Adding a Proven Power Ranking Overlay to the Mix
Massey-Peabody NFL Power Rankings, devised by Cade Massey, Practice Professor in the Wharton School's Operations, Information and Decisions Department, and Rufus Peabody, renowned professional gambler, are based, simply, on rushing, passing, scoring, and play success statistics—adjusted for home field and discounted based on situation (i.e., score differential and time remaining). See here for a full description of the creators' methodology.
This framework delivered a solidly profitable 56% cover rate between 2011 and 2018, which translates to about an 8% return on investment, assuming the standard -110 vigorish. While we do not have accuracy data for the last five years, the cover rate during the eight-season sample period proved statistically significantly different than the hypothetical breakeven rate of 52.4% (per a single sample z-test). This observation conveys a high level of confidence in the predictive value of this system.
But winning over long periods is quite different than winning right now. To this point, through the first three weeks of the 2024-25 NFL regular season, Massey-Peabody rankings are an uninspiring 22-25-1 versus the number—well below their long-term success rate. Even still, these forecasts contain substantial information, in spots. As the graphic below indicates, when there has been at least two points of implied value with home teams this year, Massey-Peabody ratings are 13-5 (this informational content diminishes greatly for road picks, however, as is illustrated in the following exhibit).
Given the combination of our Week 2 struggles and the solid basis associated with Massey's and Peabody's work, we recently instituted the modification of cross-checking our historical context-based picks against Massey-Peabody NFL Power Rankings.
Happy betting!!
Comments