Should Georgia Have Spiked the Ball?

The best game of the season — and one of the best conference championship games I’ve ever seen — came down to one final bizarre play: Georgia quarterback Aaron Murray scrambled his team to the line at the eight yard line, dropped back, and threw a fade route to receiver Malcolm Mitchell. The ball, however, was tipped at the line by Alabama linebacker C.J. Mosley, and it fluttered and landed in the hands of Chris Conley a few yards short of the end zone. Conley instinctively caught the ball, was tackled, and the game clock expired.

Tough call

Obviously, that’s not how Georgia drew it up. And, immediately, led by Gary Danielson, the chorus began: Georgia should have spiked the ball instead of running a play. But I’m not so sure. I think spiking it would have been fine and maybe even advisable, but what I don’t think a spike would have been is necessarily outcome determinative. Richt and his staff had a reason for not spiking the ball, and having the ball tipped and then caught by some other receiver very easily could have happened after a spike as well. Per Blutarsky:

To Spike or Not to Spike. That is the question.  Actually, I’m not sure why spiking is such a slam dunk decision in minds of many people today.  If you read Weiszer’s post on the play, you get a valid rationale for what they called…

“We were moving the ball effectively. By the time we got down to the red zone we didn’t really want to spike the ball. We wanted to keep the personnel they had in the game. We decided to hurry up and get to the line and get another play off. There was a little bit of confusion.”

… and you get an explanation for why it didn’t work out.

Mark Richt on spiking the ball: “Well, spiking the ball takes time. We had plenty of time to call play, so we called the play and we were taking ‑‑ the goal was to take a shot at their back right end of the end zone and the ball got batted, the ball got tipped and it landed to a receiver that was running a speed out.”

And more: “We had the play we wanted. We had a good play. The ball got tipped at the line of scrimmage and it fell in the arms of a guy in play. The ball was going to the back end of the end zone, either a catch or out of the end zone. Because if you have, I don’t know how many seconds there were, 15 or whatever it was, if you spike the ball, you might only have two plays after that.  If you throw the ball in the end zone, you probably get three plays out of it.  So once you spike it, it does take a little time to spike it, and you reduce the chance of having the third play, basically. So the goal was to throw it in the end zone. That’s what Murray was attempting to do.  Once again, the ball got batted, and landed in the arms of our guy in play.”

Plus, they thought they had what they wanted.

[Aaron Murray]: “We thought we would have time for two more plays. Obviously they were running down the field. They wouldn’t get set up real quick. I actually think I had Malcolm (Mitchell) on the fade. When I threw it, he said he had him. I thought he beat him too. If it was an incomplete pass we still would have had one more play to go.

Now, a spike may well have been advisable in these circumstances. Normally a spike is not worth it because you waste a down when time is more valuable. With only approximately fifteen or sixteen seconds, however, it’s unlikely you get all the way to fourth down, so the spike may have been fine or even the better approach. But it’s not a no brainer.

As explained in The Essential Smart Football, a lot of my thinking on spiking the ball was influenced by the great Homer Smith. I would have loved to know what the late Homer Smith would say about Richt’s decision not to spike against Alabama. But one of Coach Smith’s best points about the spike is one Richt mentions above: the spike play itself takes time, so people fantasizing about spiking and running three plays are being fantastical. As Smith explained: “Even when there is not time to use all four downs and even if you can snap the ball for a regular play as fast as you can snap it for the spike, [the spike] consumes a second that might have otherwise allowed you an additional play.” Richt’s thought was that the best shot at getting three end zone plays was not to spike. And without the ball being instinctively caught, there would have still been five-to-seven seconds left; enough for one, maybe two plays.

Again, it’s not that not spiking was absolutely correct, but the anger is better directed elsewhere, and the decision not to spike is totally defensible.

Ultimately to me it was a great play by Mosley, and an unfortunate fluke play for Georgia. One of the ironic things is that, with notable exceptions, Richt has been one of the better clock managers over the years. Much of that is because, after his first season at Georgia when his Bulldogs lost a variety of close games, he met with Homer Smith specifically to improve his clock management skills:

Richt met with Homer Smith, a longtime offensive coordinator at schools like Alabama and Arizona, over the summer and got some calming advice.

“The first thing he said is, ‘You’ll never master clock management,” Richt said.

The second?

