Re-Focused: Broncos @ Vikings, Week 13

| 5 years ago

Re-Focused: Broncos @ Vikings, Week 13

The legend continues to grow and now we are reaching a point where the Denver Broncos are no longer in a situation where the conspiracy theorists can claim that the Broncos are playing Tim Tebow to fail. After a series of low scoring victories coming off the back of stifling defense, a dominant running game and a handful of key plays, Denver won in a game that featured 67 points. Tebow was still the beneficiary of some crucial plays from his defense but his passing as, much as the running game, bailed out the defense at pivotal moments. The Broncos now head the division and a home playoff game is theirs for the taking. Who thought that possible when Tebow took over as starter?

For the Minnesota Vikings this game was as frustrating as it was familiar. Frailties in the defensive secondary and another blown lead as a few dreadful plays by Christian Ponder threw away the chance to derail the storybook season of Tebow. An early pick six and an equally horrible pick in the last two minutes handed the Broncos ten points and the victory, as the Vikings did what no team could in the last month; score more than 20 points on the Broncos defense. Minnesota had the recipe for success but threw the game away. Let’s take a look at some of the performances that set the game up but eventually turned it in Denver’s favor.


Denver – Three Performances of Note

In the absence of Von Miller

If a player’s worth is defined by what his unit does in his own absence then Von Miller’s stock should be on the rise in every single award category out there. Be it Rookie of the Year, Defensive Player of the Year or even MVP; he needs to be at least on those ballots. A defense that has been so stifling for the last month on the Broncos winning streak suddenly looked a shadow of itself against a Viking offense that has hardly been tearing up the league. Without Miller the productivity of the pass rushing nosedived, netting pressure on only 13 pass rush attempts of the 261 times that Denver defenders rushed the passer. That Brian Dawkins (+1.6) was the Broncos’ highest rated pass rushed in Miller’s absence shows the scale of the problem. To the defense’s credit they still came up with some real difference making plays in this game. Mario Haggan’s pick six in the first quarter was crucial for a Denver team that never established much in the opening half, and Andre Goodman’s interception late in the game was the defining play of the game, but in between those two plays the entire defense was shown up by the Viking offense.


Tebow to Thomas

The Vikings came out with a plan to stop Tim Tebow (+4.1); they weren’t going to let him run and they weren’t going to let him complete short passes so he’d have to beat them deep. Unfortunately for the Vikings, Tebow showed early that he could beat their plan of attack and for whatever reasons the Vikings never looked to change things up. The Vikings were probably the most aggressive team Tebow has faced thus far in terms of scheming to take away his running, and it worked well, as Tebow only ran the ball three times, outside of his late kneel down. Two of those runs were scrambles and he netted only 13 yards on the ground. But this left gaps deep and in a defense gutted by injuries, the Vikings simply couldn’t stop Tebow and Demaryius Thomas (+2.7). One touchdown (Q3 3:03) was helped by the Vikings’ fear of Tebow running, but many of the big gains were simply the Vikings allowing receivers behind and Tebow finding them consistently. Tebow went 3-of-4 on deep throws and unlike recent weeks didn’t miss on the “easy” pro throws to stall his own offense. This may not have been his most mercurial performance but it was one of, if not, his most impressive in this winning streak.


McGahee carries the load

With the Vikings clamping down on the running threat of Tebow, the burden of the Broncos’ ground game once again fell on the shoulders of Willis McGahee (+1.9) and once again he delivered. McGahee tied his season high, forcing six missed tackles from Viking defenders and saw his way across the goal-line for only his fourth score of the season; his first since he doubled up on the Raiders in Week 9 and what a timely play that touchdown run was. Just after the Vikings had established an eight point lead the Broncos drove the field on two plays with McGahee capping it off with a sharp cut at the line before scything his way through the Minnesota secondary and diving for the pylon for a score that Tebow capped with a two point conversion. McGahee appears rejuvenated in Denver and is showing some of the best form of his career.


Minnesota – Three Performances of Note

Emerging from the Shadows

He may have gained fewer than 100 yards, he may not have crossed the goal-line and he may not have ripped off a 20 yard run, but this was as good and as consistent a running display as you will see this season. Toby Gerhart (+4.2) put in a display to make Adrian Peterson proud simply by the manner in which he ran; with aggression and a sheer unwillingness to be tackled that has been so absent from Gerhart’s performances of his pro career to date. Much like his Bronco counterpart, Gerhart forced six missed tackles from his opposing defense making a Broncos’ defense, that in recent weeks had been stifling, look somewhat soft. His first down rush at 5:15 in the third quarter epitomized his performance with a purposeful cut at the line of scrimmage and then great power and leg drive in the secondary at the second level taking him to a 14 yard gain close to midfield.


