Writers of Pro Football Prospectus 2008

Most Recent FO Features

ThomasEar10.jpg

» 2013 Play-Action Defense

Are the best defenses against play action the best against regular passes too? How much impact does play action really have in an NFL game, and does it correlate from year to year?

25 Sep 2012

Any Given Sunday: Vikings over 49ers

by Rivers McCown

Minnesota overcame the 49ers and all five of their scab-gifted second-half timeouts in a stunning 24-13 victory Sunday afternoon. FO alum Bill Barnwell called this game a case of the Vikings "out Niner-ing the Niners," and while that rings true to me in some ways, the Vikings also have an offensive philosophy in the passing game that the 49ers are not really built to stop. And a unique weapon that perhaps no NFL team is built to stop.

While San Francisco's defense was able to keep the mighty spread offenses of Green Bay and Detroit in check in Weeks 1 and 2, they didn't have quite as much success against Minnesota's passing game. What that really boiled down to was that the 49ers are built to keep the deep pass in check, and that the Vikings don't really have any deep threats to throw to. In 2011, the 49ers had our third-ranked defense, but their success on pass defense was built on shutting out big-name wideouts; they finished 24th in the league on passes to running backs. And on screens and quick hitches, San Francisco allowed a 35.2% DVOA, which is roughly 22% worse than the league average. Taking away Stephen Burton and Devin Aromashodu from the Vikings doesn't really do anything: to hold their offense in check, the man you have to stop is Percy Harvin.

By the VOA

But before we get into that, let's get our weekly look at the VOA values from this game. Would you guess that Minnesota outplayed San Francisco? After watching the tape, I certainly would have. Well, guess again! We've got our third Dewey Defeats DVOA game of the year, narrowly. Most of the difference this time comes from a pair of Kyle Williams returns, one of which set the 49ers up deep inside Vikings territory to lead off the third quarter.

Dewey Defeats Vikings
Team Total VOA Off. VOA Def. VOA Special Teams VOA
SF 2.5% -19.5% -9.1% 12.8%
MIN -8.5% -7.3% -16.2% -17.4%

We should note that Minnesota loses on special teams in part because DVOA does not credit the defense for a blocked field goal; it's not a "luck" play, but it is non-predictive because blocked field goals are so rare. Give Minnesota credit for that block, and the VOA ratings for this game come out pretty much even.

Call of the Game

Minnesota faced fourth-and-goal from the 1 with 7:25 left in the first quarter. Many a head coach would be content to take the three and start the game with a lead, but Leslie Frazier would have none of it. He sent Christian Ponder out, called a play-action bootleg, and despite the fact that Dashon Goldson was barreling down on Ponder, Ponder sent a strike to Kyle Rudolph for seven.

Obviously, the "Always go for it on fourth-down" faction was appeased by this call thanks to the game situation, but more importantly, those extra four points really changed the entire complexion of the game. Despite averaging around 4.4 yards per carry, the 49ers were only able to rush 20 times because the game situations dictated throws. The Vikings were able to get pressure on Alex Smith with their front four, and they were able to play a deep shell and keep the passes that were completed in front of them. San Francisco had just six completions that went for 10 or more yards, and three of them came on their lone touchdown drive. That probably wasn't how San Francisco drew it up this offseason when they brought in Randy Moss and Mario Manningham.

Spotlight on Percy Harvin

Ponder leaves this game with a 24.8% VOA and 246 YAR, good for sixth in the league heading into Week 4. However, on tape, it's clear that his play does not back up that statistical ranking. Rather, offensive coordinator Bill Musgrave has managed Ponder expertly and put him in situations where he can succeed in spite of Minnesota's lack of receiving depth. Adrian Peterson's presence and Ponder's own natural running ability allows Minnesota to run a lot of play-action. Through Week 2, 25 of Minnesota's 77 pass plays (including kneels and plays cancelled by penalty) were off play-action.

Harvin brings his own set of challenges to the table. Minnesota has no fears about lining him up anywhere on the field. He's in the slot, he's outside, he's in the backfield (sometimes in a Full House set), and that's not even getting into putting him in pre-snap motion. If a team plays off-coverage on Harvin, it won't end well: he's got an innate ability to break tackles.

