Dalvin Cook could end up being the savior of the Vikings offense. With an ambiguous injury timeline for Sam Bradford, the overall skill of the stellar receiving duo of Adam Thielen and Stefon Diggs may not matter. And without a turnover, the defense is unlikely to provide points themselves, like they did earlier last season.
Instead, the tremendous burden falls on a second-round rookie, the third running back taken in the draft and one who must overcome some of the deficiencies of a much-improved but still struggling offensive line.
So — how good has he proven to be, and where can he improve?
Highlights and Film
Cook has put together some fantastic highlights that demonstrate everything about who he is as a runner already. His highlight run against the Steelers that was nearly a touchdown is a good example of how Cook has the ability to improvise, demonstrate second-level vision and employ one of many running styles.
While this could have been a counter play, the way that this is blocked looks very much like an inside zone that Cook wanted to cut back on because he saw a linebacker in the hole. While Pat Elflein eventually moves up to take on that linebacker, Cook chooses the backside C gap before seeing a lurking safety. Cook bounces out entirely in a race to the edge and nearly beats the safety to the tackle point — he chose instead to let that player overrun the play.
Cook could have been patient here, but his quick decision-making moved the defense out of its traditional gap assignments. That kind of movement is difficult to predict and often causes problems for both the offense and defense.
For the offense, they cannot ensure the blocks they need to make when a running back chooses a different path than intended (and quickly). For the defense, it means abandoning the run reads in order to give chase. For the most part, this kind of thing can favor the defense, but Cook made it work.
But he also demonstrates patience. In the next play, he continues to pursue an aiming point that is already lost in order to wait for Joe Berger to hit the next level and create the space he needs to get a run. It’s a way of pressing a lane to get the defense to pursue the wrong gap, but also a way of waiting for his “real” gap to open.
If it wasn’t for Cameron Jordan’s singular talent, Cook likely would have housed this run through the combined efforts of Berger’s seal, Cook’s ability to press a lane he wasn’t going to take and his willingness to wait for the lane he wanted to take.
Cook combines those qualities with superhuman hip fluidity that might end up being his trump card over the years. Below is a clip that’s slowed down and designed to emphasize that kind of fluidity.
Cook has turned this unique physical trait of his into an asset that allows him to constantly move forward as he changes directions.
Below, you can see Cook using that fluidity to escape from Tyson Alualu, who has him dead to rights. It’s been slowed down at the point of potential contact to emphasize how exactly Cook escaped.
Of course, at the clip, you see Cook demonstrating the kind of leg drive that gives him the power most people — myself included — didn’t think he could demonstrate in the NFL.
Cook’s ability to bulldoze through much larger players has been a huge asset, and a big part of making his running game more complete.
He won’t drive through full-form tackles taken head-on in the same way that Mike Tolbert will, but it’s a good asset to have when the run-blocking team is still developing consistency.
Cook has the unusual ability to switch his running style between frenetic and unpredictable to calm and patient. His physical capabilities allow him to present unusual angles of attack and fully take advantage of the benefits of multiple styles of running. His willingness and talent at taking on contact also rounds out his ability.
While Cook can still make mistakes as a runner, particularly with vision (though, surprisingly enough for a rookie running back, not as much with being too bold), he also demonstrates high-level awareness that allows him to manipulate second- and third-level defenders.
Embodying that set of running styles is difficult, and it can lead to problems for the offense as it can encourage riskier and riskier styles of play, but it’s currently an asset to be coveted and cultivated.
We know what Cook can do at his best, but what’s the ultimate impact of everything he does?
Our second look at Cook can be a dive into the data the FSU alum has provided us over the first two games. Advanced statistics are relatively difficult to use for running backs, as it is nearly impossible to parse the influence that offensive lines have on running backs and vice versa.
It is easy, initially, to look at yards-per-carry. Cook obviously scores well there; he’s fourth among running backs in yards per attempt with 5.6. The issue is that yards per attempt for running backs is extremely random and not very indicative of skill.
