Last night, you may have heard naysayers point out that the Dolphins’ third-quarter attempt to go for two after scoring a touchdown that put them up one point ended up putting them in a situation in which they were down by three in the fourth quarter as opposed to two. Here’s why you can’t focus on that, and how it relates to a big decision from this weekend’s games.

On one hand, it’s true, but it’s also besides the point; it’s one outcome amid many that can extend from that given moment when the Dolphins decided to try the two-point conversion. You obviously can’t predict the Dolphins would then kick a field goal and allow a touchdown, which is what happened. You have to play the situation at hand, and with a one-point lead and 17 minutes to go in a low-scoring game, the most likely outcome you’re going to face is a Tampa Bay field goal attempt. Going up three is of huge importance, and when a computer looks at the range of possibilities from that two-point decision with the one-point lead, it suggests the breakeven point for going for two in that spot is a relatively low 37 percent.

Even if you want to bring up that one outcome as proof, it’s unfair and disingenuous to do so exclusively in the service of conventional wisdom. There are plenty of times when playing it safe screws a team over, too. Take the 49ers-Panthers game, when the 49ers turned down a chance to go for it on fourth-and-1 from the 2-yard line, instead kicking a field goal that put them up 9-0 in the second quarter. Ex-coach Brian Billick was announcing the game and noted in the moment that points were at a premium and that the 49ers needed to make it a two-score game. (Imagine if a team went for a two-point conversion in the second quarter to make it a two-score game.) Billick is a good commentator; he’s self-deprecating, smart, capable of expressing X’s and O’s succinctly, and he often does a good job of bringing in salient moments from his coaching career to inform the discussion during a game.

He ended up being wrong here; the Panthers eventually won, 10-9, with the 49ers never getting another chance to score. Now, I don’t really care about Billick never mentioning that he was wrong; if I had to talk about things every time I got something wrong, it’s all I would ever do here on Grantland. But what does strike me as frustrating is how there’s never a discussion about how the conventional wisdom steered a team wrong. I’ll get to the play later in the column, but the numbers make it clear the 49ers should have gone for it. They didn’t play it smart; they played it conservative. I’d rather not break down decisions based upon a single outcome — it’s shortsighted and simplistic — but if the discussion heads in that direction, it needs to be mentioned when the safe call goes wrong, too.

The Three Best Calls of Week 3

3. Carolina challenges a would-be 49ers fumble in the second quarter.

Every decision to challenge a call in the NFL balances risk and reward. If the reward is a mere 10 yards in the middle of the first quarter, you better be damn sure you’re not needlessly risking a challenge and a timeout by throwing the flag when the call isn’t clearly wrong. On the flip side, if a play has the opportunity to dramatically change a game, you don’t need to be anywhere near as sure that the call will go in your favor.

That’s what made Ron Rivera’s challenge against the 49ers so smart, even though it was unsuccessful. When a red zone pass to Vernon Davis was ruled incomplete, Rivera (and a fair amount of the people watching who weren’t wearing red at the time) saw a play that looked a lot like a catch and a fumble. The announcers even initially called the play as a fumble. It’s always tough to get a judgment call like that overturned on review, but the upside was enormous, particularly in a low-scoring game: It would take a minimum of three and possibly even seven points off the board by ending a threatening San Francisco possession on the Carolina 6-yard line. The reward is so high that you should challenge, even if you’re not sure, just on the hope you’ll get lucky. I don’t know that this technically qualifies as a Riverboat Ron moment, but it was the right call from Rivera.

2. The Chargers run a fake punt in the opening quarter against the Broncos.

Mike McCoy’s Chargers probably don’t have a ton to learn from the Jaguars, but one thing Jacksonville tried during its David strategy festival against the Broncos was a fake punt on its own side of the field. The Jags’ fake punt didn’t work, but McCoy pulled his fake punt out of his bag of tricks and managed to pick up a first down with a direct snap to Eric Weddle.

What makes the commitment even more impressive is that McCoy attempted his fake punt on the opening possession of the game, on fourth-and-1 from his own 29-yard line. Against Peyton Manning! That’s begging to be second-guessed. Kudos to McCoy for seeing something on tape and trying to retain possession and keep the ball away from a dominant offense. I wasn’t thrilled about his performance as the game went along — kicking on fourth-and-1 in the red zone in the first half is basically wasting time when facing the Broncos with a terrible defense — but he got the day off to a good start.

