After getting access to the coach's film and taking a look at the game, I think it's appropriate for that reality check we all wanted to avoid just days ago.
It was a fantastic game, and a real thrill, but Vikings fans need to be prepared to understand: the offense has improved, but not by much. There were some encouraging signs by the defense as well, but also some issues that needed to be fixed. The Vikings are a growing team, and with that comes mistakes, responsibilities and changing expectations.
The Vikings were able to pull out in the clutch, which is a huge psychological difference from last year, but there are things to work on. Below the jump, we'll take a look at the standout stars, the underrated workmen, the overrated show-stealers and the folks in need of obvious improvement.
A year ago, this was a unit that found itself productive early on (both in games and in the season), but tapered off later, particularly when it mattered. Knowing what had happened a season ago, even hopeful Vikings fans turned cynical as Cecil Shorts grabbed that last wobbling spiral for a touchdown with twenty seconds left in regulation.
Turning it around was no mean feat. Credit goes to Devin Aromashodu, Adrian Peterson and Kyle Rudolph for excellent performances, and Bill Musgrave for calling the plays that needed to be called at the end of the game.
Speaking of Musgrave, we can break down the personnel packages. While the snap counts are publicly available, the situational packages are not. After hearing so much about the improvement and installation of a two tight end offense, what did we see?
Using my own counts, I found 15 instances (of 59) of two tight end sets. I found three instances of three tight end packages. The snap counts published by NFL media list Rudolph as being there for every snap, Carlson there for 18 snaps, and Ellison being there for three snaps. That concurs with the Pro Football Focus list.
These are two implications I can draw from that: 1) the two tight end package isn't that important or 2) Musgrave and Frazier are not confident enough in Carlson's return from injury to give him more snaps until he takes more reps with the team.
I'm willing to bet the second is more likely than the first; the Vikings did a lot of two tight end installation during training camp and seemed to use that versatility to confuse defensive playcalling.
On 24 first downs, the Vikings used two tight ends 7 times, and one tight end the other 17 times. My count of first downs is different than the box score, and I'm not particularly interested in finding out the differences, but it might be because I counted snaps with penalties and snaps before and after time outs. I doubt the box score did.
On all five instances of "third and short" that I counted, we used an "11" package, which is one running back and one tight end (and therefore three receivers). On the only third and long that I recorded this information, we also used an 11 package.
The 11 package is interesting, because the Vikings have done a lot with it. On four or five instances (not exclusively on third down), the Vikings formation was a "full house" formation, also called an inverted wishbone:
We had Percy Harvin line up as the deep back all but once, if my notes are correct (Adrian Peterson was the other one), and had Felton as one of the fullbacks with Rudolph as the other. The run/pass split was only a little run biased, with Harvin running routes out of the backfield on the pass plays.
The Vikings also used the 11 package in no-huddle situations, including the two-minute drill and the "20 second drill." It seems as if the Vikings find this to be the best way to move the ball.
My full counts—"11" package: 23 times, "21" package (two running backs or fullbacks, and one tight end): 13 times, "22" package: 8 times, "12" package: 6 times, "23" package, 3 times. That only adds up to 53, not 59. I do not know where the extra six snaps went.
I remember people in the stands bemoaning Musgrave's creativity and/or playcalling, but I thought this was an extremely well-called game from his end. We saw more creative uses of the most dynamic player in the league (Harvin), we saw more formations out of the same personnel packages, we saw an offense designed to both take advantage of Ponder's strengths and what the defense gave him, new wrinkles of old formations and my favorite, excellent use of pass/run alerts to bait the defense.
Receivers ran routes without traditional landmarks (the numbers and hashes), which exposed holes in zone coverages, tight ends were bounced around the formation (inside the numbers, outside the numbers, in the backfield, etc), and deception was a constant feature of the offense.
As Carlson gets more integrated into the system, expect to see more two tight end sets and different playcalling tendencies. An interesting note, incidentally: in at least one "22" call, Phil Loadholt lined up as the inside left tight end (on the left of Matt Kalil), while Rudolph lined up at right tackle. The fullback was set offset left, and the play was indeed a run to the left. Sometimes, you avoid deception and put strength on strength, and trust your guys to come out on top. On this play, they did not.
Many of you know that I have defended Christian Ponder in a big way during this preseason, and right now I am saying in all honesty, I was not extraordinarily impressed by him in this game. Yes, his completion percentage was unfathomable (74%, 5th highest in the league) and his yards per attempt was unreasonably high (10.0, 4th highest in the league), and I don't care one way or another that he threw zero touchdowns.
He made mistakes too, and not just in the first half.
By my count, he had six plays that were good and six plays that were bad. Here, "good" means something beyond what is the minimum standard for a play (in decisionmaking, ball placement, ball velocity, etc) by a quarterback in that situation and "bad" is the opposite. Both sacks and his one yard run were "bad" plays, because all three of those busted plays were on him without much pressure prompting the movement into the sack or tackle.