When he lines up as a running back, do you treat him as a receiver, or as a back? In their Week 1 game against Jacksonville, the Jaguars tended to play nickel against the Full House Harvin set, which had Harvin, Rudolph, and fullback Jerome Felton in the backfield. While that's a match on pure personnel terms, I'd be willing to wager that if we had the full scope of NFL history at our fingertips, we wouldn't find many occasions where a defense went with the nickel against a Full House backfield.

How does this affect Ponder? Glad you asked. Per Tom Pelissero, eight of Ponder's 21 completions came behind the line of scrimmage. He had 12 more completions behind the line of scrimmage in the first two weeks, and the average yards after catch on those plays was 12.7. Or, to put it another way, 131 of Ponder's 246 DYAR has come on throws to Harvin, 125 has come on throws to Rudolph, and he's got -10 DYAR on throws to all other receivers.

One good example of the way that the Vikings are getting the ball to Harvin in space came with 5:37 in the third quarter. The Vikings lined up in a double-tight end base set, and the 49ers countered with base 3-4 personnel.

Harvin went in motion towards the slot as the ball was being snapped, then followed under the formation, as the play-action was set.

Harvin finishes the end-around route, then moves clear to the outside.

With Tarell Brown cleared from that side of the play by Michael Jenkins, Harvin is currently being defended by ... open turf.

Harvin makes the catch about eight yards behind the line of scrimmage, and San Francisco has a pair of defenders in front of him, but he's already found open space, and that's all that Minnesota was aiming for.

He then easily splits the pair of San Francisco defenders (Carlos Rogers and Patrick Willis) right down the middle with his speed, and fights through Willis' tackle to get extra yardage.

End result of the play: negative-8 air yards, seven-yard gain, first down.

Let me go back to the Jaguars game again, where I can actually use the coaching tape (blasted deadlines), and ask you a question: ever seen a play-action back catch the ball on a designed non-screen pass?

Minnesota is driving in Jaguars territory, up 14-12 in the late third quarter with a first-and-10. The Vikings come out in their Full House backfield, with Harvin behind the quarterback flanked by Rudolph and Felton. The Jaguars are in nickel, with Aaron Ross playing "linebacker" and assigned to Harvin.

The Jaguars actually do a pretty good job of not biting on the play-action; their linebackers haven't budged much. (I should note: Harvin broke a couple of tackles as a running back in this game, and has 16 DYAR through three weeks on the ground.)

Ross gives Harvin way too much of a cushion on his release, and has no chance to make a play on this ball by the time it's thrown.

Harvin catches the ball with space, and again manages to create extra yards after the catch by barreling over Kevin Rutland.

End result of the play: 13 yards and another first down.

This isn't to say that Ponder doesn't have the ability to hit some downfield throws, that he's not capable of improving, or that this is what he'll be going forward. Right now, though, the passing portion of this offense is built on finding the first or second read, and putting the ball in Harvin's hands.

And more often than not, when Minnesota puts the ball in Harvin's hands, he's proved to be tough to bring down.

Posted by: Rivers McCown on 25 Sep 2012

24 comments, Last at 27 Sep 2012, 7:01pm by Rivers McCown

Comments

2
by Will Allen :: Tue, 09/25/2012 - 2:28pm

I'm too lazy to look; how many teams since the beginning of last year have had 33 rushes, by running backs, against the Niners? How many of those rushes came in the first half? In the first 3 quarters? Yeah, they only averaged 3.5 yards, but that is pretty good against that defense. It seems to me that the Niners defense really tees off when they make an offense extremely one-dimensional. Yes, Frazier likely really showed some understanding of strategy by going for the td rather than the field goal, early on.

If Simpson gives the Vikings some deep chances downfield, and forces defenses to respect that, then Ponder likely will only improve. A hostile Ford Field this Sunday will give some insight as to how far the Vikings offense has progressed.

