Editor’s Note: This article details some research recently put in by our own Eric Hardter to discover a better way to view receiving output. For a complete breakdown of his findings, make sure you visit our Premium Content section.
As many of the DLF faithful may already know, I spend my days in the laboratory honing my chemistry technique and striving to achieve new scientific goals. When it comes to the way I approach dynasty research, it’s truthfully no different. Much like how I approach my “day job,” I’m constantly attempting to refine my data sets so I can better understand the statistical landscape that is fantasy football.
No, these metrics don’t always hit, nor do they begin to tell the full story of a player’s ability (even as a stat guy, I readily admit it’s folly to ignore game film). However, much as with anything in life, you should constantly be trying to improve your abilities in whatever way possible. If coercing every last drop of numerical goodness out of a player’s raw data can help me identify trends and potential breakouts, you’ll find me with a calculator in hand.
To that end, previously I had written about and even introduced a small range of metrics that relate to wide receivers and tight ends. I consider each to be a valuable data point, but they are certainly not without flaws. Here are what I perceive to be the drawbacks of each:
Points Per Target (PPT): This is a useful metric that details a pass catcher’s efficiency on a per attempt basis, but lacks the ability to describe the offensive constraints forced upon him. Simply put, was his quarterback any good?
Yards Per Target (YPT): This metric combines and summarizes a player’s yards-per-catch and catch percentage, but fails to describe his scoring ability.
% Change to a Quarterback’s PPA (%CPPA): This metric answers whether a quarterback is more or less efficient when targeting a single player. While %CPPA is perhaps the best metric on a relative level, it doesn’t take into account the very important component of sample size.
When combined the above metrics tell a nice story, but I think we can do better. It’s in that spirit that I designed AIR, or the Adjusted Improvement Ratio. AIR masks the individual deficiencies of the qualifiers above by taking nearly every pertinent factor into account, including sample size, efficiency and the scope of a team’s offense.
To tabulate AIR, I created and followed the following formula:
AIR = % of Team’s PPR Receiving Points / % of Team’s Targets
The calculated ratio can then determine if a player was able to operate above the mean level of production for his offense. It also provided an answer as to whether a player’s fantasy points correlated to the volume of targets he received. Put another way, AIR describes a pass catcher’s impact relative to his team, as well as the efficiency at which he operated.
I know I’ve said a mouthful and there’s only so much that can be described about metrics before the actual numbers are used. So in order to provide a sample of AIR in action, let’s consider the Browns’ Josh Gordon and his sterling 2013 campaign. In order to enact AIR, we first need to know how many PPR points were available to the Cleveland pass catchers:
As the table above shows, this number can be generated by taking the quarterback data and converting it to how the points would be scored relative to pass catchers. Each reception is worth one point, each yard is worth 0.1 points, and of course every touchdown is worth six points. For the Browns, this amounted to 972.2 possible points. As Gordon was responsible for 314.4 of these points, his numerator for the AIR equation stands at 32.3%.
The denominator is more straightforward. Browns’ quarterbacks attempted 681 (!) total passes, 159 of which were directed at Gordon. Therefore his AIR denominator (% of team targets) stands at 23.3%. Dividing the two numbers results in an AIR of 1.39 for the budding superstar and also helps qualify the metric as a whole – the bigger the AIR, the better. AIR ratings approaching 1.0 (or below) represent expected, or subpar production.
In a test analysis, AIR data was derived for the top-12 PPR receivers, and is summarized in our Premium Content today if you’re interested. Also, each player’s PPR rank was included in order to highlight any discrepancies. Finally, brief analyses for each player are included, highlighting what their AIR signifies.
Conclusions and Future Directions
In the future, I plan to sift though the historical data to see if the AIR metric can surpass descriptive abilities and serve as a predictive measure as well. I’ll also expand the sample size in an effort to detail even more players, and see if AIR can help elucidate the futures of the 2013 rookies. Finally, I’ll apply AIR values to other positions, including tight end and running back.
In summation, the AIR metric was created to help further describe the play of pass catchers. I think it is an excellent indicator of how efficient a pass catcher plays, surpassing the limited qualifiers used previously. Not only do AIR values show how well each player performs individually, they also effectively normalize results to incorporate the scope of the entire passing offense. This isn’t Lord of the Rings and there isn’t “one metric to rule them all,” but I truly believe AIR data offers a better lens through which to view receiving output.
Follow me on Twitter @EDH_27
You can find his (typically strong and hopefully reasonable) opinions on Twitter at@EDH_27.