on Sunday, the Panthers added another fourth-down conversion to their long list of short-yardage successes. I’d call it another notch on the bedpost, but Steve Smith might sue me and then beat me up after winning the lawsuit.
Anyway, the Panthers just went through the usual steps. They were up 10-6 in the second quarter and faced a fourth-and-goal from the 1-yard line against the Buccaneers, who have the league’s eighth-best run defense. Riverboat Ron let the clock wind down before calling a timeout, which remains the only flaw in his fourth-down strategy; he does that more often than not, and it’s going to cost Carolina a meaningful timeout some day. When the Panthers came out of the timeout, they lined up to go for it and pulled the Cam Newton diving sneak out of their arsenal; Newton broke the plane of the goal line before the ball was knocked out of his hands, resulting in yet another Panthers touchdown. It was Carolina’s seventh successful fourth-and-short conversion of eight meaningful attempts this year, with a drop from Brandon LaFell blemishing the perfect record. Each of the seven conversions has either resulted in a touchdown or extended a drive that would eventually produce a touchdown.
I can’t even get angry at Rivera for kicking a field goal on fourth-and-1 in the fourth quarter; there, a 22-yarder from Graham Gano put Carolina up by 21 points with 6:43 to go. Tampa Bay was so interested in trying to mount a comeback after that, Greg Schiano’s men punted at the end of a three-and-out on the very next drive.
Halftime Draw Report
Demerits for four teams this week, and again, it’s often with running backs for whom pre-halftime handoffs seem like terrible ideas.
Why are the Broncos handing the ball to fumble-prone Montee Ball on their own 12-yard line with 25 seconds left in the first half against Kansas City? So they can gain the two yards Ball scrounged up? Denver was trailing at the time; a turnover and a cheap field goal for the Chiefs could have served as the margin of victory if things broke differently. And likewise, in a close game, did Minnesota really need to get six yards from Toby Gerhart on a one-play drive from their own 15-yard line with 12 seconds left? At least they gave Adrian Peterson a rest.
San Diego mounted little offense amid its loss to Cincinnati, but was Danny Woodhead really going to break a long gain with 22 seconds left and the ball on his own 26-yard line? Was he really going to outrun the Bengals defense? At least throw him a screen and try to get some misdirection going. He gained five yards. And while a 15-yard run from Kendall Hunter with 54 seconds left was nullified by a penalty, the 49ers could have just knelt on the ball and gone to halftime without handing the ball off again to Hunter, who is a year removed from Achilles surgery. Keep your backs healthy and avoid turnovers, guys.
The Three Worst Decisions of Week 13
3. The meaningless halftime draw to Montee Ball in Kansas City. That really gets me.
2. The Rams run a fake punt on fourth-and-8 deep inside their own territory. Last year, the Rams got away with a fake punt on their own side of the field just before halftime that I regarded as one of the stupidest plays of the year: It was all downside with virtually no upside, since the team wouldn’t be able to capitalize on the success of the fake, anyway.
Here, the logic didn’t make much more sense. Whether it was Jeff Fisher from the sidelines or punter Johnny Hekker making a call at the line of scrimmage, the move just didn’t pass the smell test. The Rams were down 16-6 with 14:42 to go in the contest, facing a fourth-and-8 on their own 22-yard line. You’re not desperate enough in that situation to go for it, given that you only need a touchdown and a field goal to tie, so why would you instead try a fake field goal, which puts out inferior personnel for running with and catching the football? The Rams didn’t have big guys out there moving the ball like the Alabama critics might have imagined, but they instead ended up with safety Matt Giordano taking a direct snap and trying to run up the middle for eight yards. That’s suboptimal when you consider the variety of weapons the Rams have on offense. If they wanted to try to convert, they just should have gone for it. Instead, they failed on the fake punt, lost yardage, and the 49ers scored a touchdown to seal the game on the next play.
1. Marc Trestman settles for a 47-yard field goal in overtime against the Vikings and loses. It pains me to criticize Trestman, who has been such a great in-game coach this year, but he blew this one. Even worse, he’s been so thoughtful in his answers to the media that we can pinpoint how he blew it. Trestman said after the game that his plan was to kick as soon as the team got in Robbie Gould’s field goal range, which ended up producing a field goal attempt on second down from 47 yards out, which Gould missed.
The problem in Trestman’s thinking is treating “field goal range” like a binary variable as opposed to, well, a range. You can’t just treat a kicker’s likelihood of success as either in-range or out-of-range, because that’s not the way kickers work, regardless of how good they are. Even if you think your kicker’s range is around 50 yards, which appears to be how Trestman regards Gould, your odds of successfully hitting that kick do not remain the same within that range. As Jason Lisk noted yesterday, Gould is 13-of-20 (65 percent) on kicks between 45 and 49 yards over the past five seasons, but that figure improves to 15-of-17 (88 percent) on kicks between 35 and 39 yards. Those are small samples, but it’s true that all kickers see their accuracy rise notably as the kicks become shorter over the broader NFL sample.
By immediately kicking when Gould gets “in his range,” Trestman could be suggesting that the probability of Gould succeeding from any spot in his range is roughly similar, which is patently false. He could alternately be suggesting that the likelihood of losing yardage (via a bad play or a penalty) or the ball altogether (via a turnover) is much greater than any improvement in Gould’s probability that would come by attempting to advance the ball further, but remember the range chart in the preseason TYFNC piece and how drastically a kicker’s odds of success can change by improving the field position even five yards. Trestman should have given his team two more chances at trying to convert and created an easier kick for Gould, as opposed to kicking from the moment he got inside Gould’s range.