11
by Karl Cuba :: Tue, 09/25/2012 - 8:21pm

Ithimk what you've said is probably right but it makes me wonder why the 49ers didn't adjust for the Vikings? Why didn't the niners use the sort of approach that everyone else will use against Minesotta, throwing eight in the box to prioritize stopping the run and forcing Ponder into second and third and long? If they had tried that then they might have held the Viking runners to maybe three YPC and been able to be more aggressive. I felt that the Vikings were comfortable with the passivity of the niner defense.

This is especially apt when you consider that Ponder's major weakness, perhaps his only one if he can stay healthy, is that he doesn't have the strongest arm and so he might not be able to hurt you deep like some other quarterbacks. Again, I feel the Vikings' coaching staff really had a better game plan.

12
by Will Allen :: Tue, 09/25/2012 - 9:38pm

I think a good team is most likely to be outcoached when facing a team with a bunch of guys who are starters in their first and second year, if those guys have some talent. The Vikings have four such guys on offense, one of them at the most important position, another at arguably the 2nd most important position, and the third at an offensive position that is more important than, say, the guards, as well. Throw in the known quantity of a running back with HOF caliber talent, and the Vikings seem like they may have the chance to sneak up on some folks, as long as the defensive backfield doesn't fall apart again.

1
by Jovins :: Tue, 09/25/2012 - 1:58pm

The VOA doesn't surprise me at all. Toby Gerhart fumbled three times in the last three minutes, and VOA doesn't know that the game was basically over by that point.

3
by keithwaters :: Tue, 09/25/2012 - 2:42pm

I can do without the political commentary. You must be a Democrat.

8
by Snack Flag (not verified) :: Tue, 09/25/2012 - 7:36pm

Where? I've re-read a couple of times and the most "political" it gets is using the term "scab." Perhaps I'm skimming over something. Seems like an odd thing to complain about and then immediately use "democrat" in a derogatory sense.

9
by DisplacedPackerFan :: Tue, 09/25/2012 - 7:46pm

I took it as sarcasm in reference to the Dewey Defeats DVOA (look up Dewey Defeats Truman if you haven't been exposed to that reference yet, as I know some folks who haven't).

10
by Snack Flag (not verified) :: Tue, 09/25/2012 - 7:58pm

I get the reference and now I feel like an idiot. Kudos for a good joke that went over my head.

15
by Dean :: Wed, 09/26/2012 - 1:50pm

I thought it was the scab comment. There's no real opinion being expressed by the Dewey reference. "Scab" on the other hand has clear and obvious political overtones.

16
by DisplacedPackerFan :: Wed, 09/26/2012 - 2:24pm

Scab has political overtones? Well I suppose as much as unions have political overtones, but I never think of it in political terms. Unions in general, while being the results of laws and laws being the results of politics, are something I almost always think of in business terms. Perhaps I'm just the odd one since I thought it was the clever joke.

4
by Paddy Pat :: Tue, 09/25/2012 - 2:56pm

Rivers,

Just as a point of constructive criticism, aside from the fact that it's a great column, the coaches film is really not helping you here. It's hard to see the play on those shots--the action is so tiny and you can't read the numbers. If you could circle Harvin in the backfield and mark him on the subsequent shots, it would really help. It might also be nice if you would name the other backs in the initial shot, maybe color-coded circles that you could explain in the text? Otherwise, great piece though! It'll be really interesting to see where Minnesota goes from here, especially with the scab refs fielding Green Bay a loss.

23
by Rivers McCown :: Thu, 09/27/2012 - 6:53pm

Yeah, sorry. Again, the deadline with this doesn't give me as much time to play around with the graphics and ideas as I'd like. I'll try to work some diagrams into this stuff if I have time next week.

5
by nath :: Tue, 09/25/2012 - 3:34pm

I can't tell at all where Harvin is in that second series of images. Could you circle him or something?

24
by Rivers McCown :: Thu, 09/27/2012 - 7:01pm

Harvin starts out as the deepest back, then goes up through the hole, and towards the bottom of the pictures.

6
by Karl Cuba :: Tue, 09/25/2012 - 4:21pm

That Harvin play was driving me insane, especially if you can work it into a series of end-arounds with the requisite fakes. Just a great package to keep a defense off balance while trying to defend Purple Jesus.

