Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 6 other subscribers

MN Vikings Tweets

Bleacher Report – Vikings

Offensive Line Play: Not a good debut for Vikings’ new O-line

By Derek Wetmore

BY ARIF HASAN

Follow Arif on Twitter

Last week, we pointed out that the Vikings needed to see significant improvement in their offensive line play in order for them to make good on the implicit promise that they’d make a playoff run. Their importance in protecting the quarterback and paving the road for running backs cannot be overstated, yet their influence is often left absent in discussions about team play.

As a result, we’ll be tracking offensive line performance all year.

The difference that a quality offensive line can make is enormous.

Yesterday, we looked at quite a few advanced statistics, but one that pops out for its ease of use and effectiveness at predicting quarterback play was adjusted net yards per attempt (ANYA), which combines sacks, passing yards, interceptions and touchdowns into an easy-to-understand number that is significantly less arbitrary than passer rating.

The best quarterback in the NFL last year in the statistic, Carson Palmer, not-so-coincidentally helmed the league’s second-most effective offense in points per drive. An average quarterback would generate 6.3 adjusted net yards per attempt, which resembles the lines put together by Matt Ryan, Brian Hoyer and Matthew Stafford.

But if that average quarterback put up 6.3 ANYA with average pass protection, he’d suffer significantly with the league’s worst pass protection. If he had had the pass protection the Vikings put together in 2015, his ANYA would come out to 5.7, which would have ranked 28th in the NFL last year. With the best pass protection in the league, his ANYA would have come out to 6.8, which would have ranked 10th.

Offensive linemen are far more important to the passing game than they’re given credit for, and if they alone can be the difference between the tenth-best passing offense in the league and the 28th-best, then they are worthy of serious examination.

Not only that, quick tests (nothing more than simple correlation tests in a spreadsheet program) of seven years of Pro Football Focus data shows that offensive line run blocking can explain between 40 to 50 percent of the variation between different teams’ running ability, as expressed by yards per attempt or Football Outsiders’ DVOA metric.

That’s not rock solid—perhaps PFF graders are unduly influenced by running success and gave linemen better grades as a result of a good run—but it does a good enough job to demonstrate that, generally speaking, offensive linemen are responsible for a good chunk of the running game.

So, how did the Vikings’ offensive line do last week?

Pass Protection

Statistically, pass protection is difficult to figure out. Even with the advent of tracking companies like Pro Football Focus and the increased availability of data from STATS, Inc., our ability to understand whether or not an offensive lineman did well or the offensive line in general protected the passer well.

Many would look at the fact that there were no sacks as evidence that the offensive line played well, but it gives no credit to Shaun Hill in avoiding pressure or getting rid of the ball quickly. It’s a critical quality of quarterbacks to avoid sacks and the difference between different quarterbacks behind the same offensive lines can be massive.

Last year, Michael Vick took sacks on 13 percent of his dropbacks while Ben Roethlisberger and Landry Jones only took sacks on four percent of their dropbacks. In Houston last year, each of the four quarterbacks who took starts had different sack percentages by about two points each: Ryan Mallet (2.6 percent), Brandon Weeden (4.5 percent), Brian Hoyer (6.3 percent) and T.J. Yates (8.1 percent).

Even systems designed to reduce sacks by emphasizing screen passes can see different drop-offs. In 2014, Mark Sanchez started eight games for Philadelphia and took sacks on 6.9 percent of plays. Nick Foles started the other eight games and took sacks on 2.8 percent of his dropbacks.

In Denver in 2011, Tim Tebow took sacks on 11 percent of his dropbacks, while Kyle Orton only took sacks on 5.5 percent of his dropbacks.

And so on and so forth. When quarterbacks switch teams, the statistics that stays the most stable isn’t yards per attempt, completion rate, touchdown percentage or interception percentage; it’s sack rate. The inverse is true as well—sack rates change the most often when the scenario described above occurs and a team changes quarterbacks mid-season.

None of this is to say that quarterbacks are always at fault when a sack occurs, merely that knowing a sack total doesn’t tell us enough about the job of the offensive line—knowing how many pressures they gave up is critical.

That matters more because quarterbacks can rarely control pressure rate, merely whether or not pressure turns into a hit or a sack. Unfortunately, sacks are much easier to measure than pressures and we can run into some problems with the data.

For example, STATS, Inc. only records four defensive hurries from the Titans against the Vikings. The NFL gamebook also includes four quarterback hits (which are not measured in STATS, Inc.’s “hurries” statistic). That’s eight total pressures.

Pro Football Focus has 15 total pressures, likely 11 hurries and four hits. Because many of those pressures occurred on the same play, they ended up crediting Hill with dealing with eleven total dropbacks that had pressure.