One or two of these bad plays "should have been picks," for what it's worth, but is more importantly a sign of poor decisionmaking. The first was wildly off-target, and could have been the result of an early communication problem. I mentioned earlier that receivers were not using landmarks as often as is common, and Ponder threw to the hash marks, when Harvin was several yards outside of them.
He didn't attempt a throw over 29 yards, and only two over 20. Five of his passes were screen passes to Harvin, who produced all 65 of those yards (along with his blockers).
Ponder ranked fifth of all quarterbacks in percentage of attributed passing yards coming after the catch, and this had very little to do with leading his receivers. He often threw to receivers wide open (which is yes, what he is supposed to do), and also did not have to make many reads.
The playcalling allowed the receivers to carve up the soft underneath zones (sound familiar?) in spots on the field that less experienced cornerbacks don't often find receivers. Remember that awesome 2-minute drill that sparked the team right before halftime? Three of the four passes were screen plays to Harvin, and one of the plays was a run by Peterson. That pass to Jenkins to set up the touchdown was pretty good, though. It was a one-step shotgun drop, so he didn't have the time to stare down a receiver, but he had enough zip on the ball and good ball placement to prevent defenders from getting to it.
That does not mean I didn't like what I saw. I loved what I saw. He was calm under situational (but not pocket) pressure, put the ball in the hands of playmakers, took what the defense gave him and improved in big ways—he has more ball velocity, quicker decisionmaking and avoided many more mistakes than before. He improved over his play last season, but is still not in a spot that many fans may want their quarterback to be.
He needs to look off receivers more, and we'll need to see more tests of what happens when his primary read is covered.
If the Vikings can keep playcalling like they did, this is a way to win games. They cannot use this gameplan again, however, because if it is repeated, it's fairly easily combated by inside man coverage. Ponder did what was asked of him, but was not extraordinary in this respect. Until he improves, he is not the one carrying the offense, and people should remember he wasn't asked to do much against too much.
He threw left 7 times, completing 5 of them for 70 yards. He threw right 6 times, completing 5 of them for 56 yards. He threw up the middle 10 times, completing 6 of them for 182 yards. He also threw to outlet receivers behind the line of scrimmage 4 times, completing all 4 of them for 62 yards.
I was very impressed with the offensive line. None of the sacks should be credited to them, and they gave Ponder time while also grading the road for Harvin, Gerhart and Peterson. John Sullivan was a pleasure to watch, and he made a number of blocks on the second level. He was extremely useful on screens and was perhaps the best offensive lineman that paved the way for the rushes up the gut. It's hard to overstate his value in the run blocking game, as it was nearly perfect.
If there was any lineman who played better than Sullivan, it was surprisingly Brandon Fusco. He didn't leave his blocks to get to the second level (which was not his assignment), but he did drive defensive players all over the field nearly at will. He got low and he stayed mean. Neither he nor Sullivan gave up a pressure, and they were both extremely impressive as run blockers.
Loadholt improved over his 2011 form, and was also good at sealing the edge or preventing strongside/backside defenders from making an impact on runs. He didn't do anything too impressive, but he was functional in a way that Vikings fans aren't used to. I did see him give up one quarterback hit.
With a relatively quiet game (a good thing for offensive linemen), Matt Kalil didn't give up a sack, hit or hurry. He was not as effective in the run game has much more work to do here. What's interesting about it is that it seems as if Kalil either had phenomenal run blocking (like on the third play of the game, with Percy Harvin's run left) or horrid run blocking, with complete whiffs (like on Gerhart's run at 2:07 left in the first quarter or Peterson's second run in the second quarter) that lead to contact. Very streaky, but it seems as if much of the confusion had to do with assignment more than anything else. Once he locked on, he was great.
Charlie Johnson was not a good run blocker and had some problems as well. He held up in pass protection in a very big way, and also gave up no pressures, but could not open up or sustain holes for the running backs.
Overall, the line played extremely well in pass protection, and was great in run blocking. Good to see, for the first time in years.
Harvin was the best offensive playmaker on the field for either team. He was playing outside the numbers, in the slot, as a returner and in the backfield, running nearly every route and acting as an effective rush option. He showcased excellent block reading ability on his screens, and displayed surprising strength on some of his plays, pushing players much heavier than him in order to gain extra yards.
The young utility receiver did a good job finding space and running precise routes despite the lack of landmarks. He was a reliable option and didn't drop a single pass. His run blocking could have been better.
