Welcome back to the Next Gen Advantage, where each week we dive deep into the advanced statistics created through NFL Next Gen Stats player-tracking data to find little edges for fantasy managers. This week, we're diving into route profiles and how they match up with Week 12 opponents, plus some sack projections.
Route profiles and matchups
One concept we've looked at in this space this season is route profiles: the rate at which each receiver runs each route type -- on all routes run, not just targets. Some route profiles are more productive on average than others -- a high rate of deep crossers is a lot better than a high rate of go routes, for example -- but at this point I think those are probably pretty baked in to a receiver's fantasy value unless there's been a significant shift for a player midseason. But there's another factor that changes on a weekly basis: the opponent.
And certain routes have a natural advantage against certain coverages in terms of yards per route run. It's a small edge, but isn't that what we're here for? Let's check out the matchups at the extremes in Week 12.
Upgrade: George Pickens vs. Colts' Cover 1
It's still single-high, but it's been interesting to see that Gus Bradley -- the Cover 3 king -- transition to a heavier Cover 1 defense and use Cover 3 at only a slightly above average rate. But either way it should be to Pickens' advantage on Monday Night Football. I've been critical of the way the Steelers have employed Pickens, with a ton of deep fades and go routes. Though now that we're seeing who Pickens is as a receiver -- someone who struggles to get open but is excellent as a contested-catch receiver -- maybe it makes a shade more sense. Those deep shots down the sideline are better against single-high coverages for obvious reasons, as the lack of safety help creates better opportunities. Deep fades, for example, generate 1.0 and 1.1 yards per route run vs. Cover 2 and Cover 4, respectively, but 2.6 (!!) against Cover 1.
Upgrade: Justin Jefferson vs. Patriots' Cover 1
The subheadline above actually undersells this because the Patriots run both a high rate of Cover 1 and Cover 3, two single-high coverages. Three routes that Jefferson runs frequently relative to other wide receivers are corner routes, medium crossers and deep crossers. All three of those routes share a common trait: They perform better against Cover 1 and Cover 3 than two-high coverages. The crossing routes, well, cross in front of that single safety, and the corner route against single-high is against a defender that doesn't have safety help. New England has a good defense, don't get me wrong, but from a schematic standpoint this should work out nicely for Jefferson to rebound off a 33-yard performance against the Cowboys, assuming health.
Downgrade: Jakobi Meyers vs. Vikings' Cover 2 Zone
Meyers runs more deep crossers (10% of his routes!) than anyone in the league, generally a good thing when we think of his production. But against Minnesota that advantage is diminished because the Vikings run Cover 2 more than any other team. That's a problem because those deep crosses are the least effective against Cover 2, averaging 2.3 yards as opposed to 3.6 against Cover 3. Again, 2.3 is still good, just not as good.
Downgrade: Deebo Samuel vs. Saints' Cover 2 man
Screens, it turns out, perform the worst against man coverage with two high safeties. At first, this surprised me, but I think what might be happening here -- I can't be sure -- is that wide receiver screens against 2-man suffer, given the lack of cushion and a safety in good position. And if that's right, that's not ideal news for Samuel, because no wide receiver runs more screens than him and no defense runs more Cover 2 man than the Saints. Of course, even running the most 2-man doesn't mean they run it a ton -- just 19% of the time. But still, every advantage -- or disadvantage -- matters, so this could be a slight deterrent on Samuel's Sunday.
Sack projections
Let's see if we can predict some Thanksgiving and Week 12 sacks, shall we? I've enjoyed using my sack model in this space at times this year, which uses information that includes a player's past sack production, playing time, pass rush win rate, the opponent's pass block win rate, the opposing QB's sack rate and the spread and total to forecast the likelihood of a sack in a given game. I had to make a couple of early-week quarterback calls here, like subbing out Matthew Stafford (concussion) and returning Kyler Murray (hamstring) to action. Projections could be affected if those are incorrect.
Below are the top 20 sack projections for Week 12 in the form of highest chance to record at least one sack, and I'll call out some notable names below.
Alex Highsmith and T.J. Watt at Colts
Highsmith chance at 1+ sack: 48%
Watt chance at 1+ sack: 46%* (see explanation)
Clearly Watt should be higher than Smith here, but the model isn't perfectly built to handle a large gap of missed time in-season that Watt has had this year. I'd expect Watt's true number to be higher, as opposed to Highsmith's being lower. And so this is a strong forecast for the Steelers' pass-rushers. That's in part because of each player's production, yes, but is also specific to this game. Matt Ryan is taking sacks at a high 7.4% rate this season, and the Steelers are only light underdogs.
Frank Clark vs. Rams
Chance at 1+ sack: 44%
Why is Clark's forecast so strong this week, considering he only has three sacks this season? Most important is that the Chiefs are huge favorites (favored 14.5-points, as of this writing) against what very well could be -- and I am assuming is, for the model's purposes -- Bryce Perkins at quarterback. With little data on Perkins himself, the model assumes a higher-than-average sack rate for Perkins, though even if Stafford plays, he too has a high sack rate this year. Clark's 19% sack rate as an edge rusher ranks 13th this season. If you're wondering why Chris Jones (9.0 sacks this season) and Clark (3.0) have similar forecasts despite the difference in production, one big reason is that Clark plays on the edge, where there are just more sacks to be found.
Sam Hubbard at Titans
Chance at 1+ sack: 43%
Why does Hubbard (3.5 sacks this season) have a slightly stronger forecast than teammate Trey Hendrickson (6.0 sacks)? Playing time. At first, I thought this was just because of when they played: Hubbard is a good run defender, and Hendrickson is not. But even in clear passing situations, Hubbard plays more frequently. Both have a good shot at a sack in Week 12 against Tannehill, who is taking sacks at almost an 8% clip.
Von Miller at Lions
Chance at 1+ sack: 37%
This feels low, right? Miller has recorded one or more sacks in six of 10 games this season. And this is a little low relative to the current odds listed at DraftKings. But there is one reason why Miller is a little less likely to record a sack on Thanksgiving than normal: Jared Goff. While the Lions' year hasn't been great, one nice feature has been Goff's sack avoidance, as he has only been taken down on 4.3% of dropbacks.
Seth Walder is in his sixth NFL season as a sports analytics writer at ESPN and previously covered the Jets and Giants for the New York Daily News.
NFL player tracking, also known as Next Gen Stats, is the capture of real-time location data, speed and acceleration for every player and every play, on every inch of the field. Sensors throughout the stadium track tags placed on players' shoulder pads, charting individual movements within inches.