1. The Cowboys try — and execute — a surprise onside kick against the Saints.

Jason Garrett hit Sean Payton with Payton’s own patented finishing maneuver! Payton authored the most famous unexpected onside kick in NFL history to open the second half of Super Bowl XLIV, but here, it was Payton and his Saints who fell victim to one of the league’s best underdog strategies. After scoring a touchdown to go down 35-17 with 48 seconds left in the third quarter, Garrett had kicker Dan Bailey attempt the onside kick on the ensuing kickoff, with the Cowboys recovering to give Garrett’s team a new possession from its own 48-yard line. The Cowboys went three-and-out, but Garrett can’t know that when he makes this decision. If you assume the Saints would have taken over on their own 22-yard line (the average return from a typical kickoff) and assume these were average offenses and defenses, the Cowboys would gain 2.6 points from recovering an onside kick. Given the nature of this specific game, it was more meaningful than that.

Unexpected onside kicks are usually good ideas in moderation, but what makes Garrett’s decision so valuable is the context of this specific game; namely, that his team couldn’t stop the Saints whatsoever. New Orleans finished with an NFL-record 40 first downs. It was coming off six consecutive drives of 50 yards or more before this onside kick, and would have two more after it. What that does is render field position (for the Cowboys’ defense) as far less meaningful than it normally would be. The Saints were almost surely going to march down the field and score on the Cowboys regardless of where they got the ball.

And when field position becomes less meaningful, the downside risk of a failed onside kick — awful field position for the defense — becomes less relevant. The only way the Cowboys were going to stop the Saints from scoring was to deny them the ball altogether. The only way the Cowboys could make that happen on a kickoff was to try to recover an onside kick. Garrett had just about nothing left to lose and made a decision that was perfect, regardless of whether it succeeded, for the game he was in. You can’t hope for much more than that from your coach.

Should They Have?

There were two moments Sunday when teams could have tried to seal up a victory by going for a two-point conversion in a key spot. Doing so would have been unconventional, but it might very well have given them their best opportunity to win their games. One of the two certainly would, in hindsight, rather have gone for it with the play I’m suggesting. The other probably wouldn’t.

Let’s start with the Bengals. Somehow, their game-ending Hail Mary attempt ended with A.J. Green making his easiest catch of the year for a score that put Cincinnati down 17-16, pending the extra point. The Bengals then kicked that extra point with no time remaining to send the game into overtime, where they proceeded to lose. Should they have gone for two?

The good news is that I talked about a roughly similar topic last week with regard to the San Diego–Washington game, when the Chargers passed on a game-deciding fourth-and-goal from the 1-yard line to kick a field goal that sent the game into overtime, so we can steal a lot of the logic from there. The only difference is that converting from the 2-yard line is harder than it is from the 1-yard line. Brian Burke notes the success rate for two-point conversions in the NFL has been about 47.9 percent; if you figure the Bengals are a slight underdog to win a sudden-death quarter on the road (they were one-point underdogs in Vegas at kickoff), we’re basically looking at a decision well within the margin of error, regardless of what Marvin Lewis chose.

But — but — what about the M-word? No, not Marvin. Momentum, right? If you believe that momentum’s a tangible thing, you certainly wouldn’t argue against a Hail Mary to tie the game as a play that would create oodles of momentum for a team, right? By going for it on the two-point conversion, the Bengals would be taking advantage of a shell-shocked Ravens defense that should have already been jogging off to the showers. It doesn’t change the math for me, but it might for you. (You might also point out that the Bengals had been stuffed on a fourth-and-1 early in the game.)

Of course, the Bengals would probably go back and try the two-pointer if they could. They won the coin toss and had an ugly possession in overtime, eventually choosing to go for it on, coincidentally, a fourth-and-2 from the Baltimore 33-yard line. Gio Bernard lost 11 yards on that handoff, and the Ravens promptly drove the ball downfield and kicked a game-winning field goal. The Bengals passed on an incredibly valuable fourth-and-2 and had to eventually try an incredibly valuable two-point conversion and then had to eventually try a similar fourth-and-2 with much less upside. It doesn’t change the decision, but worth noting.

Detroit, meanwhile, had an opportunity to pull off one of my favorite plays. The Lions scored a touchdown in the fourth quarter to go up 20-13 on the Bears with 2:28 left, pending the extra point. Should they have gone for two to try to make it a two-possession game?

