Originally Posted by :
The way analysts evaluate and talk about football at the NFL level has changed dramatically over the past decade or so. Access to additional data, rule changes and a greater willingness to challenge conventional thought has led people to question axioms that were once considered absolute truth.
Nowhere is this more true than how the running back position is viewed. While there’s still some debate, it has become widely accepted by now that running backs are more easily replaced within the framework of an offense than other skill positions. Additionally, common wisdom has shifted from “run to win” to “pass to win, run to bleed the clock at the end.” This is reflected in the contract negotiations of stars like Melvin Gordon and Ezekiel Elliot, who find teams reluctant to pay them top-tier money despite their production.
Recently, The Athletic’s Ben Baldwin dug even deeper into the numbers behind the passing game, examining the latest trend of teams trying to get running backs more involved in the passing game.
It’s an excellent article that’s worthy of a pause and read if you haven’t yet. The premise is that targeting running backs is inherently less efficient than targeting wide receivers or tight ends, and by a wide margin. Because of this, it doesn’t make sense to do it any more often than it takes to keep defenses “honest” and respecting the threat of a throw to the running back.
While the ideology can be debated as to whether running backs’ inefficient numbers define their contribution (I happen to believe that individual play efficiency isn’t the entire story in the macro, but that’s for another day), there’s no doubt about one thing: teams don’t do a good job getting the ball to running backs in situations that lead to efficiency.
Well, most teams. But not Andy Reid’s teams.
There’s a lot to unpack in those numbers. “EPA/play” stands for expected points added per play, which is discussed in some detail here. The short version of EPA/play is it tracks how much more likely a team is to score after a given play, with a higher number being better. Success rate tracks how far the ball moved in conjunction with the first-down marker, with a “successful” play meaning the team gained adequate yardage to make a first down more likely.
It’s impossible to ignore how much more efficient the Chiefs are than any other team in the league when targeting running backs on 1st and 2nd down. Only the Rams were even close to the Chiefs last season, and even that’s stretching it, as the Chiefs were more than one-third more efficient. Kansas City and Los Angeles were the only two teams to have better EPA/play and success rate when targeting running backs than what the league averaged targeting wide receivers and tight ends. In other words, under Reid, the Chiefs are able to efficiently utilize a whole segment of offense that other teams cannot.
These numbers aren’t the most staggering when it comes to Reid’s ability to generate yards when targeting the running back, though. A new Chiefs-centric analytics Twitter account demonstrated that when 3rd down was included, Reid’s dominance over the rest of the league in utilizing running backs in the passing game becomes even more pronounced.
The Chiefs are so far ahead of the rest of the league that it becomes comical at this point. They’re actually easy to miss if one doesn’t click on the picture in the original tweet, which is a good sign of being an outlier.
One could make the argument that because the Chiefs had a star in Kareem Hunt for much of last season they had an inherent edge when it came to targeting running backs. However, Arrowhead Analytics was kind enough to isolate the data for Hunt, Damien Williams and Spencer Ware. The result was demonstrative of the fact that when it comes to targeting running backs, there’s Reid, then there’s everyone else.
While Hunt had a higher EPA/play than Williams or Ware, the latter two had a slightly better success rate. Given the small sample sizes involved, some fluctuation is to be expected. However, the important thing to note is that every one of the Chiefs’ running backs stands out as at or near the top of the league in efficiency stats when targeted. Interestingly enough, if fullback Anthony Sherman were added to the chart — he had only eight targets, so he didn’t make the cut — he would be right next to Hunt with more than 0.75 EPA/play. So even fullback, a position many consider to be dying, is significantly more efficient when targeted in Reid’s offense than other teams’ running backs.
So how does Reid do it? A big part of it is the way he schemes running back targets. When Reid designs a screen or other RB target, he doesn’t simply throw them the ball in space and tell them to make a play. His plays are intricately designed to leave a defense hesitating and guessing as to where the ball will end up and what direction it will be traveling.
This angle allows us to at least imagine what it’s like to be a defender here. First, the Chiefs threaten to go left (defenders’ right) with a jet sweep to receiver Tyreek Hill that must be respected. Then there’s a play-action fake to Hunt up the middle. Then another fake, this one threatening an end-around to Demarcus Robinson around the left side. Finally, Patrick Mahomes dumps the ball off to Hunt.
