How Offensive Coordinator Changes Impact Player Projections


Building a model to project NFL outcomes and statistics requires considerations on many fronts, and understanding strategy is an important one. The offensive coordinator has the lead role in the design of the offense and makes the play calls on game day, and understanding the philosophy a new coordinator will bring is a key factor in accurately predicting which skill players will generate production. These skill player projections are critical for fantasy football in all forms, as well as player props, and understanding what an offensive coordinator change means can create an edge in player valuation.

As a caveat, there are cases where a head coach oversees one side of the ball, but regardless of titles, it’s important to understand the football philosophy behind each offensive and defensive unit. This article will explore the elements involved in projecting offenses as coordinators change. It will use Arthur Smith’s move from Tennessee offensive coordinator to Atlanta head coach, and Todd Downing stepping in to replace him in Tennessee, as examples throughout, and will conclude by using the minor adjustment made as Klint Kubiak takes over for his father Gary in Minnesota as an additional example.

Step 1: Study the Chain of Command

The power structure of teams across the NFL varies, and as coaches move into new roles, it’s important to consider how each new staff will work together, and how they will adapt their methods to work with their new organization. Core principles will remain in place, but the best coaches shift plans to fit players.

When Arthur Smith was Tennessee’s offensive coordinator, he ran a run-heavy offense that featured an elite player in running back Derrick Henry. The Ryan Tannehill-led passing game was also dangerous, but it did its best work off play-action. The Titans finished in the middle of the pack for total offensive plays in 2020 after two years near the bottom of the league, and while pace-of-play is not the only variable affecting total offensive plays, the Titans have seemed to favor a ball possession approach during head coach Mike Vrabel’s tenure.

While that approach made sense in Tennessee, Smith is taking over a notably different roster in Atlanta and is now in charge of decisions on tempo and other overarching strategies. Assuming it’s true that Vrabel favored a bit of a slower tempo, that decision is tailored to Tennessee’s roster makeup, and if Smith wants to move at a faster pace in Atlanta, there’s nothing to stop him. The opposite is true of Downing’s situation in Tennessee, where the Titans are likely to play at a similar pace, though it may be with different personnel groupings.

2. How will the offense work?

Whether or not a team changes offensive coordinators, it is critical to examine a team’s personnel changes to anticipate scheme changes on the horizon. NFL play-callers are constantly evolving their philosophies, and roster moves provide important clues as to what an offense or defense will emphasize in the coming season. The New England Patriots are a prime example; with Jonnu Smith and Hunter Henry added to the tight end room, the offense is set to dramatically change their personnel tendencies from the past two seasons.

In Atlanta, Smith takes over an offense with very different personnel than his team in Tennessee. His quarterback, Matt Ryan, does not have Ryan Tannehill’s athleticism but processes the game at a very high level from the pocket. His running back, Mike Davis, had some good stretches in 2020 for Carolina but is a far cry from Derrick Henry. His offensive line has several good players but has played with more finesse than the overwhelming force of the Taylor Lewan and Rodger Saffold-led line in Tennessee.

The Falcons will also start to acquire pieces according to Smith’s vision. Atlanta may have drafted Kyle Pitts with the 4th overall pick either way, but Smith’s history with tight ends suggests he will craft an ideal role for the rookie, and that veteran Hayden Hurst could play a significant role as the second tight end in an offense that features a lot of 12 personnel, particularly after the trade of receiver Julio Jones. Smith’s heavy use of play-action in Tennessee, coupled with the fact that Matt Ryan won his MVP playing in a play-action-heavy system under Kyle Shannahan, suggests Smith will bring more balance to the offense. It won’t be a shift to the Tennessee run-pass ratios, but Atlanta has thrown the ball at a consistently high rate in recent seasons, and Smith seems likely to balance that out.

In Tennessee, Downing seems likely to shift away from Smith’s tight end-heavy approach to more 11 personnel. Downing ran an 11-heavy system in his lone season as an offensive coordinator in Oakland, and now that the Titans traded for Julio Jones and added Josh Reynolds in free agency, they have the needed personnel. Since his season in Oakland, Downing has worked as an assistant on staffs in Minnesota and Tennessee that emphasized the tight end, so his views may have changed to some degree, but a bet for a trend toward more 11 still seems reasonable.

There are many moving parts beyond those listed, but these are some of the key considerations as to how teams will look in the new year. It’s difficult to translate such information directly into changes in projected numbers, which is where modeling can be useful.

3. Modeling Steps

The adjustments to the model based on these changes are applied in multiple sections of the scripts, but four key areas are the run-pass ratio, the carry and target shares, and the team rating.

The run-pass ratios come out of machine learning models based on drive-by-drive data, and the data can be subsetted and reassigned to approximate a new offense and create associated projections. These changes are made with a focus on the coordinator’s background and the projected quarterback’s style.

After each team’s projected run attempts, pass attempts, and points scored are calculated, the attempts are assigned to skill players in the Carry and Target shares, and their production on those touches against a given quality of defense is calculated. This process depends on anticipating which personnel groupings will be used, and which players will have their touches prioritized. Some coaches want a bell-cow back to dominate their carry share, some want a committee. Some teams prominently use two tight ends, and others neglect the position group almost entirely in the passing game. All of this is fluid throughout the season, so the projected roles are controlled manually based on film study. At that point,  machine learning can do its work.

As a final consideration, the offensive coordinator shift shows up as part of the rating assigned to each unit. There’s no exact formula, but whether the offensive designer is likely to make the most of his pieces has an important impact on a rating.

This ranking is the only area that shifted as the Vikings change from Gary to Klint Kubiak. The Vikings have one key personnel change, but tight end Kyle Rudolph had been relegated to a blocking role under Kubiak and Tyler Conklin is set to get the first shot at those duties. As Klint’s scheme is likely to look like his father’s, and his personnel is similar to what’s been there the past two years, the one change is a slight downgrade due to playcalling, as it will be Klint’s first season in the role.

4. Conclusion

No change in the NFL occurs in a vacuum, and it’s critical to examine the individual and the environment involved. An offensive coordinator change fits these rules, and anticipating what may be planned is a great way to get ahead of the competition.

Website | + posts

Steven Clinton, better known as "The Professor", is a former D-1 Quality Control Assistant (Northwestern, Toledo) who holds a B.A. in Economics and M.S. in Predictive Analytics from Northwestern University. He maintains an end-to-end NFL game projection model and is a film junkie who breaks down the tape of every NFL game.