The correlation in yards-per-carry for running backs in the first nine weeks of the season and those same backs in the last eight weeks is only 0.35 — which is extremely low in same-season terms. For running backs from season to season, it’s as low as 0.11, effectively random.
Jeremy Hill averaged 4.95 yards per carry in the first half of the season and 2.75 in the second half of the season. Jay Ajayi dropped from 5.98 to 4.12, Spencer Ware dropped from 5.01 to 3.66, and Carlos Hyde jumped from 3.94 to 5.18.
There are a number of reasons for this — yards per carry is very sensitive to big plays, dependent on offensive line play, influenced by situation (it’s easier to gain yards on second-and-ten than third-and-one) and complicated by differences in defensive talent.
It’s not difficult to find ways to solve a number of those problems, but it’s easiest to look at what Football Outsiders has done with their DVOA and DYAR metrics — Defense-Adjusted Value Over Average and Defense-Adjusted Yards Above Replacement, respectively.
Both account for down, distance, game situation (grinding out the clock late in the game versus establishing the run early on, for example), opponent strength and so on. DYAR is expressed as a total yardage, while DVOA is expressed as a rate statistic as a “percentage above the average running back in the same situations.”
For our purposes, that’s a good enough explanation for now, but if you’re interested in more, there’s a huge explanation on their website.
He ranks seventh among all running backs in both measures.
That’s extremely useful and could serve as a fine stand-in, but it would be good to do a deeper dive. The most important concepts to me are the notions of success rate and explosiveness.
Success rate, as one would imagine from the name, describes how often a running back’s rushing attempts are successful. This means achieving at least 40 percent of the required yardage to convert a new set of downs on first down (so, four yards or more on first-and-ten), 60 percent of that yardage on second down (six yards on second-and-10, or three yards on second-and-five) and 100 percent of that yardage on third and fourth down.
So a running back on the following series of runs would have a success rate of 50 percent:
This gives us a measure of consistency and overall impact. Run success rate tracks very well with team success as well as passing efficiency, even though it doesn’t give credit to players who are very good at busting out big runs.
It’s easy to think, just from quick recollection, that Cook has been good at busting out big runs but has suffered from some iffy unsuccessful play — especially early in games. That hasn’t quite turned out to be the case, however.
|Player||Att||Yds||Y/A||Total Success Rate|
Cook actually places fifth among NFL running backs (with at least 20 carries) in success rate. That kind of consistency thus far is impressive, especially with a poor offensive line.
And it’s not as if Cook hasn’t been explosive, either; 8.8 percent of his running attempts have gone for at least 15 yards, which ranks fourth among all running backs.
But if he makes his biggest impact on meaningless downs, like third-and-20, it doesn’t mean much. So I’ve devised a metric called “yards over expected.”
For every common down and distance, I’ve figured out what the average rushing attempt has produced and gave running backs credit for yards produced ahead of or behind that mark. So, if a running back creates five yards on first-and-10, he gets credit for 0.69 yards above expected because the average running back gets 4.31 yards on first-and-ten.
If, on second-and-two, a running back gets 80 yards, he gets credited for 76.31 yards because the average run at that down and distance earns 3.69 yards. If that running back actually gets stuffed at the line of scrimmage, he earns -3.69 yards over expected.
Here’s how those running backs rank in that statistic, adjusted to look like normal yards per carry:
|Player||Att||Yds||Yards Per Carry||Yards Over Expected|
He moves from fourth in yards per carry to fourth in yards over expected — not a huge change. That’s distinct from players like Ty Montgomery or Javorius Allen, who lose quite a bit from the fact that they tend to run out of favorable running situations, like third-and-six or second-and-ten.
On the other hand, Mike Gillislee and Kareem Hunt gain yardage by virtue of how their running situations create difficult yardage tasks.