“Not many coaches are ever going to say, ‘I’m the guru of clock management.'”

It’s unfortunate that such a great game ended on that sort of play. I give Richt credit for sticking with his strategy and then lucidly explaining it after the game. The bottom line is these decisions are extremely difficult, and his strategy was defensible. And if you’d like to know why Alabama won and Georgia lost, I can think of a number of other more significant issues, including another thorny strategic question: Nick Saban’s decision to go for two in the third quarter. Now that was an interesting decision.

  • coachcort

    Personnally, I like going for 2 early in the game. That way you know, with lots of time left, exactly what you need to do. What i mean is if you go for the 2 late in the game, and miss, then you screwed up and there is little time to do anything about it, except to pray for an onside kick recovery. by going for 2 early in the game, if you miss, now you know you have to hurry up earlier in the game, to get the scores you need

  • Aubie Tigerton

    As Chess Masters study great matches, this game should be studied by all coaches. There are a number of issues in the “end game” such as dealing with clock manaement and play selection.
    I was thinking of Smart Football when I woke up this morning because this is exactly the kind of thing which is addressed here.

  • http://occamsrazr.com Ike Pigott

    Chris, here’s one I haven’t seen anyone tackle: Georgia burned a defensive timeout with four minutes to play, Bama at 3rd and 5 from midfield. Absent that confusion, Georgia saves another 40 seconds. Out of the timeout, Yeldon gets 5.1 yards for the first down.

    What exactly was crossed up for the Dawgs?

  • Jason

    Because the clock is stopped on the first down, spiking the ball only wastes a second or two. Because they set up to run an actual play, they didn’t even get the snap off until 9 seconds. They lost 6 seconds in simply making the decision not to spike it, thus killing one additional shot at the endzone.

  • ElGranJefeDeJefes

    100% correct. Always spike after a first down. You will only lose 2 seconds instead of the 6 seconds it takes to audible to your team. Then you can take all the time you need to set up a good play. I am surprised there are actually people out there willing to defend this decision.

  • IrishBarrister

    To spike or not to spike – that is the question. At bottom, I think the answer always lies with a number of different situation specific factors, such as how long it takes for the coach to rely to the QB, the time it takes for the QB to relay the spike play as opposed to the audible, time on the clock, field position, etc. As a general rule however, I believe that whenever a team has a quick and appropriate audible, calling a play at the time is preferable to spiking the ball. The extra down and the ability to catch the defense slightly off-guard (since they are scrambling to set up) is outweighed by the several extra seconds (0-6, depending on the team and the call) saved with the spike.

    The situation presented by Georgia last Saturday, however, was admittedly unique. Georgia had a ready audible. It was a simple goal line play, with the slot running a quick out and the WR running a fade, putting pressure on the CB to make the right decision without hesitating.
    But the cost of a down was not really a concern with the time left on the clock, compelling a different calculus than my normal instincts. Personally, I would have spiked the ball, preferring to choose the right play carefully against the Alabama defense.

    But Gary Danielson’s decision was entirely defensible. I think he felt, with Alabama running down and scrambling to set up without communication, gave his team a solid opportunity to take a shot at the off-balance Saban defense. The audible did not appear to cost Georgia that much time in comparison to the spike, and appeared to be a logical, if aggressive, game-time decision.

    I think this entire situation suffers from hindsight-bias against calculated risk takers. Rather than appreciating a calculated risk for what it is, most people conclude: it didn’t work, and therefore was a bad decision. Belichick’s 4th-and-2 call comes to mind as a comparison. And when it does work, their hindsight bias compels them judge calculated risks as brilliant. We can use Sean Payton’s decision to onside kick in the Superbowl as an example of that. Neither judgment of course is correct. It’s a calculated risk: the outcome itself cannot be the correct judge of the call. But aggressive play calls will always be billed as either brilliant or a blunder based on the outcome. If you doubt it, what do you think the press (and yourself) would say about the decision had Georgia scored on that play?

  • Chase_Stuart

    Very good post.

    Re: the 2 point conversion, yes, Saban should have gone for two. Basically whenever you are trailing by 11 — but even moreso with only 20 minutes left — you should go for 2 assuming your conversion rate is close to 50% (it is almost exactly that, on average, in the NFL; I have not studied as closely in college).

    My shorthand was “whenever the major downside to going for 2 is that you will need to go for 2 again later if you miss, then go for 2.” That’s because the odds of making the 2 once is much higher than the odds of missing the 2 twice.

