Last week, I released the methodology and top-12 all-time list for PFF Plus/Minus, a new way to value wide receivers. PFF Plus/Minus is built on historical on/off the field splits for receivers, who are grouped by type in order to build larger, less-noisy samples of data that boost the usefulness of the information gathered. The full methodology is included in the link above, and worth your review to get the most out of this analysis.
We can apply this analysis to every receiver-season in the PFF era (2007-2019), ending with an estimate for the points gained, or lost, by a team for each route that a receiver is on the field. We’ve already applied the analysis to find the 20 most valuable wide receivers of 2019, and now we can use it to value this year’s rookies. In this analysis, we’ll go a step further and use the historical data to project rookies into the future by comparing them to historical rookie receiver-seasons and looking to see how those past rookies performed over the next few years.
The core of PFF Plus/Minus is building larger samples in clusters, or groups of similar wide receivers. These clusters are determined by minimizing the variance, or the differences within the number of clusters determined by the analyst. To illustrate the concept, I walked through an example of grouping the roughly 1,500 receiver seasons since 2007 with at least 200 routes run into 12 clusters.
Below I briefly walk through the 12-cluster example again, then highlight the 2019 rookie receivers in their respective clusters. Lastly, we review the rank order the rookies by Plus/Minus numbers and identify the most similar historical season for projecting them into the future.
I named each cluster from our sample after a recognizable name that has multiple seasons in the cluster, and highlighted them on the plot below.