Watch the movement of the defenders at all levels of the field. Some are carried to the wrong side of the play following their coverage assignment. Others, such as the deep safety, are forced to wait … and wait … for the play to develop, moving one direction then another as the fakes materialize. This gives the offensive linemen time to get into space in front of Hunt. When the defenders realize where the ball is ultimately going, it’s too late and the Chiefs score.
Reid doesn’t just use fakes at the snap to get defenses moving the wrong direction. He also uses “clear-out” routes to try and lure the defense away from the direction the play is actually going better than any other coach in the league. It helps that he has a quarterback who understands how to use the tools Reid puts at his disposal.
Here, Reid initially stacks most of his playmakers on one side of the field. He brings Hill from right to left, which forces the defense to show that it’s playing man coverage as the cornerback travels with him. The Browns are threatening blitz, so Mahomes knows that with the receivers on the left side of the field all clearing to the right side, the worst-case matchup he’ll get with Ware out of the backfield is a linebacker coming from out of position at the line. Best-case scenario is that the linebacker blitzes and Ware goes uncovered with no one near him.
After the snap, the Browns blitz and Ware is all alone after Hill and tight end Travis Kelce drag their defenders to the right side of the field. The safety (in typical Gregg Williams fashion) is playing a mile off the line of scrimmage, and the result is a big gain for the Chiefs. This is partly due to Mahomes recognizing what was developing before the snap, but it was also largely Reid’s design taking advantage of the exact defense that was called.
Reid doesn’t just use his playmakers for misdirection, though. He also employs the offensive line to sell deception and get defenses looking at the wrong place on the field.
On this play, Reid again employs a fake end-around in order to get the defense guessing, as well as an initial play-action (he loves utilizing play-action, then throw the ball to the running back). However, the line plays a part in selling the fake as well. On most screens, the line quickly gives up its blocking in order to try and sprint down the field immediately. This can serve as a tell of sorts. Here, much of the line stays in to block as though it were a typical pass play.
Adding to the deception is Kelce, who moves right to left at the snap as though he’s the lead blocker around that edge. The linebackers follow him, and the result is perhaps the easiest 20-plus-yard gain I’ve seen.
When discussing Reid’s screen game, former Chiefs offensive lineman Geoff Schwartz — who has a podcast with The Athletic that is a must-listen — stated bluntly why other teams can’t copy Reid’s success: They don’t have Reid.
This tweet led to a fascinating discussion between Schwartz and once-again Chiefs offensive lineman Jeff Allen, who compared Reid’s system to other coaches in a favorable light.
The reality is that Reid is simply better than other coaches at designing and teaching the screen and short-yardage game. He pays more attention to details and appears to have accounted not just for what the playmakers are supposed to do, but everyone else on the field. That sort of obsessive attention to detail results in plays like the ones above. Moving the football shouldn’t look that easy at the NFL level.
Reid places his running backs in a position to succeed with alignments and routes as well in the screen and short passing game. He was one of the first NFL coaches to send his running backs vertical, something NFL defenses aren’t always prepared for. He’ll also frequently line up his running backs in the slot or the boundary to give them the opportunity for a mismatch in space.
As Reid’s tenure in Kansas City continues, the narrative around him has begun to shift. Rather than focusing on perceived playoff failures or more well-deserved criticism of clock management, people have started to appreciate that Reid is one of the most innovative offensive minds of the past 20 years. The availability of analytics that demonstrate how superior he is at dragging offenses to efficiency in inherently inefficient parts of the game should only further that conversation.
Loves me some Andy Reid more than I have the patience to explain here
His first act of genius was taking a legit NFL shithole on a 9 game winning streak, and it’s only gotten better from there
His guys love him without fail, and he now has the ultimate weapon who can make his craziest, most diabolical ideas possible... we should be embarrassed to have it so damn good
We had lots of issues today, but I didn't feel like playcalling was one of them. You can nitpick the clock management at the first half, but overall Reid was solid as usual. [Reply]
Originally Posted by DaFace:
We had lots of issues today, but I didn't feel like playcalling was one of them. You can nitpick the clock management at the first half, but overall Reid was solid as usual.
Play calling is a really tough thing to judge unless it’s really good or really bad. Most of the time it’s somewhere in the middle.
Fact is, poor execution can make the right play call look like the wrong play call. Great execution or superior talent can make the wrong play call look better than it actually was. [Reply]