    Down 11, there is only one major downside to missing the 2 point conversion. Down 4 or down 5, two field goals do the same thing. The real concern is that down 4, a FG by the other team makes it 7, vs. down 5 making it 8. But the point is this only comes up if you score another touchdown. And whenever you are down 11, you need to score 2 touchdowns (or, alternatively, whether you make or miss that XP won’t matter). Therefore, you go for 2 to make it a 3-point game assuming your rate is very close to 50% (there are other small factors involved, but as long as the rate is close to 50%, you’re looking at a 50% chance vs. a 25% chance; with such a wide gap, the “other factors” are irrelevant.)

  • http://twitter.com/TylerLedon Tyler Ledon

    It means this – YES it took those extra six seconds to snap it, but if the ball falls incomplete then you are still at 6-7 seconds with the clock stopped. Then it will not begin until the next play, which will take probably only 4-5 seconds, so you DO get a third play off it the second pass is also incomplete. So virtually, both scenarios can get you three shots at the endzone, but only running a play on first down and not spiking allowed Georgia to force Alabama to keep the personnel they wanted on the field and not allow them to substitute in fresh legs. The Dawgs had them on their heels and it was a gutsy call to keep the drive alive without spiking. Ultimately, we will never know what the RIGHT call was because a tipped pass at the line that is caught by a completely different could never be anticipated.

  • Kevin

    Come on guys, who is kidding who here??? Look at the result; this is why you spike the ball in this situation, every time. What do you mean the same play could happen if you spike the ball?? No it would not. The purpose of the Spike is to stop the clock, get settled down and call three plays in the endzone. Just ask yourself what would the Bill Brothers have done???Bill Walsh, Bill Parcells and Bill Belichek would all have spiked the ball. There really is no discussion here. Listen to BOBO, even he admitted he made a mistake!!!! Murray wanted to spike the ball!!!!! The whole stadium was shocked when they ran the play. Time to get serious guys, Mark Richt or Aaron Murray are never going to win the big game, reminds me of Jim Kelly and Marv Levy.

  • smartfootball

    Thank you for this bit of mature wisdom.

  • Kevin

    I will say to people who think they should have run a play and keep the Alabama defense on the field; throw the ball in the end zone. That play was at the goal line. Coaches should only run safe plays in the end zone. Looking at the play called, I did not think it was a good play regardless of the spike issue. That was not a good play in the first quarter.
    I have been watching football a long time, baseball also. Mistakes are what cost teams games. When two teams play this close the team that makes the mistakes is going to lose. Alabama made plenty of mistakes but got lucky. Georgia made more mistakes. Still a great game!!!!

  • salt_bagel

    I think you’re right that either strategy could earn you three plays, but I think the spike option is a far surer way to do that. The first down was made with 15 seconds left, and after a spike you probably have 12. As it happened, the snap was made with like 9 seconds left. Since everything else is hypothetical (whether a ball gets tipped, or the QB gets flushed and wastes seconds, etc), I’d rather do everything I can to maximize the chance of having three plays to run. Georgia is not an uptempo team. Unless you are practicing that kind of 2-minute drill on a regular basis and execute it perfectly, you’re going to waste time getting lined up and calling the play in comparison to just running up quickly and clocking it.

  • salt_bagel

    I think the audible cost them like 3 or 4 seconds in comparison to a spike. In my book, getting three plays off with 12 seconds left on the clock (clock stopped at 0:15, minus the time to spike) is a far more sure thing than doing the same starting with 9 seconds (which, in effect, is what happened). With 0:09 on the clock, if you run a 5 second play, all of a sudden you have to get the next one off in 3 or less in order to run a third. I guess you have to decide if your advantage in running that quick play outweighs the risk that you won’t be able to get three plays off.

  • salt_bagel

    One other question that I haven’t seen anybody bring up: Why a fade? Did UGA have an athletic advantage at receiver that made this a high-percentage play?

    The way Richt talks, it was “the play they wanted”, but it’s basically a one-route play. He’s telling his QB to just line up and throw that route and make it so his guy is the only one who can catch it, but that pretty much ensures that it’s going to be a low-percentage throw.