Offensive line lockdown

The Vikings offensive line must surely have breathed a sigh of relief when they found out that Von Miller was absent for this week’s game. Without Miller the Broncos’ pass rush is far less potent and the Vikings capitalized on that as only left tackle Charlie Johnson (-1.8) failed to put forth a perfect game in pass protection as he yielded one sack, one hit and one pressure. Without Miller getting the early disruption on both run and pass plays the Broncos struggled to make much ground on the Vikings’ offensive line and as impressive as any individual display this week was that from Viking C John Sullivan (+3.2). Going up against Brodrick Bunkley (-1.2) he was charged with one of the toughest assignments for a center this season and he showed that his resurgence in 2011 was no fluke. Sullivan had help from some double teams but never yielded an inch to Bunkley and even when the Bronco defender took his helmet off (Q4 4:00) he still maintained his block.


Handing the victory to Denver

A display of such promise at times from Christian Ponder (+1.7) left him as an obvious scapegoat after the game as two errors, both egregious and costly, cost the Vikings victory against the Broncos. He was not perfect for the rest of the game but his play was more than good enough for victory as he made best use of the fine displays from Devin Aromashodu (+4.2) and Percy Harvin (+2.8) to give the Vikings numerous leads, but it will be his two “completions” to Broncos defenders for which his performance will be remembered. On both plays, his pick six and late interception, it can only be assumed that Ponder misread the play or simply didn’t read the defender’s intentions as both passes were straight to the defender,Haggan and Goodman respectively. Haggan hovered under a curl by Aromashodu and Goodman drifted under a route by Harvin, ironically the two receivers who had such big games between those two plays. There was plenty of promise for Ponder to take forward from this game but it will be these two throws that overshadow that promise in the minds of many.


Game Notes

Devin Aromashodu’s 90 yards receiving were his most in a single game since Week 16 of 2009 when he amassed 150 against the Vikings.

DJ Williams was the only Bronco defender to play all 86 defensive snaps for the Broncos this week.

– A number of firsts for Demaryius Thomas this week; his first career 100 yard game, his first career multiple TD game and his first forced missed tackle of the 2011 season.


PFF Game Ball

He has failed to deliver in lieu of Adrian Peterson until this week, but for one week at least, Toby Gerhart fulfilled his promise. He ran with the sort of purpose and aggression he displayed in college, a performance that was deserving of a victory.

Follow us on Twitter: @ProFootbalFocus


| Director of Analysis

Ben joined Pro Football Focus in 2007, and has since been in charge of the company’s analysis process. He also contributes to PFF’s weekly NFL podcast.

  • gesboulder

    While I understand the quantitative characteristics of play-by-play, player-by-player ratings; as N=Snaps becomes larger we should see higher correlations between various ratings and outcomes. The 2-10 Minnesota Vikings have one of the highest rated offensive player grouping (and overall teams) in the NFL through Week 13. Against the Broncos the Vikes get 25.9 total offensive, 0.2 total defensive and 4.0 Special teams player points versus Denver’s -1.0, -19.2, 13.0, and lose the game 32-35? For the season, Minnesota appears to be the NFL’s most over-rated PFF team and Denver the among the most under-rated. The week 13 game highlights this error. 25.9 total offensive points for the Vikes in the Denver game (against a -19.2 defense)? Out to lunch, PFF people!

    • Ben Stockwell

      If individual player performance correlated to team performance then that would be a reasonable conclusion to draw. However the real world situation is that the two are not inextricably linked. The Vikings had comfortably the better of this game but a few quick strike plays and drives wiped that out. On a play by play basis for grading an individual player by our methodology one big play doesn’t wipe out 65 solid to good plays.

      We’re all about individual player performance and not team performance. Our “team” grades are just a sum of the individual player grades and as we all know a team can be more, or indeed less, than a sum of their parts.

  • gesboulder

    I never asserted that correlation meant “inextricably linked” … Surely you are not suggesting with “if” that individual and team performance are uncorrelated? Your concluding paragraph is cliche and patronizing. A +25.9 rated offensive performance is dominating; as is a +30.1 3-phases performance in an individual game. Place that against a 3-phase total score of -7.2 for the winning team and would you stipulate at least that the game is anomylous?

