Week 8 Hitchhiker's Guide

Some thoughts on RB efficiency metrics, and making sense of an HHG fallen star

NOTE: If you are a paid sub, the TAT Dynasty Ranks are updated as of today

I tweeted this week about my frustration with the Colts’ continued RBBC deployment and was met with a lot of push back defending the decision on the basis of Zack Moss’s impressive play this year.

My favourite form this push back took was people accusing me of whining about fantasy, which is one of ‘football twitter’s’ favourite tropes. It’s usually a ridiculous retort to me in the best of times, because what’s good for fantasy football is frequently also what’s best for the team. Have you watched the best offenses in the NFL lately? The top 4 offenses in the league by EPA are a tier in front of the rest of the league. Three of them in particular - Miami, San Francisco, and Philadelphia - are three of the most concentrated offensive schemes in the league. Each team has multiple playmakers, those playmakers play a ton of snaps, and they receive lion’s share of the touches.

Clearly there is a chicken and egg dynamic at play here that I want to acknowledge. Not every team has an A.J. Brown, Devonta Smith, Dallas Goedert and D’Andre Swift on their team worthy of consolidating toward. So naturally the teams with the best play-makers are also going to be the teams who consolidate the most, and the ones who score the most points. But this level of consolidation is also a choice.

The Atlanta Falcons are a great example. They certainly don’t have the type of QB play those teams have, but they do have three dynamic playmakers in Bijan Robinson, Kyle Pitts, and Drake London. I would argue that when your QB is Desmond Ridder (or Taylor Heinecke), you are even more incentivized to focus your scheme around your top skill-position players who have a chance to flip the field for you and provide the type of explosive plays which will mitigate the inevitable negative plays and inconsistency you’re going to get with that QB room.

Of course Atlanta does precisely the opposite of this, putting Kyle Pitts on the field for just 60-70% of the pass plays, and allowing Bijan Robinson to be regularly out-carried by the solid but unspectacular Tyler Allgeier.

The problem with this approach is that it starts from the wrong mindset. Theoretically, Arthur Smith allocates his personnel in this manner because he does not trust Kyle Pitts as a blocker. To be honest, I have no educated opinion on Pitts’ blocking ability so I’ll take it at face value that he’s a negative in that area.

In theory then, if you can’t trust Pitts to block, you don’t want him playing on run plays. And because you don’t want to tip your plays, you instead allocate snaps based on personnel packages: no Pitts in 12P for example because it’s typically a run-based personnel grouping, and you keep him out on 12P pass plays, because if you subbed him into 12P sets only when you passed out of it you would tip the play call. Add up all those pass plays from run-focused personnel groupings and you wind up with your most, or second most, talented pass catcher missing 35% of the pass plays.

This mindset is how someone identifies one fairly small problem, and then commits multiple levels of accommodation to to solve that issue, without any consideration for the fact that you’ve created far larger problems along the way. When you take Pitts off the field to ‘disguise’ your pass play, you are taking away one of the best players on the field not only from yourself, but from the consideration set of the defense. If you’re Drake London and you’re the primary read on the play, would you rather the defense be choosing how to allocate their resources between stopping you and stopping Kyle Pitts, or between you and Van Jefferson?

Since there is nothing the appeal to authority football twitter reply-guys love more than results-based thinking, consider the last several weeks of Falcons football: in the Bucs and Texans wins, the game-winning drive consisted of explosive plays to Pitts and London respectively putting them in position for the game-winning field goal. This week their 4th down desperation attempt found its way into (and out of) the hands of Van Jefferson.

You probably get enough Falcons bashing for a lifetime from Stealing Signals every week so I won’t go on any longer about that team, but they’re unfortunately the best example to illustrate my point: which is that the sharpest coaching staffs in the league right now are building their offense around their best players, and finding ways to get them the ball in a way that sets them up to succeed. The worst staffs are focused on the weaknesses of their best players, and capitulating to the defense by spreading the ball to other players and letting them off the hook before they prove they can stop it.

Getting back to the Colts example, the ironic aspect of these angry twitter comments is that I am of course never approaching Colts games from a fantasy-first perspective as I’m in fact one of the “laundry bros” myself.

I am a huge fan of Shane Steichen and I have a lot of confidence ‘we’ have the right coach over the long run. But I was irked by his approach this week in going away from Taylor (just one second half touch) due to what he perceived as a ‘hot hand’ in Zack Moss (never mind the fact Taylor had 12 carries for 95 yards). He also blamed the game script, and to his credit the defense simply could not get stops and the game snow-balled rather quickly. But from my perspective, the preferable approach when you have a star player is to give them a chance to keep you in a game like this. I’ve heard people call it a “players not plays” mindset, and for me, I’d have liked to see the response to falling behind rather quickly in an important game be: how do we get the bal into Taylor’s hands with a chance to give us an explosive play?

There were good faith replies to my post emphasizing how well Zack Moss has played and I want to acknowledge that. He’s been excellent this year. But I also reject the notion that we have to play prisoner to the statistics of the moment.

Zack Moss is in some ways an ironic cause for the appeal-to-authority types on twitter to champion. If he was smaller and faster I’m sure people would write him off as some sort of aberration of pixie dust created by nerds, but since he’s big and slow and owns the libs I suppose his RYOE is allowed to be taken at face value. The “hot hand” theory effectively asks us to believe that the most recent results are the most important results, and we have to allow ourselves to be guided by those until the results change. This is a reactive mindset.

Moss has been legitimately impressive this year, I want to emphasize that again. He’s 3rd in RYOE/A in the NFL and 12th in Yards after Contact per Attempt (YAC/A). But if we give ourselves the benefit of critical thought, we can watch some of his plays and ask ourselves what we think would happen if Jonathan Taylor was carrying the rock instead. On his un-touched 51-yard TD run vs. Tennessee, I promise you Taylor would have scored. On some tough carries between the tackles? Maybe Moss would have more success on a couple of them, or at least comparable. But what happens when a Moss carry is untouched for 20 yards, before being chased down from behind on path to the end-zone. What is going to happen? Who do you want with the ball in their hand?

We should have the courage in our conviction to say the clear and obvious answer to that question which is the 230-pound back who runs a 4.39 and is one of the most explosive runners of our generation.

This all leads me to my next airing of grievances which is the recent popularity of EPA/rush, or as it’s occasionally called “RB EPA.”