ESPN Stats & Info said Hill was pressured on only seven of his dropbacks.

We won’t know what Football Outsiders has on the game just yet, but knowing that information will give us more data about the specific players who gave up and created pressure.

From the perspective of evaluating Hill’s performance, determining how pressure is calculated—and whether or not that is high for the tracking organization in question—is critical. PFF, for example, tends to record higher pressure rates in general than STATS, Inc. does.

It could be (and this is conjecture) because STATS, Inc. only counts pressure if the quarterback is moved off his spot, while PFF counts it for all moments where a defender is bearing down on a quarterback or has moved an offensive lineman.

Having 32 percent of dropbacks under pressure, using the PFF statistic would be better than the NFL’s average of 35.5 percent last year, however, and represents a significant improvement from the offensive line.

I’m not as sure, however. If instead of looking at how often pressure was bearing down on the quarterback, we look at how often offensive linemen are beat on the play, we can get a different understanding of their performance.

I looked at all the snaps from the offensive linemen again and recorded whether or not each offensive lineman was beat in pass protection, while also noting whether or not it was a 3-step, 5-step or 7-step drop (there were three “now” plays that were one-step drops, as well as a rollout).

For example, I’m not sure it would be accurate to say that Shaun Hill was pressured by Derrick Morgan here, but it’s clear that Andre Smith fails from the get-go to contain Morgan. This play would probably still count as a pressure by ESPN, PFF, etc. because Jurrell Casey drives through Matt Kalil so Hill gets hit, but it demonstrates the point nicely.

Smith had an awful game. It’s easier to see in the All-22 than in the broadcast, because the broadcast rarely replayed video from the pocket and cut away all too often to focus on the legitimately much more exciting arc of the ball and receiver/defender battle.

There is a moment in the third quarter where Smith allowed three pressures on three consecutive passing snaps—the final one of which forced Hill’s hilarious scramble.

Later, Smith even gives up a pressure from the left as the Vikings showcase an unbalanced formation (the second of the day) where a tight end plays right tackle and the right tackle goes to the left in order to overload that side with powerful bodies.

There are eight snaps of 36 passing downs (34 attempts, an accepted penalty and the scramble) where Smith hurts the team, either with pressure or a penalty.

Incidentally, Hill was clearly hit on nine passing snaps, not the four snaps ESPN or the official NFL gamebook report. Most of these aren’t found easily in the broadcast but are there in the All-22. I tried to figure out whether or not Hill falls to the ground on the hit is the reason there was a discrepancy in the hit totals, so I recorded that information as well.

Quarter Time Play Description Fall?
1 9:26 Shaun Hill incomplete pass to the left intended for Stefon Diggs. Falls Down
1 0:46 Shaun Hill incomplete pass to the right intended for Kyle Rudolph defensed by Sean Spence. Falls Down
2 7:16 Shaun Hill pass to the right to Kyle Rudolph for 16 yards to the MIN 28. Tackled by Daimion Stafford. Falls Down
2 6:18 Shaun Hill pass to the left to Adam Thielen for 2 yards to the MIN 16. Tackled by Brice McCain. Hit But Doesn’t Fall Down
3 8:57 Shaun Hill pass to the right to Stefon Diggs for 10 yards to the TEN 34. Tackled by Da’Norris Searcy. Falls Down
3 0:22 Shaun Hill incomplete pass to the right intended for Adrian Peterson. Falls Down
4 15:00 Shaun Hill pass to the left to Adam Thielen for 16 yards to the MIN 42. Tackled by Brice McCain. Hit But Doesn’t Fall Down
4 8:26 Shaun Hill incomplete pass to the left intended for Charles Johnson. Hit But Doesn’t Fall Down

There were five times Hill fell as a result of the hit and at least once where he gets hit hard enough to fall but gets propped up by bouncing off of a lineman.

Given that quarterbacks are hit on 15 percent of their dropbacks, even the less generous number of six hits means that the Vikings allowed more hits than a typical team.

Overall, I recorded a failure from at least one offensive lineman on 15 of the 36 passing snaps, or 41.6 percent. That doesn’t mean Hill was pressured over 40 percent of attempts, though.

The Vikings did something in this game that they did with Hill in the preseason but rarely did with Teddy Bridgewater in the preseason or the 2015 regular season: they shortened the quarterback drop.

Including the playoffs, the Vikings took 606 dropbacks on non-rollout passes in 2015. On those passes, they only took a three-step drop on 8.4 percent of attempts and five-step drops on 37.5 percent of attempts. That forced the offensive line to block a more difficult angle for a much longer period of time than other teams. In this game, the Vikings shifted from an 8.4 percent share of three-step drops to a 42.9 percent share.