    • Ben Stockwell

      No they are not uncorrelated but it is also not unusual to see one team dominate a game and still lose. It is not a common occurance but at the same time it is not uncommon either. The Broncos profited from a couple of gifts from the Vikings but also put some very poor play by the Vikings to the sword as well. The tale of the tape in terms of consistency was the Vikings controlling the game but making key mistakes that allowed splash plays by the Broncos back into the game.

      That is not to denegrate the Broncos performance, they made the plays to win the game but on a play by play basis they were the second best team on the field compared to the Vikings. That level of disparity is somewhat anomalous, but not in the least bit unheard of.

  • gesboulder

    Thanks Ben, I understand and appreciate your point, but I still think the PFF numbers on this game are not adding up. Consider the scale, PFF is saying that the positive play of the Vike’s offensive players in Week 13 (+25.9) exceeds the cummulative offensive player ratings of either the 7-5 Dallas Cowboys or 7-5 Atlanta Falcons for the entire season (through wk 12). Keep in mind you all scored the Denver defense in the game at -19.2 (which is bad day).

    How many times this season has the winning team had a -37.8 or greater PFF 3-phases team score greater than the losing team? I don’t believe its “not uncommon” – I’m not sure what you mean by “it is not a common occurance but at the same time it is not uncommon either.” If PFF player rating for the game are accurate, then the explanations of the PFF ratings and the actual game outcome also are a problem. I believe you are going with an efficiency argument, that is, how well a team translates its players individual performances into scores. But if this “dark matter” to use a physics analogy of explaining NFL outcomes is that great, it verges on rendering PFF individual player ratings as an exercise in futility. Either this “dark matter” is due to humans (coaching, scheme, or some unmeasured attribute of individual performance) or it is do to randomness (the sum is only greater than the parts when a model is wrong or incomplete).

    I appreciate your thoughts, responsiveness and effort. I regret any tone on my part which may not have expressed so.

  • Neil Hornsby

    When someone first asked me to add up all the individual player rating into a “team” score I told them to go forth…etc.
    It was very much akin to my lack of desire to add up run blocking, pass blocking, receiving and penalties to produce a Tight End score.
    One of the reasons is I felt to have any credence at all they would have to be weighted and I have never been someone who felt particularly qualified (or indeed interested) enough to do that.
    I always wanted to know who pass blocked the best or who was the best run defender; who the best player is depends on your view as to how all these individual facets of play are weighted.
    In the end I was persuaded it was “just a bit of fun” and hence the poorly understood and generally badly used Overall Positional Ranking was born.
    To then take these un-weighted sums and further bastardise them by trying to draw some form of correlation from them is fraught with danger.
    In it’s simplest form it’s the Peyton Manning factor. The quarterback sits above everything else to such a huge degree it makes much of the rest redundant.
    Bottom line: If you want to know who the best pass rusher in the NFL is then you’ve come to the right place.
    If you want to look at team grading seriously I’d recommend you go back to the raw individual grading, devise a weighting to them all, apply these and I think you’ll get something far more to your liking.
    Hope this helps, Neil

    • gesboulder

      Thanks Neil, this is what I have been doing for years. This year I’m using PFF as my base source for player ratings. I start with a talent distribution curve with a mean of 80, 200 points of positional value (inclusive of rotational substitutions) adjust weekly performance for injury and the results do track against NFL Vegas betting lines. Here’s the Broncos:

      2011 Current Roster – Actual Broncos / Depth Chart
      Group Inj Adj Player PosVal PFF-11 M ’12 Model Points Primary
      OL -2.0 Clady, Ryan 8.0 67.9 90.0 74.0 576 LT
      OL Clark, Chris 74.9 60.0 64.0 0 LT
      OL Beadles, Zane 6.0 65.1 76.0 70.0 420 LG
      OL Hochstein, Russ 75.0 66.0 0 LG/RG
      OL Walton, J.D. 7.0 60.0 70.0 65.0 455 C
      OL Ramirez, Manny 61.0 63.0 0 C
      OL Kuper, Chris 6.0 71.1 85.0 75.5 453 RG/LG/C
      OL Franklin, Orl. 8.0 71.1 71.0 73.0 584 RT
      OL Hills, Tony 64.0 63.0 0 RT
      TE Fells, Daniel 3.5 62.8 69.0 68.0 238
      TE Green, Virgil 1.5 76.6 63.0 72.0 108 TE/FB
      TE Thomas, Julius 1.0 74.4 64.0 71.0 71
      TE Rosario, Dante 1.5 87.9 66.0 75.0 113 TE/FB
      RB -4.0 McGahee, Willis 5.0 85.1 86.0 85.0 405 RB
      RB Ball, Lance 2.5 72.5 69.0 74.0 185
      RB -2.0 Johnson, Jer. 0.5 72.7 65.0 67.0 33
      FB Johnson, Quinn 67.1 67.0 0 FB
      FB Larsen, Spencer 1.5 67.7 66.0 77.0 116 FB
      WR Willis, Matthew 5.0 80.2 64.0 78.5 393 WR
      WR Thomas, Dem. 5.0 82.9 75.0 81.0 405
      WR -4.0 Royal, Eddie 62.4 77.0 67.0 0 WR
      WR Decker, Eric 6.0 79.8 80.0 80.0 480 WR
      WR Cosby, Quan 66.0 0 WR/KR
      QB Tebow, Tim 32.0 73.3 79.0 78.0 2496 QB
      QB Quinn, Brady 72.0 72.0 0 QB

