NFL Rookie WR Model - Using an Advanced Model for Rookie Wide Receivers

Our goal is to identify the top rookie prospects based on data points that correlate most with future NFL production.

If you want a complete breakdown of how the model works, check out the Super Model Inputs & Methodology below the table.

Last Updated Aug 19th, 2024 8:10 EDT

With the NFL Draft quickly approaching, we are releasing our updated Fantasy Life Rookie Super Model!

Our goal is to identify the top rookie prospects based on data points that correlate most with future NFL production. I have been working on NFL rookie models for the last three years, and over that time, I have studied and measured hundreds of predraft variables against future NFL production.

The truth is that most variables don’t carry a strong signal, or they overlap too much with an existing variable to make it into a model. Even once you define a list of relatively strong inputs, it is hard to accurately predict which college athletes will be the best NFL players.

Football is a sport with countless dependencies played by notoriously unpredictable creatures known as human beings. When you add in plain old variance, you can see how this activity can become challenging. But that is what makes it so interesting, and it fuels me to test new ideas every offseason.

So, without further ado. Let’s dive into the inputs used for the 2024 WR Super Model.

WR Super Model Inputs & Methodology

The inputs are in order of their correlation to fantasy production in a WR’s first two years in the NFL.

  • Projected draft capital (NFL Mock Draft Database)
  • Collegiate program quality
  • Adjusted career receiving yards per team pass attempt
  • Career targeted QB rating
  • Career TDs per game
  • Age

Because the model includes advanced data that isn’t widely available before the 2018 class, our sample focuses on WRs with at least two years of play since then. So, our correlations to future performance currently derive from WR data from 2018 to 2022.

For all production stats, the data comes from the game log level rather than the season.

Draft Capital Value

The model uses Chase Stuart’s Draft Value Chart for draft capital, which is essentially a better version of what many know as the Jimmy Johnson trade chart. The value of a draft pick isn’t linear, and this methodology helps us capture that. The dropoff in value is steeper in the first round and becomes much flatter around the end of the second round. Draft capital value is the most weighted input in the Super Model.

Before the NFL Draft, we used expected draft capital based on mock drafts. After the draft, we update with actual capital. Pick 275 means a player was an undrafted free agent.

We index this data to give a score between 0 and 1, with one being the best.

Program Quality Index

Program quality uses the draft capital value to determine the total value each collegiate program has contributed to the NFL Draft at the WR position since 2014. Those scores are then indexed to form the Program Quality Index.

WRs who come from stronger programs score better. Program quality has been a factor in the model before, but this is a better way of quantifying it. Additionally, this metric helps offset lower production numbers from WRs with more target competition.

Adjusted Career Receiving Yards Per Team Pass Attempt

RYPTPA helps us normalize receiving yards based on the team environment, which is very important since how much a team throws can vary drastically.

The adjusted version of Career RYPTPA accounts for four critical variables that showed to impact performance:

  • Age and class (i.e., first-year, second-year student, etc.)
  • Average depth of target (aDOT) and alignment
  • Team passer rating
  • Teammate score (competition for targets)

Receivers who performed well in RYPTPA earlier in their careers enjoyed much stronger hit rates in their first two years. In fact, WRs who didn’t perform well until Year 4 and Year 5 correlated negatively with NFL success. To account for this, the model assigns heavier weights to the first three years.

This measurement also allows us to move away from breakout thresholds, which have a nasty habit of barely missing prospects, barely too low or high–everything is now on a scale.

The other three variables quantify an expected RYPTPA from game-level data since 2014. Then, we can perform an over-expected calculation. 

The higher a WR’s aDOT, the higher their expected RYPTPA. The higher a team’s passer rating, the higher a WR’s expected RYPTPA. The stronger the teammate competition, the lower the WR’s expected RYPTPA.

These four factors are then weighted and combined into one data point and indexed (placed on a scale from zero to one).

Career Targeted QB Rating

This metric tells us the passer rating when a WR was targeted. There is an inherent overlap between targeted QB rating and RYPTPA data points because both use yards. 

However, RYPTPA tells us how a player performed in the context of their team, while targeted QB rating tells us how well a WR performed when targeted. That critical distinction allows these two metrics to work well together. 

Career Total TDs Per Game

The data showed that using a normalized metric like RYPTPA was superior for receiving yards, but that wasn’t true for TDs. Instead, per-game data demonstrated a stronger correlation than share, per-team attempt, and other options.

There is a correlation between yards and TDs, so once again, there is some overlap in signal between our metrics. However, not all WRs who are strong in RYPTPA score a lot of TDs.

Intuitively, this makes sense because we expect WRs who can score long TDs and provide value inside the ten-yard line to have an advantage over a small slot WR with a ton of targets.

Additionally, we account for the broadness of a WR’s utility by including rushing TDs in the career total.

Age

A player’s age derives from how old they will be at the beginning of the upcoming NFL season. It doesn’t carry as much weight in the WR model as it used to because we already account for age in adjusted career RYPTPA.

You will notice that dominator rating, which combines a player’s percentage of yards and TDs, is no longer in the model. Career total TDs and RYPTPA offered stronger correlations to future production, and dominator rating was duplicative, so it didn’t make sense to keep it in the model moving forward.

You might also be wondering why target share wasn’t included–especially considering how important it is at the NFL level. The answer is twofold: 1) RYPTPA is stronger, and the two correlate strongly. 2) The targeted QB rating was stronger and offered a holistic view of efficiency that we can’t get from target share, which made it a better pairing with RYPTPA.

`