The NFL playoffs kick off this weekend, which means it’s time for an annual tradition: Feverishly debating who will win the Super Bowl.
Everyone from Boston to Santa Clara, Calif., has their opinion on NFL playoff predictions, including us. But in a recent NFL.com column, analytics specialist Nasir Bhanpuri decided to leave emotions out of it and predict the participants of this season’s big game using cold, hard statistics.
What he concluded was rather interesting: According to his model, the Arizona Cardinals and Cincinnati Bengals will meet in Super Bowl 50.
Without getting too bogged down in numbers, Bhanpuri’s model produces a “rating” for each team based on three main factors that measure offensive and defensive efficiency: expected points contributed by offense, which accounts for the “importance” of each yard gained; simple rating system (SRS), which accounts for a team’s margin of victory and strength of schedule; and offense simple rating system (OSRS), which is a variation of the SRS.
Here’s what the model produced:
So, yeah. Those stats aren’t too kind to the Denver Broncos and New England Patriots, who are the No. 1 and No. 2 seeds in the AFC, respectively.
Bhanpuri admits there are factors his model doesn’t account for — including, most notably, injuries and personnel changes. The Patriots are a much different team with Julian Edelman at wide receiver, as are the Pittsburgh Steelers without DeAngelo Williams.
You don’t have to give this model any weight if you don’t want to, but it’s worth noting Bhanpuri correctly predicted the last two Super Bowl opponents.
So can we count on Carson Palmer squaring off against Andy Dalton in California? It would take a lot, as the Bengals need to get past a tough Steelers team that beat them in Cincy less than a month ago. But hey, it’s fun to predict.
Thumbnail photo via Aaron Doster/USA TODAY Sports Images