  • http://www.facebook.com/ginoarcaro Gino Arcaro

    Great article. I give Richt credit as well. Second-guessers are just that – guessing second, after the fact. They’re simply guessing about what they would have done. Most second-guessers simply follow the crowd of conventional thinking. Great website.

  • Lawrence LaPointe

    Mr. Luck didn’t throw it in the end zone Sunday.

  • Kevin

    Well we know for sure Aaron Murray is NO Andrew Luck and likely never will be. 3 and 11 against ranked teams. Clock Management is a big part of the game and can be very costly. By the way if Georgia won we would be talking about Saban’s clock management at the end of the half. Or Alabama with a delay of game on a made fake punt.

  • Kevin

    Exactly my point. If this “was they play they wanted” well they deserve to lose. That was a horrible call anytime in the game. That ball has to be thrown in the end zone regardless if you spike or not, period.

  • http://www.facebook.com/bmarsh07 Brandon Marsh

    Just wondering…would spiking the ball have prevented the tipped ball at the line of scrimmage? We can debate whether it was best to err on the side of caution, or continue your momentum aggressively- but I’m not sure any of THAT would have changed what happened during the play- a defensive player simply made a play.

  • D S

    Whatever the odds and risks involved, the 2 point play swung the momentum of the game (or rather, amplified dramatically the momentum gained in the TD drive). That edge remained in place until Saban went prevent and handed it right back to Georgia in the game’s final minute. Almost exactly the same mistake Miles made against Alabama 6 weeks prior. Murray’s a completely different QB when he’s not processing the pass rush constantly, and he showed it on that final drive.

    What are the odds and risks in shifting defensive strategies to beat the clock rather than your opponent?

  • optionbets

    “Even when there is not time to use all four downs and even if you can
    snap the ball for a regular play as fast as you can snap it for the
    spike, [the spike] consumes a second that might have otherwise allowed
    you an additional play.” No way, it takes longer to call a formation play than it does to signal spike as you are running up to the ball. One you have to call the formation as you run up to the line and THEN the play, the other you signal spike which is the formation and then you hike it. There are other reasons not to spike such as the other team may not be ready and it’s the best chance to get a mismatched but that must be countered with the idea that your players are also perhaps not ready and may be less able to make the correct read without the time to read presnap, and the possibility for an illegal formation could end the game and/or take time and yards off the clock (not sure if college has same 10 second run off rule as NFL). They should have spiked it, but you are right, it may have not made a difference., the decision by the WR to catch the ball was idiotic, the QB’s decision not to fire one into the endzone or out of the back of it was risky to say the least

  • optionbets

    But that advantage is negated by opponent’s ability to do the same and gain the same knowledge… Flip the situation… By the same token, you would want your opponent to go for two at the first score because if he fails and after other scores like opponent getting 2 fieldgoals to make it 12 points and you getting 2 touchdowns, now you would get to go for it to go up by 3 and would have informational advantage.

    Additionally, if you go for two, the information of the short yardage play is available which will decrease your chances slightly of future short yardage and 2 point situations. I think a superior idea is to have an idea of odds GIVEN the match up. Run a swinging gate for example to see if you can catch defense off guard giving you very high chance of making it and if so, go for it. OR perhaps running offenseive play and having QB aware that he will audible formations 2 or 3 times and if he gets a matchup where he has a much better chance than average to go for it, otherwise take the delay of game and have kicking team kick.

  • optionbets

    I might agree if say you are down 2 TD and you want to go for the win to do it after the first touchdown because it is a parlay you are betting on. Your odds of losing exactly TWO 2 point conversions out of 2 are about 25%. So assuming you score and get the chance to score again, 50% of the time you get the win, 25% you get the tie and 25% you get the loss by the go for it the first time. (If you look at a tie as half a win and half a loss, you win 62.5, lose 37.5) You are 50% to win, 50% to lose if you wait until the 2nd time. This assumes a situation where your opponent isn’t going to have time to score and put the advantage against you, i.e if he goes for it himself after a TD (or finds enough time to kick 2 fieldgoals to turn 1 point advantage into 7 point advantage). It might happen in a situation where you have 3 time outs and 2 minutes left, but realistically still occurs with 3 time outs and 3 minutes left. You attempt the onside kick with contingency of a 3 and out and if you succeed then you will be able to compensate for advantage by going for two.

  • andrewfly

    The other thing you can do if you spike is tell your receivers to not catch the ball if they aren’t in the end zone.