Buffalo Bills analytics-based 2023 NFL Draft predictions

Alright draft prediction connoisseurs. We have something a bit different for your palate today. Below you’ll find my annual attempt to make predictions about the Buffalo Bills’ draft using a nearly pure analytics model. Check out this week’s Bills Mathia video for more information on the process, but the gist is simple.

NFL teams are constrained by a number of factors, most notably time. While the idea of teams creating “smokescreens” by meeting with players they don’t have any actual interest in makes a great story line, it doesn’t make a lot of sense as a common strategy. Every hour with a “fake” candidate could have been spent with a real one. A “fake” Top 30 visit uses the same slot a real one would have. I’m not saying that there isn’t an occasional smokescreen visit, but behavioral science would suggest it’s uncommon.

Note: Check out the Buffalo Rumblings visit tracker here for a great list of who they’ve met with and in what capacity.

---

By school

To be even more clear, a real interview might be for reasons other than interest in that specific player. There’s likely real fire to the smoke that scouts might want to find out more about someone by talking to a teammate or a former opponent. To that end, I tallied the number of players from each school with which Buffalo’s personnel interacted, with some thoughts below.

Unique Players

The numbers in this section are simply a count of how many players from each school the Bills met with. For now, if the team met with a player more than once, it still counts as one. I’ll only discuss schools with a value greater than “1” for this as the repetition would be what’s meaningful.

• The two schools that jump out are Florida and Tennessee, with four players and three players the team has interacted with respectively.
• Eight schools had two players Buffalo interacted with (Alabama, Georgia, Iowa, Michigan, Penn State, TCU, Texas, and USC).
• It’s hard to define trends here too clearly to be honest, so let’s move on to the next step.

Total Interactions

This is very similar to above but if they met with a player more than once it suggests there was follow-up the Bills wanted. Unlike above then, if the Bills interacted with a player two or three times, it’s counted that many times. That could suggest more buzz related to that school for one of the reasons noted above (player, teammate, or opponent of interest).

• There’s honestly not much movement here, and we’ll discuss specific players below. Florida and Tennessee go up by one for instance and a few more schools move to two. Essentially, any movement here suggests Buffalo met with a player more than once.
• I’ll discuss two potential targets of the Bills below. For the time being I’ll indicate that I did look to see if the schools they played for and against had any major numbers in this section or the one above. I couldn’t find anything that looked significant, which might suggest Buffalo doesn’t spend a lot of time talking to players about their teammates or opponents.

By Position

This is mostly straightforward. A general manager and scouting department simply cannot afford to use a ton of time and energy on positions their team doesn’t need. For example (and spoiler alert), the Buffalo Bills have invested essentially no resources on quarterback this year. That’s drastically different than what we saw when I did this exercise in 2018. Hmm... I wonder what changed?

There may still be some wrinkles, like interviewing opponents of a position of interest, etc. But, overall, the expectation is that teams spend the most time on positions they need most.

Unique Players

This is the same idea as above and just a raw count of how many unique players by position team personnel interacted with. For the sake of clarity, because the Bills tend to use their linebackers in a hybrid fashion, I counted all linebackers in one category (there also weren’t too many of them). Unlike linebackers, I have offensive linemen very specifically sorted, as there’s often less messing around by the team for these positions (though still a good deal of messing around).

I like charts because they really cut down on what I need to type. Wide receiver is the obvious area of focus. Safeties come in for a distant second. It appears that corner, linebacker, defensive end, tight end, and running back have all had reasonable focus as well. Based on current team needs, I don’t see many surprises.

Offensive line is fragmented as noted above. Combining the group leads to eight interactions so there’s some likelihood they’d like to add someone there, but may not be looking at a particular position necessarily.

If we’re combining positions though, that makes defensive backs intriguing as the combination of safety and corner is hard to ignore.

Total Interactions

Also like above, now we’re counting all instances if they interacted with a player more than once.

No spoilers on who was a repeat interaction until below when I reveal my analytics-based pick. The biggest jump, however, is wide receiver. That means not only did they interact with the most unique players in this position group, they also had more re-interviews with this group than any other. Feel free to dissect more, but this is pretty clear which direction my data pick needs to go.

The Pick Is In!

Surprise! With 16 total interactions and 11 unique players on the list, analytics strongly suggests that the Buffalo Bills are spending a ton of time on wide receivers. The level of resources in this direction is undeniable and it’s my analytics-based pick for their Round 1 target. But which receiver?

The Bills have met with Jaxon Smith-Njigba and Jalin Hyatt, who I’ve seen some fans crossing their fingers in hopes the team lands. Neither are the analytics pick though, having only met with the Bills once (more on that in a second). Jordan Addison, Zay Flowers, and Tank Dell have all met with Buffalo more than once, with Addison topping the list at three.

Locking it in, my official pick for the Bills’ target this year is wide receiver Jordan Addison. Don’t get me wrong. If Smith-Njigba or Hyatt fall, they could easily become the actual pick. But even the analytics pick can’t be completely pure.

Addison and Flowers are currently projected as late-round choices whereas Smith-Njigba and Hyatt are gone earlier in many projections. In a similar but contrasting vein, Justin Marshall met with Buffalo twice. As a local prospect the extra meeting makes sense, but doesn’t move the needle for me based on reality. He’d be a real surprise pick in Round 1. What I’m getting at is the analytics-based pick needs to factor in what’s reasonable to some extent. Smith-Njigba and Hyatt don’t seem as reasonable as Addison or Flowers at 27.

Notably; Smith-Njigba, Hyatt, Addison, Flowers, and Dell all were private meetings — which my premise would suggest are the most important (most limited “resource,” best availability to the team). I could also add that top talent doesn’t require as much digging, which means Buffalo may be as comfortable with meeting Smith-Njigba once as they are meeting Addison three times. Ultimately, the draft can be chaotic and teams need Plans A-Z not knowing what everyone else will do. Why add all this extra? Here’s my “official pick but with qualifiers.”

• Plan A: Jordan Addison — Buffalo expects him to be available and have committed a ton of resources to the position and Addison himself. Perfect marriage of draft prognostications and analytics.
• Plan B: Zay Flowers — Like Addison, a good blend of resources invested and draft fit. Just a little less of both for Flowers who Buffalo met with one less time and many draftniks have behind Addison.
• Plan C: If Jaxon Smith-Njigba or Jalin Hyatt fall the Bills jump on it, maybe even moving up a bit for one of them. This plan relies more on hopes and dreams than an actual plan should, hence it being third on the list.

Miscellaneous Predictions

Real quick hitters as the article has already broken 1,300 words:

• Despite having Tre’Davious White, Micah Hyde, Jordan Poyer, Taron Johnson, Kaiir Elam, and Christian Benford on the team, I think Buffalo is a tad worried about the future of the defensive backs — and they add another.
• I’m a bit shocked at how few interactions they had with linebackers. It’s possible this means the Bills really aren’t looking to replace Tremaine Edmunds and realize they may need to shift the defense a bit.
• There’s likely some substance to the running back theory, but intriguingly they met with more tight ends. Offensive weapons seem to be highlighted, and I think Buffalo may use two of their more premium resources here.