I'm not going to touch Sean Payton's late field goal attempt, down eight, into the wind, with only one timeout remaining. After the miss, on the ensuing Seahawks drive, Marshawn Lynch tore down the left sideline and it appeared he was let into the end zone by the Saints. The question is, did the Saints make the right decision letting him score once the first down was achieved - and should Lynch have gone down at the 1-yard line? The answer to both of these will always be the same, and in this case, it's yes.
Lynch scored with 2:48 remaining in the game. Up by 15, kicking off to the Saints, the Seahawks had about a 99% chance of winning. The Saints would need to score, recover an onside kick, and score again just to tie.
Update: Lynch actually scored with 2:40 remaining, not 2:48 as originally listed in the play-by-play. That means that the Seahawks could have milked the entire clock as listed at the very bottom of this article.
But, what if Lynch had gone down at the one? Using the time calculator, we can see how much time the Saints would have had following a chip shot (99.7%) field goal from the 1-yard line. The Saints had no timeouts left, so we are looking at the blue line.
Seahawks Should Not Have Scored
The biggest element in this equation is how plays sync up with the two-minute warning. Assuming Lynch could not burn another eight seconds off the clock before going down (similar to DeSean Jackson running along the goal line or Madden plays where you avoid defenders to waste time), the following 1st down snap would occur around 2:05 (which is a huge advantage for the Saints as no time would run off the clock after the 1st down play). Since they are on the one, we would be looking at kneel downs, which take less time than a standard play. Let's assume three seconds run off the clock every play.
2nd Down Snap: At or just after the two-minute warning
3rd Down Snap: At 1:18
4th Down Snap (FG Attempt): At 0:36
So, the Saints would likely get the ball back with 30-35 seconds left depending on a kick return with no timeouts left. That means the Saints would need to score 11 points in that time period. While I will not say that this is impossible, the win probability is essentially 0%. The Saints would basically need to either return the kickoff for a touchdown (which could be avoided with a likely touchback coming from kicking the ball with the wind) or a long touchdown on one play, recover an onside kick, and a hail mary or field goal.
After Lynch's touchdown, the Saints scored and recovered the onside kick, getting the ball back with 0:26 seconds only down eight. The win probability here is still only 1-2% (the Saints would need to score, convert the two-point conversion, and then win in overtime), but that is a much better situation for the Saints than having to score twice in 30-35 seconds.
Also, if Lynch did burn eight more seconds before going down with the Madden trick, the Seahawks could run the clock out entirely (Update: Turns out he actually scored at 2:40 so this is how the clock would have run out had he gone down):
1st Down Snap: At two-minute warning
2nd Down Snap: At 1:18
3rd Down Snap: At 0:36
4th Down Snap: 0:00
One last qualifier is that if Lynch goes down at the one, the Seahawks could botch a snap or the field goal and keep it a one-score game. The response to this is that both are < 0.3% chance of occurring, especially when being mindful of it. Even so, the Saints would still need to drive a full 99-yards
Keith Goldner is the Chief Analyst at numberFire.com - The leading fantasy sports analytics platform - and creator of Drive-By Football. Follow him on twitter @drivebyfootball or check out numberFire on Facebook.
published on 1/11/2014
in
game analysis,
strategy
Subscribe to:
Post Comments (Atom)
The Seahawks have had a non-neglible number of blocked FGs this year, including one for 6 at IND. Pete also mentioned that the kicking game was the phase most affected by the weather. In an ideal, "frictionless" world you probably should take the knee at the 1, but I don't expect lynch to make these calculations on the fly.
Fucking idiot!!!!
Nobody would need to make those calculations on the fly. Just add some hand signals or calls that are specific to end-game situations to inform players whether or not to score if they get the chance. Regardless of sportsmanship, these type of stats are going to become as central to sports play calling as they have become for making business decisions. When it's all about winning I'd rather these tactics be used as opposed to the head hunting and fake injuries that teams have implemented in the past (and potentially in the present, aka. the Harvin head hunting of today).
lmao. during the game i was joking when i said ' i bet this site says that he shouldn't score the TD'.
sadly, this site said that he shouldn't score the TD.
Worst Article about Football Analytics of 2014
At that point, they were essentially milking the clock. Does kneeling at, say, the 15, and giving yourself the opportunity for another first down, make more sense, from a win probability stand point?
Good article with one huge mistake. Lynch scored with 2:40 left (play started at 2:48). So if he goes down short of the end zone the game is over.
Thanks, Justin, the play-by-play originally had it listed at 2:48, big difference in terms of time remaining.