The Vikings outcoached the niners. The whole league knows that the 49ers call two plays in the huddle and then let Alex Smith choose between them. The Vikings were perfectly happy for that to happen as it meant that they could make the 49ers pass the ball, often out of 22 personnel and six linemen packages. The end result is that they forced the 49ers into throwing their small ball plays with the wrong players on the field. The 49ers' coaching staff might have to rethink that approach.

7
by Little Bobby Tables (not verified) :: Tue, 09/25/2012 - 7:18pm

My question is: How do you contain a player like Harvin? You'd think that the 49ers, with two of the best ILBs in the league, would have a good shot at tracking down Percy in the open field. I've got no idea what the 49ers could have done differently.

I'd be curious to know how other teams stop the Sproles, Harvins, and Spillers of the world. My guess is that it involves quick sacks and zone coverage, neither of which the 49ers do on a regular basis, but honestly I'm not sure players like this can be stopped.

13
by LionInAZ :: Tue, 09/25/2012 - 10:36pm

One way NOT to do it is to leave him uncovered on a fake end-around, especially since the Vikings don't have many other receiving threats. Giving him room to catch and run just invites disaster.

I'm very impressed with what I've seen of Kyle Rudolph. He looks to become one of the league's top all-around TEs over the next few years. I wish Brandon Pettigrew would show more of that kind of consistency.

21
by jimm (not verified) :: Thu, 09/27/2012 - 3:28pm

Rudolph is a fine pass catcher but he's a way below average blocker at this point

22
by LionInAZ :: Thu, 09/27/2012 - 6:09pm

He seemed like a good blocker in the games I've watched but I probably haven't seen him as much as you.

14
by David :: Wed, 09/26/2012 - 3:00am

They could have schemed to stop Harvin, but they don't do that. The niners' defensive success is built (mostly) around a pretty vanilla strategy. Unlike the Belichick approach, where the strategy is to take away whatever the offense is most comfortable with (and inevitably leaving larger gaps elsewhere), the niners' approach is to try to stop everything.

This means that they don't overcommit to anything in particular, so there are always gains available, but (hopefully) nothing that can be exploited by the offense over and over again.

This obviously has advantages in being easier to teach, consistent from week to week, and isn't so susceptible to one bad game. However, it also means that offenses with unique approaches may be able to consistently exploit a mismatch.

If they played the Vikings again, they could commit to a more aggressive approach around the LOS, leaving vulnerabilities deeper downfield (for an ultra-simplistic example, shadow Harvin with one of the safeties wherever he goes, and play a single high safety). In general, the niners prefer not to do this, trusting their superior play to show up over 60 minutes. In this game, it didn't.

FWIW - I didn't notice this while I was watching the game, but Danny mentioned in the Audibles thread that Bowman had a particularly bad game - maybe that resulted in some of the breakdowns

20
by Eric Quanstrom (not verified) :: Thu, 09/27/2012 - 9:18am

Good write-up and good comments. Ironically, Bowman led the team in tackles, with 18.
One thing that hasn't been mentioned is the use of virtually any blitz packages, which the 49ers used sparingly--but effectively-- last year. This year, I can probably count on one hand the times the 49ers have sent 5 or more. Does a defense suffer from being too predictable here???

17
by Jimmy :: Wed, 09/26/2012 - 3:08pm

One way might be to try to force him into staying in to pass protect with heavy pressure - ie rush and hug him. The Vikes do leave themselves plenty of options to use other blockers but in theory you could force them to either abandon Ponder to the blitzing players or keep Harvin in to protect. If this did work the worst option left open should be a player getting the ball way behind the LOS, a well drilled defense should be able corall him for little gain.

18
by Thanks4Nothing (not verified) :: Wed, 09/26/2012 - 3:25pm

If you don't think Unions are political you should try working in China for a while

19
by Karl Cuba :: Wed, 09/26/2012 - 4:58pm

It might be a violation of rule one for you to expand on that but I'd be interested to read it. Assuming that you are aware that unions are proscribed in China as enemies of the people.