Don't worry, the Jets still have an analytics department ... for now

This news from last night may have had a few people scratching their heads.

When the Jets first hired Joe Douglas, there was some talk that the Eagles were moving to a more analytical approach and that was part of the reason Douglas wanted out. If those concerns lingered, then the above news could signify the Jets are shifting away from an analytical approach. This would be shades of the Giants, who have been widely derided for their closed-minded and old school approach under Dave Gettlemen.

However, let's not forget that Douglas already set aside these concerns at his introductory presser, where it was noted that the Eagles had been embracing analytical processes throughout his tenure, as had the Ravens while he was with that organization. Douglas called it a great tool that he does value for how it helps to paint the whole picture objectively.

Mulholland actually left the organization in December and his role within the football administration department has not been replaced. However, Brian Shields, a senior manager who worked above him in the hierarchy, is still with the organization. His official title is "Senior Manager, Football Scouting Research Analytics".

Officially, the Jets didn't have an analytics department until 2018, at which point it was reported that the organization was embracing the data game and made it sound like Mike Maccagnan was warming to the idea having originally introduced a more conventional approach.

Mulholland had been dabbling in draft prospect analysis for some time though. Here's a paper the former Wharton grad co-published back in 2016, while he was working as an analyst in the Jets' organization, albeit in the finance department. The article states that he's been conducting statistical football research since he was still in college.

While there's no suggestion that the Jets took any of Mulholland's analytical modelling into account when making any of their draft picks in 2016 and 2017, it seems likely some of his analysis made its way up the organizational chain and led to him getting a role within the newly-established department.

The modelling itself is impressively complex, especially to someone who might not be from an analytical background. However, the results of this particular study didn't bode too well for his future success.

The model created identifies several wide receivers drafted after the first round who would have been expected to go on and have great careers. Four years on, though, this is far from an impressive list of recommendations:

These include Devin Smith (to the Jets in round 2), Dorial Green-Beckham (to the Titans in round 2), Tyler Lockett (to the Seahawks in round 3), Chris Conley (to the Chiefs in round 3), Sammie Coates (to the Steelers in round 3), DeAndre Smelter (to the 49ers in round 4), Rashad Greene (to the Jaguars in round 5), and Titus Davis (to the Chargers as an undrafted free agent).

The inclusion of Smith sticks out like a sore thumb there, and one wonders whether the former Ohio State product was selected by the Jets at least in part based on the recommendation from one of Mulholland's models.

The fact that, apart from Lockett, most of these players have had poor starts to their career doesn't necessarily mean it's a bad model. Most of these, including Smith, had reasons that went beyond their potential and skill-set for not having been successful; primarily injuries.

All of which just goes to underline that the analytics process is just part of the puzzle and nobody should be a slave to it without also doing extensive conventional scouting and due diligence on the player's character and medicals.

Mulholland is now gone from the Jets organization and we have no way of knowing whether he was forced out or left of his own accord. Nor can we determine whether any of his work is to blame for some of the Jets' bad moves since 2018 - or even before that if he had the ear of any front office decision-makers as we suspect.

Are the Jets losing faith in analytics? Probably not. However, Douglas came from an organization that used them successfully, so perhaps he's streamlining the process or looking to make improvements to the kind of data he gets from this department.

With a longer than usual lead-time before his first draft in the general manager role, Douglas will hopefully have his processes and procedures in place with a big influx of new data points about to drop over the next week or so at the combine.

Whether this approach is going to be successful is anyone's guess. Someone should put together a statistical analysis model to see whether or not they're on the right track.