Intro Note: Football is a matchup and idiosyncratic game where teams change their behavior over the course of a season, players improve, and random results and injuries can skew outcomes that don’t necessarily reflect underlying value.
I personally find that so much of my success betting at this point in the season relies on sifting through the data to understand the sample sizes that are the most relevant/predictive going forward. With 5-6 weeks of data, off-season priors are not yet fully shed, and we are already starting to see teams change their methods of attack and in some cases personnel changes which can make all the difference
By taking an intensive look at our data and understanding the context in which it lives we can (hopefully) stay a bit ahead of the market. This column, then, still an experiment for now (feedback appreciated) will look at some of the stats and angles I’m monitoring as I parse through the data and try to stay ahead of the market. And it should hopefully provide a window of thought into a process of digesting football data and the various applications
And to recap from last week as some are still applicable: (we had some fun with 1) Nix not running when he scrambles 2) the Panthers being legit contenders (still futures value!?) and 3) the Eagles run game and it's problems with a focus on Hurts (very sad).
Anyway, I won’t capture everything (there will prob be typos) and the structure might be different week-to-week, but it’s a start nonetheless and something to think about midweek.
KC:
It is of course obvious that Kenneth Walker has been dominant on the ground for KC. But what is perhaps more interesting is 1) how Walker is winning and 2) how that interacts with the overall offense and how that has changed from year to year for Mahomes.
We have here how often running backs have Perfect Blocks (plays with no blocking mistakes) and Yards Before Contact. In some sense, this is a decent proxy for vision and the ability to to take advantage of open holes. This is an area where Walker has been awesome this year as you see a wide gap between his placement and the trendline.
And as teams have adjusted to load the box against the Chiefs— Walker sees more box defenders than all but 2 running backs— he has maintained his ability to hit the open holes, all while leading the NFL by far in missed tackles forced.
So the running game is good and that is obviously a good in and of itself. But it has also clearly leaked into other areas of the offense making the entire offensive unit more efficient.
If we look at Mahomes through 6 weeks, we see a play action game in the 90th percentile in 2026, following 3 years (the averages are the same in the 5 game sample from these years) topping out at 60th percentile EPA in play action with the run game weaker. Now perhaps it is a coincidence and there is no causal effect, but I’ll go out on a limb here and say the improved running game, more box defenders in conflict, is contributing to a far better overall passing game.
And with ground game allowing them to stay ahead of the chains, Mahomes has also faced (by far) the fewest obvious passing situations (xpass >=75%) of any QB in the NFL, where is dominating with a 0.25 EPA/play.
Now SEA will be a hell of a test for this running game having perhaps the games best run D over the last 2 years, but the Chiefs offense has so many ways to beat you, I’m bullish on them this week and going forward.
SEA:
And speaking of SEA and run games, the opposite phenomenon has occurred for the Seahawks early this year. The run game has struggled mightily, bottom 3 in EPA, bottom in nearly any metric. The result is Darnold frequently in obvious passing situations. And where last year Darnold was seeing only a ~40% obvious passing situations, this year the number is closer to 55%.
Darnold has been unable to overcome the obvious passing spots losing close to 0.2 EPA/play, as he is more frequently under pressure (it is easier to pressure when you know a pass is coming) where he is the 2nd least productive QB.
Likewise, without the run game the play action game has suffered.
Where Darnold last year (again, same thing with same trend if we account for same length sample this year to last) was 80th percentile producing with play action, that has nearly collapsed for the offense and the overall production has tanked as a result. Until this run game can get back on track this is an offense that I think has taken a massive step back from last year and will be in trouble against opponents who can stop the run/bring pressure.
DAL:
The Cowboys defense has been miserable for the better part of the last year and a half now. Last year, this was a defense 31st in EPA/Pass and this year so far are 32nd in EPA, 30th in YPA and at the bottom of basically any pass defense metric.
But there is perhaps some reason for optimism following the Cowboys trade for Joey Porter Jr.
Porter has been tasked with a clear assignment: play press often coverage and win. So far, Porter has been dominant with the best mark preventing separation of any CB in the NFL.
From the Cowboys perspective, it of course helps to have a dominant CB lock out an opposing WR. But what elevates this is the second order effects and how Dallas can now scheme their defense.
With faith that one guy will hold up and lock up a WR, Dallas has been able to shade coverage help elsewhere. The result in 2 games with perfect coverage (plays where everyone fulfills their assignment) above 50% with the previous season high at 25% in this metric. And likewise, the defense last week was more willing to blitz last week which they did a season high 42% of the time, with a 65% percentile EPA performance against the blitz. And this is a defense that has been far better with blitzing than without.
Is the Cowboys defense all of a sudden going to be a top unit with Porter in there? No, of course not. But is it possible for them to improve out of the basement in pass defense? Absolutely. And don’t be too shocked to see the defense improve ever so slightly…. which might make all the difference in a very winnable NFC East.
SF
Some of this is part anecdotal, some of it more data driven, but the 49ers offense has operated entirely different this year moving more away from the YAC guys and leaning into Purdy→ Evans more contested catches in key situations.
Where this is most obvious is on late downs:
To the Niners credit, they haven’t seen very many 3rd and longs this year, but when they have they’re converting them at an outrageous clip. Where is the production coming from?
Well it is Mike Evans. He has an outrageous 45% target rate on late downs and is somehow converting (from that large a sample) nearly 70% of the targets. Without Mike Evans for the next bit, the team will need to find a new weapon to convert— and with the injuries in SF, there aren’t too many possible answers.
But what is perhaps most interesting to me and something I’ll be monitoring, is that without Mr reliable— throw-descievely-in-Evans-direction-and-let-him-make-a-play, Purdy took his some more sacks in the second half of MNF. Where they have been so good staying ahead of the chains, avoiding 3rd and longs etc, I would think, without Evans, the sack rate is going to pop up a bit and Purdy will be a bit less decisive. The offense should still be great (and the defense is a topic for another time as a legit unit) but I think the team takes a bit of a step back without Evans and regresses over the next few weeks, especially on late downs.
That’s all for now! If I have some tails thoughts for tomorrow, I’ll be back in this space, if not will be back with gameplan at the regularly scheduled time. As always, appreciate the support and spread the good word about throwthedamball!
Great post keep these coming mid week