It protected Hill, and made the offensive line’s job easier. Furthermore, it didn’t prevent deep passing, as Hill’s average passing depth was over 10 yards and among the deeper passers in the league. Play design — plus a quick release — decreased pressure even as the offensive linemen only marginally improved.

Even when pressure came in under two seconds, Hill wasn’t waiting for receivers to get open as he was dropping back to pass. He could get rid of the ball to an outlet receiver and wasn’t shackled to a play that would doom him to pressure or a hit.

To Hill’s credit, he also got rid of the ball quickly and prevented some of those hits listed above from turning into sacks.

Outside of Smith, Kalil had a rougher game than I initially remember, failing on six snaps. Brandon Fusco had three failed snaps in pass protection, all of them in the third quarter. Alex Boone and Joe Berger both only failed once in pass protection.

Overall it was a rough day for the offensive line, but that didn’t materialize into brutal pressure or quarterback punishment. All we asked of them last week was to get better, not even reach average. They definitely hit that mark.

One of Fusco’s mistakes may also simply be a communication error as the nose tackle slants to the opposing A gap and Fusco passes his assignment off to Smith, who cannot logically pick it up

Run Blocking

In the advanced statistics recap, we noted that the running game suffered and that both the offensive line and Adrian Peterson were at issue. Often, it is difficult to parse the different impacts that an offensive line or running back have on running production.

To my eye, it seemed obvious that both were at issue, as the offensive line consistently lost battles at the line of scrimmage, while Adrian Peterson looked hesitant when he should have been aggressive — and aggressive when he should have been patient.

For now, we’ll just look at the offensive line and their contribution to the unremarkable 1.6 yards per carry.

After looking at all the running failures (outside of the scrambles and receiver end-arounds; there were 17 runs that didn’t accomplish their goals), Alex Boone was involved in eight of them. He continued to lose leverage to Jurrell Casey, who was an incredible force in the game. Boone lost against swim moves, push-pull moves and through simple missteps. He lost against strength, agility and technique.

There’s at least one run that wasn’t Peterson’s fault

The Vikings ran through Boone more than any other offensive lineman, increasing both his importance and the likelihood that his failures would be obvious and critical. When running to the right, they would often pull Boone as a lead blocker, and he would occasionally make the wrong decision on those runs.

There are also times where the play design would doom the Vikings. For example, the Vikings motioned Stefon Diggs inside the box twice, which brought additional defenders into the box—who would be better defenders than Diggs is a blocker.

Because Boone and Berger lose their blocks, Peterson is even forced to run in the direction when they had a numbers disadvantage and a skill disadvantage in terms of blockers.

There was no worse player in the run game than Boone and that held true by a significant margin. The Vikings asked the most of him, in fairness, but he didn’t deliver on the burden that was put on him. On a different note, Matt Kalil made the fewest mistakes, but they didn’t ask much of Kalil or run too much behind him.

Andre Smith was the next-best blocker, but he wasn’t particularly great and the Vikings didn’t feel comfortable moving him to the second level or asking him to block outside his gap. The interior blockers, who were largely very good in pass protection, were the biggest weaknesses in the run game.

My initial assessment that Berger made few mistakes in the running game was incorrect, and there were times he clearly gave up critical blocks. He was an important player who made good plays—especially in short yardage running situations—but he definitely allowed himself to be worked back too often to be considered “good” in the running game.

Brandon Fusco was by comparison a ray of light. He’s the only player that seemed to have a lot asked of him while also making relatively few mistakes. He still wasn’t a very good run blocker, but he certainly seemed to be the “least bad” among those who were given relatively difficult assignments.

There’s very clearly more than one play where Peterson was at fault for the short yardage the offense produced, but it’s also difficult to imagine this kind of run blocking producing a league-average 4.3 yards per carry for any running back, no matter how good.

It was an abysmal showing by the offensive line. The only positive seems to be the fact that they were good in short-yardage situations (successful on two of three runs), but the blocking throughout the rest of the game suppresses any praise that can be drawn from it.

Before the game, you would be forgiven for thinking the Vikings would be far better at run blocking than pass protection. They weren’t close to average at either responsibility, but they definitely improved in pass protection while they regressed in run blocking.

Knowing that the worst performer in pass protection (Smith) and the worst performer in run blocking (Boone) are new players might give one hope that the issue may be chemistry or adapting to a new scheme, but with a divisional game coming up, they have little time to fix their problems.

The post Offensive Line Play: Not a good debut for Vikings’ new O-line appeared first on 1500 ESPN Twin Cities.

Source:: 1500 ESPN Sportswire

Leave a Reply

You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>