Michael Jenkins was good in space, too. He was an effective blocker downfield and when plays broke down, and was the possession receiver the Vikings signed him to be. He made veteran moves to get open, and was particularly effective on a bootleg play to the right where he sat between several zones to allow Ponder to make the play (in the third quarter, at 4:35). His best play was in the fourth quarter, with 9:35 left to go—he ran an outside route and his savvy route running baited the corner in biting outside for half a second too long. The CB tripped, but that would not have changed the result of the play, which included a tip-toe move to stay inbounds to maintain possession.
His worst play was in the second quarter with 5:21 remaining, where Ponder threw a bad pass to his back shoulder and Jenkins fell on the adjustment. Both of them are at fault on the play, although Ponder more than Jenkins. Jenkins' penalty in the fourth quarter should not have been called; he blocked above the waist.
Devin Aromashodu showed up when needed the most. His two biggest assets—speed and precise route running—were enormous factors in his three catches. The first was a great pass to DA, who ran a route designed precisely to attack the area between zones, and Aromashodu's high-pointing of the ball completed the process to guarantee the catch. The other two were good routes, but he was helped by poor defensive playcalling and a bad secondary unit.
Kyle Rudolph had a fine game, although he did have Ponder's only drop of the night (giving Ponder a relatively low drop rate in the league right now). That drop was on third down and would have converted it in the red zone, possibly (though unlikely) for a touchdown. He has good hands, but he needs to catch those specifically. Other than that, his adjustment to the ball in the air was great overall, and he provided a reliable option. His route running is OK, and it shows a lot of faith that Musgrave is willing to line him up outside as often as he does.
Rudolph was an OK blocker in the run game. He was great in containing backside blocks on run plays, but did not always take defenders on the playside out of the game. Pro Football Focus has a much harsher evaluation than I do of his run blocking game, and I am OK with that. He was inconsistent, but not entirely bad. I did not pay as much attention to his pass blocking plays as I should have, but I've been hearing bad things about it. For now, I will trust that evaluation. If it is the same problem as last year, it is scheme communication and pad level that provided most of the problem. Find your guy, get low.
Carlson was only in on 18 snaps, and was a relative nonentity. I did not like his run blocking and he did not manage to find separation. As he gets back up to speed, he should do better. His pass blocking was fine, but ultimately unnecessary, given the amount of times he was required to pick up a blitzer (once).
Rhett Ellison was in on three plays, so that's something. In his limited time spent doing fullback duties, he was bad at it. It was part of the reason that Gerhart's first play was for no gain. He also massively screwed up the lead blocking on his second play.
Adrian Peterson is back.
He's not completely back, but he has maintained almost all of the qualities we have seen in his running over the years. Also, I don't care what he says, he is better running behind a lead blocker. Period.
AP has been looking for the home run since his days in Oklahoma, but he has clearly maintained his improvement over the years in reading his blocks and staying patient. There were still a number of plays where he did not follow his fullback, and this was a good decision as often as it was bad. He did not set up all of his blocks, but he did set up a good number of them.
As much as the Vikings wanted to reduce his cutting and planting with the left foot, his style of running didn't pay particular attention to where the Vikings wanted him to run the ball. He planted just as well with his left foot as his right, so does not seem to be slow or tight in this respect. He still needs to get back up to 100%, however.
Had Stephen Burton finished his block on the 20 yard run in overtime, Peterson would have had a touchdown and 40 more yards, incidentally.
Toby Gerhart, largely as a result of random noise, got stuck with worse run blocking. He only had six running attempts, but did as could be expected given his blocking. I mentioned Ellison's need for improvement in this respect, but should also mention that with Felton, Carlson and Fusco leading the way, Gerhart had some really good gains, including a 7-yarder, followed by a 6-yarder a quarter later (Felton was the fullback, Fusco and Carlson pulled. The plays were almost identical). There was only one play (of six, granted...) where I thought Gerhart underperformed (2 yard play, with 2:22 remaining in the third—he incorrectly bounces outside).
Jerome Felton was a beast. I only saw one or two plays were I did not like what he did, but none of those plays were problems with the technique of his blocking, and entirely about his reads on the play. Minor things that don't affect his overall performance.
There's not much I can say to emphasize how effective his performance was on Sunday except for this: if he continues like that, not only will he play himself into another contract, he'll be the best fullback in the game. If he plays like that for an entire year, he will be the perfect replacement for (dare I say it?) Jim Kleinsasser.
He burst through holes and held up good run-stopping linebackers, including downhill backers like Paul Posluszny and Kyle Bosworth. Felton dominated his blocks and held them more than long enough to conclude the play. He made the run offense go.
Offensive MVP: Percy Harvin
Unseen Offensive Player of the Game: Jerome Felton
Offensive Rookie of the Game: Matt Kalil
Offensive Honorable Mention: John Sullivan
Later this week, if I have time, I may be able to diagram a play, much like I did here. Which play would you like me to take a look at?
Defense and special teams later (where I will mention Kalil's block). Either tonight, or tomorrow morning.