      LS Paxton, Lonnie 0.8 86.0 65
      FG/KO Prater, Matt 8.0 83.6 81.0 82.0 656
      P Colquitt, B. 6.0 99.0 81.0 99.0 594
      PR Decker, Eric 2.0 81.0 162
      KR Decker, Eric 0.8 79.0 59
      ST Bruton, David 0.5 82.0 41
      ST Woodyard, Wes. 0.5 82.0 41
      ST Haggan, Mario 0.5 80.0 40
      ST Green, Virgil 0.5 80.0 40
      ST Larsen, Spencer 0.5 80.0 40

      DL Ayers, Robert 4.5 74.3 79.0 75.5 340 LDE
      DL Hunter, Jason 4.5 75.5 75.0 76.0 342 LDE
      DL Bunkley, Brod. 5.5 98.1 84.0 98.0 539 NT
      DL McBean, Ryan 2.0 63.7 76.0 68.5 137 NT
      DL Thomas, Marcus 5.0 71.6 73.0 74.0 370 UT
      DL -2.0 Unrein, Mitch 0.5 64.8 54.0 66.0 32 UT
      DL Dumervil, Elvis 6.5 83.9 91.0 84.0 546 RDE
      DL Harvey, Derrick 68.1 58.0 69.0 0 RDE
      LB -2.0 Miller, Von 6.5 99.1 88.0 99.5 634 SLB
      LB Mays, Joe 5.0 84.5 80.0 85.0 425 MLB
      LB Haggan, Mario 89.8 82.0 83.0 0 SLB
      LB Irving, Nate 65.0 66.0 0 MLB
      LB Muhamed, Mike 63.0 0 MLB
      LB Williams, D.J. 4.5 62.7 87.0 69.5 313 WLB
      LB Woodyard, Wes. 3.5 72.4 71.0 74.0 259 WLB
      DB Bailey, Champ 7.5 92.5 98.0 93.0 698 RCB
      DB Wilhite, Jon. 4.0 84.0 72.0 81.5 326 RCB
      DB Goodman, Andre 4.5 69.3 82.0 72.0 324 LCB
      DB Harris, Chris 4.0 91.8 69.0 87.0 348 LCB
      DB McCarthy, Kyle 0.0 63.0 65.0 0 S
      DB Dawkins, Brian 6.0 91.1 90.0 91.0 546 SS
      DB -2.0 Carter, Quinton 6.0 67.1 68.0 77.0 450 SS
      DB Bush, Rafael 65.0 0 SS
      DB Moore, Rahim 79.8 75.0 76.0 0 FS
      DB -2.0 Bruton, David 65.0 70.5 0 FS

      200.0 Off 75.3 -5.9
      Def 83 2.8
      Calibration ST 86.9 1.7
      Factor 125% Power Rating Against NFL -1.3

      • gesboulder

        Apologies for the horrid formatting, it looked like it was going to post correctly.

        • gesboulder

          Regarding Manning, I do adjust positional values to match schemes and substitution/subpackage for a player’s usage, that is, the 200 positional weighting points are distributed based on snap counts played not just “starters” and also by how a team emphasizes the position, e.g. a 3-4WOLB is weighted greater than a 4-3WLB. In the Colts system, I’d probably give Manning a 34 point positional value (slightly more than the average NFL QB weighting of 32) if he rated a 99 and the two replacement QB roughly 69, the total change in power rating points, 13.8, against the median NFL team rating of 20, accurately measures the Manning effect.

  • mvillar

    So I should’t be using this to gamble? Now you tell me.