NFL Plus/Minus: Is Christian McCaffrey or Tyreek Hill the Offensive Player of the Year?

Using propriety NFL Plus/Minus methodology to concretely assign value to all non-quarterback offensive players

I (re)introduced my favored methodology for measuring player value last year, and then applied it to all the non-quarterbacks in the NFL. As the title hints towards, Hill and McCaffrey are two of the most valuable players by this methodology this year. Hill was at the top last season on the offensive side of the ball, and he’s taken his production and efficiency to the next level in 2023. His OPOY campaign in voters’ minds likely stalled out with injury and the Dolphins offensive struggling the last couple games, but the numbers here don’t see the competition as very close.

The likely 2023 OPOY Christian McCaffrey also looks fantastic in this analysis, leading all running back in both rushing and receiving value added, and it’s not really close. George Kittle had a great comeback year for tight ends, and the biggest name had a down year (at least on the field). This is somewhat of a down year for offensive line value, but there are a few emerging, younger tackles already positioning themselves to be the most valuable for years to come.

Before getting into all the point-specific numbers across non-quarterback offensive players for this season, let’s big into the method behind the madness.

METHODOLOGY

I explain the methodology for valuing one facet, or phase, of player contributions in the NFL Plus/Minus introductory post: receiver route running. I don’t want to restate everything from that post, so please look back for clarification on the process. In this analysis, I’m applying that same methodology to the four aspects of non-quarterback value contribution for offensive players (rushing, route running, pass blocking and run blocking).

Before we get into the results of the analysis for every non-quarterback offensive position, let’s walk through a quick, specific example of how clustering works, using the route facet for wide receivers, highlighting some 2023 player results.

Without getting bogged down in the details, the graph above divides all wide receivers seasons from 2016 to 2023 with at least 50 routes into five different, color-coded clusters. For each receiver season, I calculated a number of efficiency and volume statistics and settled upon six for best differentiation receiver types: routes per game, slot rate, yards per route run, touchdown as a percentage of routes, first-down percentage, deep rate (percentage of 20+ yard targets) and yards per reception. I translated these seven features into principal components to minimize multicollinearity and make for easier visualization. The technique I used to form groups of similar receiver seasons is called k-means clustering. With this clustering technique, you choose the number of clusters, or groups, to form. 

The dashed arrows extending out from the center of the plot show the directional force of the different features and the length of the arrows indicates the importance of each feature on the cluster assignment. The volume of routes and route-based efficiency for yards, first downs and touchdowns all pull in the same direction to the upper-left. Touchdowns are the least impactful feature, which makes sense as they are the noisiest stat of the group. Higher slot rate is directionally indicated by player seasons at the top of the plot, with yards per reception and deep rate mostly being opposing features. It makes sense that slot receivers are less likely to go deep and have higher yards per reception, and vice versa for receivers who mostly line up out wide.

For the remainder of this analysis, I will refer to the five clusters by the last names of the representative players, rather than cluster number. With the clusters sorted, we then calculate plus/minus splits per play for each by adding together total EPA for all members when on the field, dividing that by total snaps on the field (running routes) and then subtracting the corresponding off-the-field figure from it.

Unsurprisingly, the cluster with Tyreek Hill and CeeDee Lamb (plus other top receivers) adds the most average value on a per-snap basis. This is followed by the Gabe Davis cluster, which includes efficient, outside receivers who typically run routes and are targeted further down the field. Primarily slot receivers like Wan’Dale Robinson add some value, but not much on average. Rarely targeted deep threats like Jalin Hyatt are a slight negative. The biggest negative cluster, by far, is the least productive and represented by Jonathan Mingo.

Remember that this is just a simplified example of one clustering sample. In the total analysis, I repeat this exercise a 100 times, with randomly assigned numbers of clusters, for each of the facets (route running, rushing, run blocking, pass blocking) for every position group. I then average up all the cluster value estimates or each player to get an overall number. There are also additional steps to add stability to the results, like application of the results through a regression, that bring us to the final numbers.

THE RESULTS

With the nerd stuff out of the way, let’s dive into the results for each position, before bringing them all together to see the most valuable non-quarterbacks on offense in 2023. First, let’s start with wide receivers.