Who would have thought that Ted Ginn Jr.'s absence might have made all the difference in the NFC championship game? Kyle Williams' two fumbles on punt returns kept the Giants in the game and all but won it for them in OT. During the course of this 22-punt game, Jim Harbaugh was forced with a few 4th-down decisions. Earlier this year, Brian wrote about Harbaugh's decision to keep the 3 points after David Akers made a 55-yard field goal and the Cowboys were called for a 15-yard penalty. In the third quarter, down 10-7, the Niners were faced with a similar conundrum, this time with a punt. On 4th-and-6 from midfield, Andy Lee hits a beautiful punt the to the Giants' 7-yard line. Justin Tuck is called for running into the kicker, but Harbaugh declines and takes the punt, pinning the Giants deep. But, was this the right decision?
After the punt, the 49ers win probability was 37% (and their expected points were +0.34, meaning they were actually expected to be the next team to score even though the Giants had the ball). So the question is as follows: does going for it on 4th-and-1 after the penalty increase the Niners' chance of winning? The estimated success rate on 4th-and-1 is 74%. If San Francisco succeeds, their win probability jumps to 47%; if they fail, it falls to 31%. So, if we let x be the chances of converting on 4th-and-1, we have the equation 0.31*(1 - x) + 0.47*x > 0.37. Thus, the 49ers should go for it if x > 37.5%. Since the estimated conversion rate is 74% (almost twice our break even point of 37.5%), this seems like a no-brainer: the correct decision would be to take the penalty and go for it.
Now this should be qualified slightly, since we are using the league-average historical rates and this game has somewhat different circumstances. Things like the sloppy weather, how well the Giants' front four had been playing, the defensive nature of this game in particular, will all have an effect on the numbers. Regardless, the sum of those things do not cut the probability of converting in half.
Looking at it purely from an efficiency standpoint, the San Francisco expected points on a success would have been +2.49 versus -1.7 on a failure. Doing the same math, purely based on the expected points, it makes sense for the Niners to go for it if their probability of converting is above 48.7%. This is just one play in the game, and it could have easily gone either way, but the right decision would have been to take the penalty.
Harbaugh was faced with another decision at the beginning of the 4th quarter, on 4th-and-1 from the Giants' 46 and elected to take the delay of game penalty and punt (a punt that only netted 31-yards). If the 49ers had gone for it and succeeded, their win probability would have been 79%, if they failed, 61% (versus the expected 70% from a punt). The break-even point was a 50% conversion rate, 24% less than the estimated conversion rate.
Tough loss for the Niners, but a hell of a season for the team that finished 6-10 last year with a first year coach. Now, on to the Super Bowl.
Keith Goldner is the creator of Drive-By Football, and Chief Analyst at numberFire.com - The leading fantasy sports analytics platform. Follow him on twitter @drivebyfootball or check out numberFire on Facebook.
Should The Niners Have Kept The Punt?
published on 1/23/2012
in
analysis,
game analysis,
strategy
Subscribe to:
Post Comments (Atom)
One key factor in the decision I'm sure was the fact that the Niners were 0-for on 3rd down conversions. I don't think their expected success rates would have been anywhere near the league averages in this game. Maybe even <50%
I agree that the Niners' 3rd down failures may have played into that, but based purely on my recollection, I think the majority of their third downs were third-and long passing situations, where they had to rely on Smith to make a play, rather than third-and-shorts where they could rely on a RB (or even try Smith on a bootleg). I only remember one third-and-very-short, which was the run where Dixon got stuffed. Speaking as a Giants fan, I was definitely relieved each time they punted on fourth-and-short, even if that meant holding my breath every time Blackmon or Ross settled under one.
Brian, can you do an analysis of Coughlin's decision to punt it on 4th down at the end of the game inside 49ers territory? That seems like a play that you would advocate going for.
The breakeven conversion rate is based on expected points from a league-average offensive game. Since this was more of a defensive struggle, the expected points should fall and the equation would be altered. You can't just do a sensitivity analysis on the conversion rate. I recognize that this is still probably directionally correct, but just wanted to point out a caveat that wasn't made in the article and that I have seen glossed over a few times.
Agreed, Unknown, which is why I mentioned that those factors affect all the numbers, not just the conversion rates. That being said, they are not going to affect the equation enough (at least not in terms of win probability) to make the punt the correct decision. It's possible that it could affect the equation purely based on expected points enough.
Great analysis, Keith. We were debating the same thing during the game. I think supporting to the quantitative analysis is the qualitative consideration that the Niners defense was good enough to stop the Giants if the conversion failed.
hey keith,
how'd it feel like betting against the giants?
just wondering,
reader
I know you pay lip service to tweaking the numbers for the specific circumstances of this game, but I'd suggest that the backdrop was such an outlier that it makes the numbers almost irrelevant. There were 20 punts outside of the 2 you mention, the Niners had no success on conversions of any length and the Giants didn't have all that much offense either. Right decisions in the unique context of this game imho.
