Through four weeks, there have been 28 cases where a team, trailing in the 4th quarter, sent in the kicker or the punter. In general, that’s a pretty low rate — it’s just under once every two games. But while going for it on 4th down isn’t always the right decision when trailing in the 4th quarter, even 28 kicks/punts is too many.

Twelve of those 28 plays were field goal attempts, and all were successful.  Four kicks were to tie or take the lead, but only one of them — Rian Lindell’s 37-yarder, trailing by one point with 38 seconds remaining in week one against the Jets was an obviously correct decision (not that Greg Schiano’s conservative play-calling on the prior three plays deserves the same treatment). Two other kicks were noncontroversial: Facing 4th-and-9 from the Vikings 11, trailing by 10 with 3:40 remaining, I don’t blame Mike Tomlin for sending in the kicker. Even more obvious: Jeff Fisher having Greg Zuerlein kick a 38-yard field goal on 4th and 8 from the Arizona 20, trailing by 3, with 9 minutes left in the game. What about the other 9 field goal decisions? Let’s start first with four end-game strategic blunders:

1) John Harbaugh sent out Justin Tucker to kick a 30-yard field goal on 4th and 4 trailing by 18 with 5:33 remaining at the Broncos 12-yard line. Yes, Harbaugh thought Baltimore’s best chance of winning was to kick a field goal, stop Denver, score a touchdown, stop Denver, score another touchdown, convert the two-point attempt, and then win in overtime. Even though a 30-yard field goal is close to automatic, this one is pretty easy to analyze.  In both situations, you need to stop Denver twice and score two more touchdowns. So the question becomes, it is easier to:

(a) kick a 30-yard field goal and (assuming the other events all unfold in your favor) then have only a 1-in-4 chance of winning (i.e., convert on the 2-pointer and win in overtime); or

(b) score a touchdown on a drive at the Denver 12, facing 4th and 4?

2) Mike Smith had Matt Bryant kick a 25-yard field goal with 3 minutes to go, trailing by 10 against New England, facing 4th and 1 at the Patriots 7-yard line.

This one is not terrible, but I agree with Bill Barnwell’s analysis that going for it would have been the preferred move. Note that you don’t need the benefit of hindsight: Jason Lisk correctly predicted in real time that by going conservative on 4th-and-1, Atlanta would have to go for it (and fail) in a more challenging 4th down play later in the game.

3) Mike Shanahan chooses a field goal, trailing by 10, from the Lions 3-yard line with 1:44 remaining.

The Redskins then held the Lions offense, and got the ball back with 38 seconds left at their own 20. Here you have to consider the two options:

(a) Kick a 21-yard field goal (okay, we can assume 100% success rate here) and then drive 80 yards in 38 seconds to tie the game and then win in overtime; or

(b) Convert on 4th and goal from the 3 (given Robert Griffin III, probably a 45-50% success rate)  and then drive 80 yards in 38 seconds to win the game or drive 50 yards to get into field goal range to tie the game and then win in overtime?

According to Brian Burke, teams have a 2% chance of winning when needing to go 80 yards in 38 seconds, versus a 6% chance of winning when down by three in that situation. Again, Shanahan chose to stay in the game longer rather than maximize his team’s odds of winning. Unless you think it’s not (at least) twice as difficult to drive 80 yards for a touchdown in 38 seconds as it is to drive for a field goal — a pretty untenable position, in my view — then this is a pretty obvious blunder.

4) Harbaugh does it again. Jason already covered how pathetic this decision was, as Harbaugh kicked a field goal, down six with 4 minutes remaining, facing 4th and 5 from the Bills six-yard line last weekend. Again, you don’t need hindsight for this one: I tweeted it as soon as I saw him send out the field goal unit.

The other five field goal decisions came with less late-game pressure (which is essentially a function of deficit and time remaining), but does that mean these were better calls?

1) Trailing by 1 at the start of the 4th quarter, the Chargers kicked a 23-yard field goal facing 4th-and-goal from the Cowboys five yard line. According to Burke, this one is essentially a toss-up: San Diego has a 57% chance of winning if they go for it or if they kick.

2) The Cardinals had a 4th-and-2 from the Lions 15-yard line, trailing by 5, with over 14 minutes left in the game. The odds here favor the bold — an EPA of 2.61/WPA of 41% if you go for it, versus 2.06/37% for kicking the field goal. Giving the Cardinals terrible ground game (and the presence of Ndamukong Suh and Nick Fairley), running it is probably out of the question; that means do you trust Carson Palmer (and maybe Larry Fitzgerald) to complete a short pass? I don’t think this was n unreasonable decision given the personnel involved, but I do think it was suboptimal.

