On the business of analytics

Last weekend marked the beginning of the 2026-2027 NFL season, and with it came a number of new and interesting narratives to watch out for over the next few weeks. Was Week 1’s offensive outburst a fluke, or are we in for a year of no defense, just vibes? Will the Kansas City Chiefs kickstart their dynasty again after a year out of the postseason? Did the Knicks magic somehow infect the Giants and Jets and restore respectability to the teams of New York City?

These questions are, of course, rather irrelevant here at Pinstripe Alley, which is, as you know, a baseball site focused on the New York Yankees. That being said, though, there is one topic of conversation that intersects rather neatly, and which bears some highlighting: the controversy surrounding Pro Football Focus (henceforth referred to as PFF), the most famous NFL analytics site.

For those who are unfamiliar, PFF is a sports analytics company that provides quantitative and qualitative analysis of NFL and NCAA Division I football. Known primarily for grading each player on a 0-100 scale, they provide data to NFL teams, players, coaches, and agents (presumably, for a fee). Their ratings are regularly cited by journalists and bloggers alike, and indeed, after Chris Collinsworth bought a majority share in 2014, have often been included on NBC’s Sunday Night Football broadcast, particularly during the player introductions towards the beginning of the game. When you talk analytics in football, chances are, you’re talking about PFF.

The introduction of analytics to baseball was the subject of much debate, and indeed, even today, there are quite a few holdouts who insist that analytics are overrated and believe that the truly successful teams ignore them. PFF ratings have arguably been more controversial, as at the end of the day, they lean on analyst’s interpretation of a play to create their analytical data, not on raw data taken from the box score or from the football equivalent of Statcast. Week 1, though, saw this controversy take on new heights: Trent Williams and Myles Garrett received almost identical grades despite the fact that Williams basically shut Garrett down all game, Jaxson Dart received an average grade despite a near-flawless game, and Caleb Williams’ performance differed so greatly from the eye test that the site had to publish an article explaining their reasoning.

Normally, this would be bad enough to spark extensive conversations among fans and members of the media, but after a few days, everyone would move on. Earlier this year, however, PFF was sold to Teamworks, and their new ownership began the era with mass layoffs that represented the culmination of a years-long brain drain discussed by both Defector and Check the Mic (the latter being a podcast with former PFF employees on as guests). Because of this, fans have begun to openly question the validity of PFF grades, with many fans on social media speculating that the company is using AI to help generate its grades; here, PFF’s refusal to typically reveal how they calculate their grades, plus the fact that most of PFF’s data is behind a paywall, adds fuel to the fire. If that’s the case, well, then a lot of bloggers and analysts who use PFF grades as a means to discuss the performance of players that are hard for the layman to analyze through film, such as offensive linemen or run-stuffing defensive tackles, will suddenly be without their main analytical tool.

The discourse surrounding PFF this week has gotten me thinking about the publication of analytics in baseball, and really, I can’t help but thinking how lucky we are. Of the three major sources of baseball data, one of them — Baseball Savant, aka, Statcast — is owned by Major League Baseball, and provides the same raw data given to the teams (well, most of it, anyway; we don’t exactly get to see how exactly the Outs Above Average sausage gets made).

The other two, Baseball-Reference and FanGraphs, are privately owned entities, but have cultivated strong relationships with fans. Baseball-Reference has made all their season-wide and game-by-game data, including season and career splits and historic box scores, easily accessible and freely available; only the advanced search options, most of which are not needed on a day-to-day basis, are behind the Stathead paywall. FanGraphs, meanwhile, has made some quality-of-life tools (including, unfortunately, a monthly article limit) exclusive for members, but again, the main draw of the site — in this case, the projection models and Roster Resource data — freely and publicly available.

As a graduate student and teacher, I don’t exactly have the disposable income to spend on Stathead and FanGraph memberships; trust me, if I did, I would, if only to remove the ads that sometimes cause the sites to freeze. Because of the sites’ commitments to easily accessible data, however, that has not hindered my ability to engage in analytics at all, both as a writer here at PSA and as a fan talking baseball with my brother and friends. Had Baseball-Reference and FanGraphs taken the PFF route and kept their data as private as possible, well, the world of baseball blogs, and honestly, even just the world of baseball, would be very, very, very different. And not in a good way.