It absolutely amazes me the ability of some people to throw out cold hard numbers just like that, because they "know" the right answer. Obviously, because we have this extremely small sample size here we should "throw out" the numbers. Throw out? Not let them inform our decisions whatsoever? Really? That's a great idea on a website called Advanced NFL Stats. Lets throw out the numbers and just stick with our preconceived notions about what our gut tells us. Geez!
I agree that adjustments for specific situations are necessary, but the stats are the starting point for those adjustments. They should not be thrown out.
Great comment Boston Chris.
Also I agree with ff. The Coughlin decision to punt it on 4th down at the end of the game was by far the stupidest mistake in the game. I would love to see an analysis on that one just to show how boneheaded it was. Had the Giants lost, that would have been the reason.
To Coughlin's decision to punt, it depends on how much time he left on the clock, but assuming it was around :30 seconds here's how it breaks down:
Estimated Success Rate on 4th-and-4: 53% (probably high in this case)
WP Success: 80%, WP Fail: 33%, WP Punt: 49%
Break-even point is a 34% success rate on that 4th down (.8*x + .33*(1-x) > .49). I think that the correct decision was to go for it, especially given the Giants ability to convert, however, it is far from a no-brainer given the situation.
Re: Keith's most recent comment.
In a defensive struggle, a punt is not necessarily a bad decision. The last thing I would want to do is to fail, give them good field position, with TO's, and the chance to move ~25 yds for a chance at a ~50 yd FG, which Akers can make.
First, punting takes more time off the clock than a failed 4th down. Second, he might fumble it, and give you the game winner. Third, with <30 secs, make them go another 25-30 yds--there's not a lot of time. They might be able to run 5 plays at most, with one being the FG. If a player is tackled in bounds, they prob. get 3 plays + the FGA. It is almost assuredly not happening.
With the matchup of the Giants D vs. the 49ers O, I'll take the Giants D every time--which is what ended up happening. The Giants had advanced the ball some on their 1st 2 drives of OT (in other words, they were gaining net field position--a huge thing in this particular game), then got the gift from K. Williams.
Keith - WP Fail is 33%? That seems incredibly low. Maybe I'm not correctly remembering where the Giants had the ball..
Joseph -
Yeah agree. I still think going for it is the better option, but like I said, definitely not a no-brainer and could be seen either way.
Failure is 33% WP since they would be giving SF the ball at their own 46, great field position for a last second field goal. Again, SF had trouble moving the ball, so that WP may actually be higher on a failure.
They had the ball at around the 46 of the 49ers. That's like 20 yards away from FG territory.
The real question is why didn't the 49ers use their magic Special Teams/Turnover sauce in this game like they did every other game?
As we were informed last week Special Teams and Turnover Numbers from the regular season absolutely correlate to performance in those area sin the post-season, so the 49ers sudden failure is inexplicable.
If by inexplicable you mean 'explained by the fact that their punt and kickoff return man was out', then totally.
I think Niners in 4th and 1 might be outliner. They had very, VERY efficient defense (in 2nd half they OWNED Giants offense) defense. And they had very weak offense. usually the left side of Niners OLine is the better one, but Iupati wasn't 100%. So accepting the penalty isn't such obvious as usual.
Of course putting in a power formation and throwing to Sopoaga/Staley would probably gain 15-20 yards, but I can't fault Harbaugh for playing it safe. It was the strategy that had worked for Niners the whole season. Win the field position battle, let the defense hold them, kick the field goal.
And if not for Williams misfortune (I don't blame him for the fumble, that happens, but the muff was a braindead) the strategy would be probably successfull again.
Is 74% on 4th and 1 true even when on the goal line? If so, it was incredibly stupid in the AFC game for the Ravens to kick a field goal on 4th and goal from the 1, down 3-0 in the first half--right? I figured this site would judge that as a mistake, but I had no idea the conversion rate was anywhere near that high making it an absolute no-brainer.
There were also the 2 near-picks, at least one of which probably should have been caught, so that was kind of unlucky.
Alan - if you pull up the WP graph for that game, people did comment on that decision. I noticed it while watching the game for sure.
Another thing worth noting. It wasn't 4th and goal from the 1, it was 4th and 1 from the 3. Meaning they could have potentially scored, but if not, they could still have gotten a fresh set of downs to try 3-4 more times.
And even if you don't get it, the Pats are pinned very deep, at the 3 yard line. So your next drive will probably have good starting field position and there's a chance for a safety.
Coaches shouldn't wait until halfway through the 4th quarter to start making bold decisions.
There is a strong correlation between NY Giants fans and douchiness...
Not as strong as the correlation between anonymous commenters who insult entire fanbases and douchiness.
No they shouldn't have went for it. The Giants did nothing offensively after the first half, only scoring off turnovers. Harbaugh knew the Giants had nothing and they'd just get the ball back anyway which is what happened. Harbaugh coached a great game and the 2 turnovers (the first more than the 2nd) killed them.