NFL Draft: Using Metrics to Assist the NFL Draft Process
If you follow me on Twitter (which you’re always encouraged to do @DraftChiado) you’re probably aware that I have taken the time since last summer to develop an athleticism metric that quantifies test scores into a single factor for NFL Draft prospects. Before diving into the post, I’ll leave you with two definitions:
Analytics– The discovery and communication of meaningful patterns in data
Metrics– In mathematics, a function that defines a distance between each pair of elements in a set
These two definitions perfectly describe how analysts are using metrics to assist with the NFL Drat process. In a single draft class, metrics are meant to do exactly what the definition suggests; analyze the distance between individual players in terms of athleticism.
In the larger sense of the process, in terms of analyzing multiple classes, the analytics definition fits more properly. The members of so called #MetricsTwitter (draft analysts on Twitter who incorporate metrics into their evaluations) are trying to use these data-sets and test scores to determine if there is, in fact, a direct correlation between athleticism and success, and if so how?
Now that you understand the basis behind the idea, let’s tackle the motivation and specifics behind my metric.
The Motivation
Anyone who keeps in touch with the football and draft community on twitter or around the internet is certainly familiar with the two people behind two of the more well known metrics.
Justis Mosqueda (@JuMosq on Twitter), who writes for Bleacher Report, Optimum Scouting, and Draft Breakdown, among others, has worked on his metric specific to front-seven players. His metric is given the designation #ForcePlayers#.
What Justis has figured out through his fantastic work is that there is a correlation between the success of front-seven
More from Still Curtain
- Steelers news: Rashard Mendenhall calls out Big Ben, Brian Flores interviews for HC job, and more
- Steelers play it safe in recent 2023 NFL mock draft
- Kenny Pickett gets his guy in latest Steelers mock draft with trades
- Steelers face potentially dark reality if Matt Canada doesn’t pan out
- Senior Bowl sleepers the Steelers must have on their radar
players in the NFL and their athletic testing numbers coming out of college. Now that he has established a database and basis or his metric, Justis gives the front-seven players in each draft class the designation of #FP# or Non#FP#. The somewhat obvious final product is that players who achieve the #FP# are more likely to succeed than players who don’t
The second analyst who is behind one of the more well known metrics is Zach Whitman (@zjwhitman on Twitter), who writes for Field Gulls and 3 Sigma Athlete. Zach uses a reverse-engineered version of Nike’s athleticism metric for college football recruiting called SPARQ.
The back-story behind how Zach and other members of Field Gulls (Danny Kelly and Davis Hsu) began their work on the variation of the SPARQ metric, which is called pSPARQ is too long to get into. The basis, though, is that they hypothesized that the Seahawks used some sort of variation of the SPARQ metric to influence their drafting, so they reverse engineered the concept to come up with a reasonable variation.
Although Justis’ metric is focused on a single position, while Zach’s focuses on the entire prospect pool, they both are trying to predict who the most athletic players, and therefore the players most likely to be succesful, in a given class are.
What can be seen through the work done by Justis, Zach, and others is that there is a correlation of some kind, whether it is big or small, between athleticism and success in the NFL. The correlation is not necessarily the more athletic, the more likely to be successful in the NFL, but is probably related to a baseline athleticism that a player must meet.
What this means is that as long as players reach certain baseline speed, agility, explosion, and athleticism scores then other factors of their skill-set determine their success level. Meanwhile, players who don’t meet the baseline scores will need to have exceptional traits in other areas in order to succeed.
Seeing the work done by Justis, Zach, and many others prompted me, being the math nerd that I am, to see if I could comeup with my own reasonable metric.
The Specifics
Most athleticism metrics that analysts are using today test three specific areas of a player’s test scores: speed, agility, and explosion. Most individual metrics also combine these three factors into a single athleticism number. Now there are certainly different ways that analysts are approaching metrics, but most that I’ve seen separate scores into these three categories.
The metric that I have created takes speed, agility, and explosion into account and forms a single athleticism factor that I call qABP. Since it isn’t obvious, qABP stands for Quotient of Athletic Based Production. The idea behind this metric is to show how successful a player would be if NFL production were based solely on athleticism. Now, obviously, that isn’t the case, so what qABP can be used for is to show who the most athletic players in a given draft class are.
When looking at combine test numbers and watching games it is somewhat obvious who are the fastest and most athletic players. When sorting combine data, skill position players and defensive backs rise to the top with the fastest times. That is why when creating a metric it is necessary to adjust for weight in some way.
Feb 23, 2015; Indianapolis, IN, USA; Washington Huskies defensive back Marcus Peters does the vertical jump during the 2015 NFL Combine at Lucas Oil Stadium. Mandatory Credit: Brian Spurlock-USA TODAY Sports
In my metric I adjusted for body density, an idea which I got from Justis, because a 6’1″ and 300 pound player is different from a 6’6″ and 300 pound player. It’s necessary to adjust for density in the metric scores in order to see who is the most athletic for their body size.
In a similar idea to adjusting for body density, I also incorporate z-scores of each category for each individual position. What the z-scores are basically showing is how each players score in the individual categories. In mathematical terms a z-score measures how many standard deviations away from the mean a given data point is.
For those of you reading that aren’t as fluent in math talk, the z-score measures at what percentile a certain data point falls in a set. Through calculating the z-score and looking at a table of normal distribution, which converts z-scores into percentiles, you can tell where a player falls in the given data set of the qABP at a certain position.
What you can do with the qABP score and a score is tell a player’s athleticism with respect to weight and the position they play. For reference, here is the scale I use:
Elite: 83.50 and above
Above Average: 75.50-83.49
Average: 64.00-75.49
Below Average: 53.50-63.99
Bad: 53.50 and below
For another reference point, anything above 100 is ridiculously good and anything below 25 is ridiculously bad.
Why Metrics Are Important
Just like any theory on how to approach the draft process, there is no one answer that is fully correct. What I, and all of the others who use metrics in their evaluation process, are trying to do is use all of the information at our disposal to make the most informed decision.
Referring back to the definition of analytics above, if there is a pattern that exists in athletic measurements correlating to success in the NFL, it would be counter intuitive to overlook it. If there really is a pattern in the data, like all data gathering so far suggests there is, then why wouldn’t you want to incorporate it into your evaluations?
With as large of a player pool as there is in each draft class and how much of a risk even the safest of draft picks is, it would help to have all of the information possible to make an informed decision.
The NFL Draft is and always be an imperfect process that we are trying to perfect as much as possible. Even with the progress that’s been made with analytics, there still isn’t a definitive answer on how to approach the process.
My goal, and what the goal of every draft evaluator should be, is to find the best way to approach the process and right now that is with incorporating analytics into evaluations.