Generally, I love the contrarian thoughts. But in this case, everything needs to go perfectly for the Saints just to tie it (TD, onside kick, 2nd TD and a 2 point conversion). Stuff happens. I saw the Giants lose to the Eagles like 35 years ago. My recollection was Czonka fumbling with 25 seconds to go. Or this by Jerome Bettis:
http://www.youtube.com/watch?v=FUcDGLHCOi4
I think he did the right thing by scoring.
It would have been interesting though if he stopped at the 2 yard line and the Saints had to push him into the end zone. :-)
That entire 4th quarter was surreal, so of course it ended with a very obvious illegal forward pass for no apparent reason.
The week before, I was surprised that the Eagles didn't allow the Saints to score a touchdown once the Saints had driven to around the Eagles 20 with under 2 minutes to play. Instead, they defend and hope Graham missed from inside 40.
Math is math man. If the RB for Auburn woulda run out at the 1 yard line on there would be a different narrative.
Normally I agree with not scoring in this situation, but in this case with how close Lynch was to the sideline and how close the defenders were, you'd be asking a lot for Lynch to stop his momentum, execute and time his fall perfectly so that he wouldn't either not score or go out of bounds. In this case not scoring and going out of bounds would be the worst of all possible worlds, better to not take the risk and just score.
I completely disagree. What it comes down to IMO is the fact that FGs are not automatic. There can be a bad snap, botched snap, slip, block, simple miss, etc. So while the author says <0.3% chance? But I highly doubt thats an accurate number for this situation (wind, rain, playoff intensity, etc), especially considering Novak slipped today, Pats long-snapper missed the snap yday, GB almost blocked Dawson, and Dawson missed a short FG against Arizona just a few weeks ago, in a game that felt like it had playoff intensity. And considering the fact that almost every playoff game has had bad rain, wind, snow, or cold.
Would you want to pin your entire teams season on 1 roll of the dice with lower odds on a single roll, or would you prefer to bet against the 3-parlay of (1) Your prevent defense lets a deep passes go for a TD (2) Your hands team fails to retrieve the onside kick (3) Your prevent defense again lets a deep pass go for a TD. I'll bet against the 3-parlay.
And if I'm Payton, my plan is: Load the box with 8-9 and house blitz. What this does is either: Stop Lynch, Force a fumble (Lynch is prone to fumbling, 4 fumbles this year, 7 last year including in each of the play off games), get to the QB for a sack, force a fumble fromt he QB, force a quick bad throw into an interception or incompletion (stop the clock). OR lynch breaks free or Wilson completes the pass...either one goes for a TD. At least attempt to force a turnover, instead of just sitting back and letting em score an easy TD.
As the math can argue, taking a knee helps to lower the T.O.P. for the Saints when they get the ball. That being said, if whenever a team is only up by one score I would always take the points. I would argue that the difference in T.O.P. for the Saints would be less of an advantage than comparing it to the fact of how many plays in a row the saints would have to get right to complete the comeback. A botched snap, handoff, or blocked kick and the T.O.P. argument is right out the window. I always would look to make it a two score game ober almost anything else. To me two scores is always better than one and almost always better than time of possession (albeit a QB taking a knee)
Wow, there are some really uninformed comments in this thread--and that is putting it nicely.
I guess some of them are written before the post was updated. I would hope that once you know the clock could have been milked down to nothing, without ever letting Drew Brees or anyone on the Saints have the ball again, that you would change your position.
I totally called it as I was watching, before even the extra point was kicked. I said to my son that it was a mistake to score. He is only 13, so I don't blame him for disagreeing at the time. I pointed out to him that while it may be fun to run up the score, trading in a certain win for a 99% or 98% chance of winning is not smart. By the time the Saints recovered the onside kick, he was converted to my way of thinking.
But I am glad there are so many people with trouble evaluating these things logically. It allows me to squeak into qualifier positions in poker all the time. Players who should be folding even pocket aces, just folding their way into the money, instead just can't resist playing--even against bigger stacks. They end up running into bad luck with their big hands and letting short stacks like me leapfrog past them.
I don't want the coaches and players of the football teams I root for to make this mistake, however. You don't get to save those extra points you score and apply them to your next playoff game. It is a binary condition in the playoffs: win by any margin, and survive; or lose, and begin the off-season.
The legendary basketball coach Dean Smith understood this, which is why he created the four corners offense. Now, of course, that was an abomination against the intent of the game and extremely boring for the fans. So the rules were changed. But I can't blame him at all for working within the rules he was given.
And maybe the NFL needs to change some rules to prevent these kinds of anticlimactic decisions from being made. If they do, they should include any scenario where a team goes into victory position. Just spitballing here: maybe in the last five minutes of the game, any play that loses yards stops the clock, similar to how out of bounds plays only stop the clock at the end of halves. That would certainly make things exciting. Yeah, I kind of like that.
