PFF is criticised as much as it is loved around the NFL so we thought we’d ask the question, just how reliable is the PFF grading system?

PFF has undoubtedly brought analytics to the mainstream. With its sleek graphics and well-known personalities endorsing the service, such as Cris Collinsworth they gain a lot of attention. Their grading system has been brought under a lot of scrutiny for various reasons which I’ll get into later. These gradings are now making it onto to the TV broadcasts and NFL teams are using them.

Why do people love PFF?

I think the main reason that people love PFF is that it gives them a definitive number to rank players by and they don’t have to do the analysis. Whether the numbers transcend positions or not, PFF does rank the 101 best players each year regardless of position by their grading system.

The other main thing that PFF has recently started is providing help with fantasy football. It’s about £40 for a year’s access and they give you the best matchups and most likely players to breakout based upon their advanced statistics. Whether they’re any more valuable than any other fantasy analyst is anyone’s guess because fantasy football really is a crapshoot.

Criticisms of PFF

The rating system at PFF can sometimes seem dodgy because you might go to a certain position and right at the top of the rankings is some player you’ve never heard of on the Bengals. Ratings for positions such as corners and offensive linemen are especially scrutinised because PFF doesn’t know assignments so how can they accurately grade whether a play was good or bad. This was explained perfectly by Belichick in an interview, don’t rely on the analysis without seeing the playbook! There are players you see ranked highly that look like trash which makes PFF seem like they know nothing.

Their mathematical rating system isn’t exactly foolproof. They try and assign values on a pseudo normalized distribution -2 to +2 range. This implies that two -1 graded plays and a +2 graded play are equivalent to three 0 graded plays where in reality the first set of plays is more valuable. Here’s an example: If a corner misses two tackles against a runner but then jumps a curl route and takes the ball to the house that’s better than being on the opposite side of the field for three plays in reality but not in their grades.

They also miss a lot of context, they grade things based on just what happens not comparing against what would’ve or could’ve happened. A player can do the exact same thing but because of an external factor they get a different grade

So is PFF reliable?

I wouldn’t say that PFF is totally reliable or unreliable. I think the service has its good and bad points like everything else. Whether we like it or not, analytics will continue to shape the NFL going forward. At this point, it seems like PFF will be the leaders in that market. Game tape should still be the ultimate test of whether a player is good or not. However, for data capture and other mechanisms, PFF is a great measure and an invaluable tool for analysts, fans and NFL teams alike. I think with things like this there are always going to be biases baked in but you can decide how much stock you want to put into PFF rankings.

Keep up with all the NFL offseason news with Gridiron Pulse!