Thus far, as a runner, Cook has been consistent, explosive and ahead of the chains.
Can we separate that from line play?
Football Outsiders tries to do it in a fairly sideways manner. They assign full credit for runs up to two yards to the offensive line (with a penalty for those that end behind the line of scrimmage), and then give successively less credit to the line for further yardage, with the running back getting full credit for runs over eight yards.
It’s not a particularly elegant system, but it is a smart way to approach the problem. We know that that assignation is not quite accurate — running backs have certainly been more responsible for tackles for loss at times, and at other times we’ve seen running backs perform extraordinary individual feats in order to turn a five-yard loss into a one-yard gain… something that would be fully credited to the offensive line in this system.
Nevertheless, it gives us a rough outline of how much credit the offensive line deserves and how much the running back deserves.
|Rank||Team||Adjusted Line Yards||Stuffed||Stuffed Rk|
|11||New York Jets||4.1||15%||7|
|16||Los Angeles Rams||3.99||19%||16|
|22||New York Giants||3.71||15%||11|
|32||Los Angeles Chargers||2.56||34%||31|
This implies that the Vikings are below-average at creating yards for their running back. Not only that, they seem to be responsible quite a few stuffed runs.
There is another measure that can help capture individual performance: broken tackles and yards after contact. Both Football Outsiders and Pro Football Focus measure broken tackles, and PFF also provides yards-after-contact numbers.
Those terms are fairly straightforward, so let’s look at how Dalvin ranks in each of those categories.
|Player||Attempts||Broken Run Tackles (PFF)||Yards After Contact per Attempt||BT/Att (PFF)||Overall Score|
The issue with using “broken tackles” and yards after contact is that good running backs also have the ability to avoid tackles and prevent contact. Still, it’s another way of isolating a running back’s play from the blockers around him. In this metric, Cook ranks just above average.
The final way we can isolate players from the blockers in front of them is by using run-blocking grades from Pro Football Focus. This year, one can explain about 30 percent of the variance in yards per carry with the variance in team run blocking.
The Vikings rank 20th in run blocking grade for their offensive linemen and tight ends (adjusted for snaps) over the past two games, which implies that Cook’s excellent running performances are largely independent of the offensive line’s blocking capability.
Cook hasn’t been good in the passing game thus far. That’s largely because of his receiving issues, though Pro Football Focus will also argue that he’s a poor pass protector.
Vikings fans (and the Vikings themselves) have lauded Cook’s ability to pick up his assignments, but — according to PFF — he hasn’t performed those assignments well.
Cook ties for 13th of 16 running backs with at least 10 pass blocking snap in PFF’s Pass Blocking Efficiency Metric, which essentially divides the number of pressures given up (with extra penalties given to sacks) by the number of pass protection snaps.
It’s such a small sample (he ranks this low because of two total pressures allowed), and I suspect that he may be given credit for a hit that was ultimately Elflein’s fault for his indecision. As a result, he may be getting tagged for a pressure that he’s not responsible for. With that pressure taken away, Cook places about average among running backs.
For now, I’m OK with one pressure given up as he seems to actually doing a fairly good job on assignments that are unambiguously his.
As for his reception capability, Cook has been a liability. In drop rate, he’s tied for 31st of 34.
One might be able to absorb drops from receivers if they make up for it with big plays, but for only occasional receivers like standard running backs, forced incompletions are more damaging. To measure his full impact, one can look at how many yards he’s produced per snap in route.
He ranks dead last in receiving yards gained per pass route run.
If his drop issues can be resolved — and because it wasn’t a big problem in college, they likely will be — Cook will become a true multi-capable threat as he’s genuinely dynamic enough to be a premier back in the modern NFL.
As it stands, Cook demonstrates the ability to run with multiple running styles, find consistency even when the offensive line is struggling, generate big plays and finish with power.
Dalvin Cook is already a pretty good running back.