To me a more interesting question is what's is the optimal play for Lynch if the team is *down* by one or two points? I think you are still best off in that scenario to just kneel at the 1 yard line (followed, of course, by three kneeldown snaps and a chipshot field goal as time expires). Or if the game is tied, same thing.
The funny thing is I can picture more coaches going for that strategy then for the one in the scenario that actually happened. I think there is a magical thinking around being up by two scores that is increasingly unwarranted as kickers and coverage teams become better and better at onside kicks.
I think to expect Marshawn to know to kneel in that situation would be a bit much. If Seattle was clearly within the time where they could run out the clock, then the answer is clear. In this situation when the play started, Seattle was not in that situation. A normal running play would not last 8 seconds, which leaves some time (but not a lot of time) for the Saints to get the ball back. I realize that most here are not placing that burden on Lynch, as he should not really concern himself with the game clock while running. I do agree that Carroll should probably have the team prepared for this scenario.
The interesting call, in my opinion, was whether the Saints should have gone for two when they scored with 26 seconds left in the game. I would like to see the numbers on that. I think that going for 2 immediately would have been a reasonable call. If you miss, it is as close to a certain loss as there is (big deal when you have a 2 % chance anyway). If you get it, you have the option to go for 2 upon another TD to win the game.
Its almost like no one here has heard of a botched snap or a blocked FG before.
MaddenDude...How many botched snaps happen during a victory formation?
That is all Seattle had to do to win the game.
Exactly right, Jerry. And sure: it's conceivable that there could have been a botched snap in victory formation. It's all about comparing risk, though. For Seattle to lose in that scenario, they would have to commit that extremely rare mistake (unheard of but possible) of botching a snap in victory formation. Then New Orleans would still have to recover the fumble, go 99 yards to score, get the two point conversion, and win in overtime. I'll take that microscopic risk over the still small but not nearly as small risk in what actually happened.
Anonymous, I agree that the fault was with Carroll and not with Lynch. (Another element to this is that the player has a countervailing personal interest in adding a playoff "game sealing" TD to his resume; but if he has been specifically instructed in the huddle not to do it, one presumes he will follow orders.)
Alan "totally called it as I was watching, before even the extra point was kicked". Big deal. DId you call it as Lynch was running? In time to communicate over the crowd noise and have him stop at the one?
Anonymous thinks "Carroll should probably have the team prepared for this scenario". Do you really want to try to get players to memorize all of the score differentials, time breakpoints, etc.? Or, should they just worry about playing football?
If this was such an obvious decision, why did a top football analytics writer get it wrong (the first time) in this post, written in the calm of a post-game atmosphere? I think people on their couches tend to underestimate the speed at which things progress and the situations change (yes, the silly challenge did give Seattle some time to think, but the previous play had been a hugely important third down conversion, and Seattle still wanted another first down above all else).
Other anonymous guy, by Carroll having his team prepared, it means that Carroll tells the guys in the huddle immediately before the play. If you get a first down, either go down or stop in bounds before you score. You seem to think that "worrying about playing football" does not include considerations like this. I disagree.
Lynch scored at :40, if he was to get tackled at the 3 yard line, it wouldve been a few seconds before :40...meaning you can't milk the clock down all the way...giving NO a chance. No one considered that huh? Probably the most important thing to consider. Pete Carroll isn't going to tell his RB, hey watch the scoreboard, if its <40s, then kneel, if its >40s then go for the TD. If he got tackled inside of the 3, he would've gotten tackled out of bounds or into the endzone. Had it been even 1 second above :40, the entire plan doesn't work.
On top of that, you guys assume a playclock runs the full 40 seconds + 2 seconds of kneel. In reality, they have to snap it with :01 left on the playclock. That 0.5 seconds + about 1.5 seconds to run the kneel = a total of 41 seconds per play, not 42. Every second counts. If you give NO the ball with even 1 second left, in a 1 score game, they can come back and win it. Maybe brees launches a hail mary and they get the DBs on a PI, suddenly he'd attempting another hail mairy but inside FG territory. If you attempt the FG on 4th down, you risk the FG getting block and the FG snap getting botched. You say thats unlikely? NE just had a bad long-snap against Indy, GB almost blocked Dawsons kick, and Novak just missed an easy FG against Denver. Anything can happen on special teams. And running a play to milk the clock always runs the risk of a fumble....which Lynch is prone to do.
All your guys' assumptions are based on a huge "what if"
Apparently, Maddendude thinks it took Lynch "a few seconds" to run 3 yards. I'm impressed he's as successful as he's been with his 50 second 40 yard dash time. He also clearly doesn't understand that his "what if" scenario is 100x less likely than the "what if" he's protesting. I think he's just really really really bad at math