You might recall this was the game where the Cardinals had a negative Game Script but won, thanks to some excellent special teams play. On the next drive, the Lions drove down into field goal range, but a block kept the Cardinals within two points.  Arizona eventually won the game, 25-21.

3) With 7:18 remaining, and the Titans trailing by 7, Rob Bironas kicked a 37-yard field goal on 4th-and-6 from the Chargers 19-yard line. The move worked out well in retrospect, as the three points meant that Jake Locker’s touchdown to Justin Hunter with 21 seconds left was the game winner. But was it the right move?

Following the field goal, the Chargers did their best to channel Norv Turner, calling six straight running plays. One first down helped bleed the clock, and a good Mike Scifres punt (and a penalty) pinned the Titans down at their own 6 with just over two minutes remaining. Banking on Locker to deliver a 94-yard drive seems riskier to me than asking him to convert a 4th-and-6, but this is actually one of the more complicated situations to analyze.

Advanced NFL Stats views this one is a toss-up, and I’m inclined to agree. According to EPA, kicking the field goal is the right move (+1.86), in part because getting the first down still could leave you 13 yards away from a touchdown (EPA of going for it = 1.72). On the other hand, a field goal is not very valuable here, so the win probability is actually slightly higher (22% to 21%) to go for it. Had the Titans scored a touchdown, that would have emboldened Mike McCoy, which would have been a bad thing. So while the decision to kick was conservative, a hidden value was that it encouraged San Diego to play conservatively, too.

4) With 7:18 left and trailing by 6, the Bills attempted a 48-yard field goal at the Panthers 30-yard line. Going for the field goal makes sense because even if you convert the long fourth down, you still may end up with only a field goal. On the other hand, a 48-yarder is no chip shot. Dan Carpenter connected, making this decision uninteresting in retrospect, but the odds still say that going for it was the right call.

5) One of the more interesting ones to analyze straddles the border between late-game optimization and just trying to score as many points as possible. On  opening day against Tampa Bay, the Jets faced 4th-and-2 from the Tampa Bay 12-yard line, trailing by two points, with five minutes remaining. I have no doubt that conservative Rex Ryan never gave this a moment’s thought: take the lead, it’s late in the game!

But realize that after Nick Folk’s field goal, both the Jets and Bucs would score again before the game was over. Five minutes left isn’t as late in the game as it used to be, if you know what I mean. Considering the state of the Bucs offense that day, a good argument could be made that it wouldn’t have been that bad had the Jets failed on 4th down. Tampa Bay would have played conservatively, and the Jets would have regained possession needing to travel only 30 or so yards for a more valuable field goal: one that would give them the lead later in the game.

According to Burke, the WP is 55% if you go for it (70% on a success, 33% on a failure, with a 59% success rate) versus 51% on a field goal attempt (53%, 31%, and 91%, respectively). In this situation, Geno Smith and the Jets offense is below average, while Gerald McCoy, Darrelle Revis, and the Bucs defense is above average. The break-even rate to go for it is 44% for EPA and 49% for WPA. Again, I would go for it, but this is one is right on the border.

The punting situations are less easy to attack: all of the punts in opposing territory came in situations where the offense needed at least seven yards.  Going for it on 4th-and-2 instead of punting is still a good decision in many cases, but I won’t get too angered by coaches who are punting in their own territory with lots of time remaining. Jeff Fisher’s punts against the 49ers were pathetic, but St. Louis wasn’t winning that game, anyway. Two of the most questionable punts were Rob Chudzinski’s decision at midfield trailing by 8 with seven minutes remaining and Bruce Arians’ call on 4th-and-7, trailing by 2, at the Lions 44-yard line. The Browns passed on 4th-and-6 at their own 49 but then went for it on 4th-and-10 from their own 22 on the next drive. Being conservative is often just a way to kick the can down the road, with interest. Arizona was fortunate to pin the Lions inside the 10, but punting that late in the game in opponent territory is rarely advisable.

I’ll also note that Marc Trestman punted on 4th-and-7 from his own 40 — normally a reasonable decision — but he did so when trailing by 21 points with 13 minutes remaining. On his next drive, the Bears had to go for it twice on 4th down (and converted both), but was it not obvious that the Bears were already in desperation mode? Mike Tomlin also sacrified three points of win probability when he punted on 4th-and-5 against the Titans, from his own 49, trailing by 8 points.

You can see all of the punts and kicks in the table below.