You wouldn’t want to do this in a vacuum unless you were super-confident you were a great team on two-point conversions, but the time constraints make this an interesting proposition. If you succeed, the game is essentially over; if the Bears take over on the 22-yard line on the next possession, their chances of winning the game (per Advanced NFL Stats) down eight are at 16 percent; down nine, they fall all the way to 4 percent. The Bears would have to score, get the ball back via a stop or an expected onside kick, and then drive down the field again to pick up the other points they need to win. And if you fail, you’re still exceedingly unlikely to lose in regulation; the other team is still down seven points, and even if it goes down the field and scores a touchdown, it’s almost never going to try a game-winning two-point conversion (like the Bengals didn’t). You’ll have to go to overtime, which is what would happen if you kick an extra point and the other team scores a touchdown and a two-pointer.

That’s almost what happened to the Lions; they kicked the extra point, went up eight points, and then proceeded to allow the Bears to drive the length of the field, scoring a touchdown with 47 seconds left. Then, even after the Lions committed a personal foul on the first two-point conversion attempt, Chicago failed on its second attempt to convert the two-pointer, sealing the game for Detroit.

I think you can make a case the Lions should have tried the two-pointer in this situation, given the specific context of their matchup at this moment in time. They were facing a Bears team that is down to backups at a minimum of four of their front seven spots. Chicago’s run defense is putrid, and the Lions had been gashing it throughout the fourth quarter. Detroit also has Calvin Johnson and Joseph Fauria, two good options on a passing play near the goal line. Furthermore, the Bears had only one timeout; if the Lions succeeded, Chicago would not be able to score and then try to come up with a defensive stop to get the ball back. It would have to recover an expected onside kick, the odds of which are roughly in the 10 to 15 percent range.

It’s also fair to mention that this isn’t a place where the numbers and I agree. The Football Commentary model I often refer to as a good source for two-point conversion ideas suggests that teams should go for it in this situation only if their odds of succeeding are 58 percent or higher; it’s hard to imagine that Detroit’s chances, even in this context, were that high. Chase Stuart has also written a smart critique of the possibilities on his Football Perspective blog.

In both cases, I don’t think it’s patently obvious the coaches making the calls were either clearly right or clearly wrong to skip the two-point conversion. That might not be the most satisfying answer, but it’s the truth. I do hope, though, that they at least gave some thought to the possibility.

Halftime Draws

Two great teams sneak into the most disappointing section of TYFNC this week with a rare pair of combo halftime plays! Let’s start with the Jaguars, who probably got a little too excited about taking a lead into halftime after getting the ball back on their own 20-yard line with 41 seconds left. With three timeouts, they ran Maurice Jones-Drew into the line for one yard. Then, instead of letting the clock run to halftime, they handed the ball to MJD with 24 seconds left for six more yards. Having proved their point — they don’t like Jones-Drew or want him to stay healthy — the Jaguars floated into their locker room with all the momentum from two carries for seven yards.

The Giants had a weirder ending to the half. They got the ball with 1:18 left on their own 20-yard line, only for Eli Manning to be sacked on first down. The Giants, for some reason, decided to use their first timeout in that moment. OK. Second down was a draw to Peyton Hillis for 12 yards, setting up a third-and-4, which was a handoff to Hillis for five more yards and a first down. That’s fine; you don’t want to punt to the Raiders, but you need to pick up the first-down yardage. That play ended inside of 40 seconds, meaning the Giants could then go into halftime without running a play unless the Raiders called timeout, a move they had exhibited no intention of doing (otherwise, they would have called timeout before third down). Instead, the Giants lined up and handed the ball to Hillis again with 13 seconds left for a four-yard gain before heading to the locker room. Is Hillis about to bust off a 69-yard touchdown run? What’s the upside to that carry? And how did this team win two Super Bowls again?

Oh, and there’s actually one more useless halftime play …

The Three Worst Calls of Week 10

3. The Chargers kick an extra point down 28-19 with 11 minutes left. This isn’t Thank You for Not Announcing or Thank You for Your Graphics Department, so I don’t really want to spend too much time on CBS’s use of a mysterious graphic known as “The Book,” nor do I want to be too harsh on Jim Nantz for suggesting the book should be burned if it was telling the Chargers to go for two in this situation. I will just gently suggest that I don’t think Nantz is building expected points models under the table during down moments at Augusta and move on.

Why should the Chargers go for two in this situation? Well, it’s a decision that comes up in this space all the time. If you go for the two-pointer now and succeed, you can kick the extra point later. If you go for it now and fail, you’re in bad shape, but at least you have 11 minutes knowing you need two scores, not just one. If you kick the extra point now and wait until later to try the two-point conversion, the only way the situation changes is that you’re far more screwed by failure. You’re bound to have way less time left on the clock to pick up that extra possession you suddenly need. You’re not “extending the game”; you’re procrastinating, putting off a key moment until a time when you have no choice but to face the moment of truth. Burn the book all you want. Funny how that comes up as a side product of ignorance.

2. The Lions don’t use their timeouts … twice. Jim Schwartz has a bad habit of forgetting (or choosing not) to use his timeouts in key moments on defense when good timeout usage might create an extra possession for his offense. At the end of the second quarter, the Bears were finishing up a lengthy drive when a Matt Forte run after the two-minute warning gave Chicago first-and-goal from the 7-yard line. That’s the point of no return for using your timeouts; you have to assume Chicago will run a maximum of three more plays, and if you use your timeouts, the Bears will either score a touchdown, kick a field goal, or turn the ball over on downs with about 1:40 left, giving your talented passing attack a chance to score before the half. The only downside is that calling timeouts might allow the opposing team to run the ball, but the Bears had two timeouts and 1:55 or so left in the half with three plays to go; they could very easily huddle and run the ball to their heart’s content, and they did in fact run the ball on first down before Jay Cutler threw his tipped interception on second down. Detroit got the ball back with 24 seconds left, at which point they threw a checkdown to Reggie Bush, their irreplaceable running back with a history of injury woes, for one yard.

Then, during Chicago’s drive in the fourth quarter, the Lions watched as Chicago marched down the field for their late touchdown. This one’s a little trickier; the Bears were never in a goal-to-go situation, and if they failed on their two-point conversion, you wouldn’t want to leave extra time on the clock in the case of an onside kick. With that being said, it’s probably the better option for Detroit to start using its timeouts with about 1:20 left, with the Bears setting up for a first-and-10 play from the Detroit 15-yard line. The combination of a touchdown and two-point conversion is far more likely to occur than the combination of a touchdown, failed two-point conversion, and expected onside kick recovery by the kicking team, so you want to put your team in the best situation possible by leaving time for a possible short drive and shot at a field goal after any scoring drive from the Bears.

Instead, the Lions used one timeout all day, calling it before the onside kick attempt that came after the two-pointer failed. I hope the other five timeouts they left on the board were donated to a good charity.

1. The 49ers kick a field goal in the second quarter to go up 9-0. On fourth-and-1 from the 2-yard line, already up 6-0, the 49ers lined up to go for it. They were facing one of the league’s best defenses, which was doing its best to hold up a Carolina offense that had exhibited no ability whatsoever to move the ball; the Panthers’ first four drives had produced an average of 6.25 yards before punting or turning the ball over. Not per play. Per possession.

Just as the 49ers were about to earn their spot on the positive side of TYFNC, Colin Kaepernick stood up and started doing that really fake barking that comes with an obvious attempt to draw the opposition offside. You, my fullback, move to a weird spot where you would never normally stand on the opposite side of the formation, Kaepernick gesticulated. Let me point out that the defensive end is going to rush on this play! Now I will quickly get back down under center to receive the sna— no, just kidding, I am standing up and pointing out things about the defensive front again! Surely all these smart reads I am making will lead to an obviously successful pl— oh, the clock ran out? Bummer! The 49ers then kicked a field goal from the 7-yard line to go up 9-0.

In a vacuum, the 49ers should have gone for it in this situation if their chances of succeeding were greater than 35 percent. Given the value of field position in a low-scoring game, I would suspect their breakeven point was probably even lower. They kicked a field goal, allowed a touchdown to Carolina on the ensuing drive, later allowed a field goal to give Carolina the lead, and never even got into field goal position again the rest of the game. San Francisco could have picked up a first down or a touchdown by going for it, and even if it failed, it would have backed Carolina up on its own goal line, likely producing another possession with better field position than the field position it would have the rest of the way. It was a mistake when the decision was made, and the 49ers ended up punished by the outcome.