One particular area of interest in the analytic realm is the use of pre-draft measurements and tests to forecast player behavior and future potential. My work in this area is based on SPARQ, a metric developed by Nike to apply a single composite score to a player’s athleticism. Most notably, the Seattle Seahawks are reported to use SPARQ-like metrics in their player evaluations, with Seattle Head Strength and Conditioning Coach Chris Carlisle involved in the creation of SPARQ. The idea is simple: quickly assess the athleticism of a player with a single number.
What’s the use of SPARQ? What we see often in pre-draft analysis is an over-emphasis on the forty-yard dash, for which there are two main reasons: (1) speed is important, and (2) we’re familiar with the common forty benchmarks. A 4.4s 40 is fast and sounds good, and there’s an inherent understanding of what it means. The problem is that the forty-yard time isn’t fully indicative of a player’s overall athleticism. Most people don’t know off-hand what a good broad jump is for a wide receiver, and even fewer are aware of what they should expect from a defensive end. SPARQ is a way to standardize these different parameters and gain a more circumspect view of a player’s natural ability.
I’ve written extensively on SPARQ in the past, and will link to those articles rather than re-hash the basics of the model here. The following point is key to understand, though:
SPARQ isn’t perfect. Player test results have error and, even if they were perfect, don’t fully represent the ability of an athlete. The goal here isn’t to build an airplane. SPARQ is just a method by which we can better understand players, and it’s important to not let perfect be the enemy of good.
Terms to know:
pSPARQ – The SPARQ formula isn’t published, but there’s a significant data set available from SPARQ high school camps. From this data, I back-calculated a formula that’s very close to SPARQ. I then worked to keep the SPARQ weighting the same while applying the tests that we use for NFL athletes. The eight inputs for my metric, pSPARQ, are as follows: player weight, bench press, broad jump, vertical jump, forty-yard dash, ten-yard split, short shuttle and 3-cone (L) drill.
simScore – This is a measure of how a given player compares to another using a SPARQ-based similarity algorithm. As far as I know, the idea of similarity scores is originally credited to Bill James, a pioneer in baseball sabermetrics. With my metric, a simScore of 60 shows a vague similarity, 80 represents a significant match, and 90 is a near-exact result. The excellent Tony Wiltshire has provided me with player data from the last 16 drafts to use in the simScore study.
Similarity scores aren’t perfect, but they do provide an interesting perspective and allow us to view a player through an objective lens. Had we used this method in 2013, the main breakout candidate would’ve been Zac Stacy, mainly due to his high Doug Martin simScore.
Editor's Note: For rankings, projections, exclusive columns, mock drafts and tons more, check out our jam-packed online Draft Guide or Draft Guide iOS app, or follow Rotoworld Football on Twitter for the latest news.
Jerick McKinnon, RB, Minnesota, 147.5 pSPARQ
No other running back has an athletic profile similar (i.e., over 80) to that of Jerick McKinnon. Backs with McKinnon’s speed, lower-body explosiveness and upper-body strength just don’t exist. He’s truly unique.
He’s also a freak. An average NFL skill position player will ring in at around a 110 pSPARQ, a good athlete at 120, and very good at 130. Elite is about 140, and the 150+ range is reserved for Calvin Johnson, Vernon Davis, and a select few others. McKinnon’s 147.5 is one of the top scores among all running backs over the last 16 years.
An option quarterback at Georgia Southern, there’s certainly a fair amount for McKinnon to learn at the NFL level to become a running back. The transition isn’t guaranteed. But out of 1000+ RBs, my database shows that only 56 have recorded a sub-4.4 40-yard-dash at either the combine or their pro day. The top 6 pSPARQs from those 56 are: Willis McGahee (tests taken pre-ACL), Chris Johnson, Edgerrin James, Jamal Lewis, Reggie Bush and Jerick McKinnon. When you’re the kind of freak McKinnon is, you tend to do pretty well in the NFL.
Toby Gerhart, RB, Jacksonville, 130.6 pSPARQ
This is one of the more interesting SPARQ results. Gerhart’s 131 pSPARQ puts him at the level of a very good NFL athlete, which does not appear to be an opinion shared by the general public. Consider the following two athletic profiles.
Now, test results aren’t a perfect indicator of athleticism, and Gerhart is likely not quite the athlete he was in 2010. Still, his measurables compare very favorably to those of Steven Jackson, and we don’t question Jackson’s athleticism. Gerhart is a well above-average NFL athlete and his profile suggests that his ceiling extends beyond second-stringer.
Andre Williams, RB, New York Giants, 137.7 pSPARQ
Williams doesn’t flash with a 4.5 40-time, but his overall athletic profile is very impressive. He recorded elite results in the short shuttle, vertical and broad jump categories, more than making up for the average 40. Out of 1000+ running backs dating back to 1999, the two players with significant similarity to Williams both ended up with good NFL careers. The Giants would be pleased if their fourth-round pick ended up with the career of either Correll Buckhalter or Marion Barber.
Latavius Murray, RB, Oakland, 132.9 pSPARQ
I hesitate a bit to make the Peterson comparison. Obviously, All Day is one of the great athletes at the position, and the test results don’t fully reflect what he does on the field. With that caveat, Murray’s top two simScores are very promising.
Good fantasy strategy can improve your odds, but no one will ever be perfect at predicting the future in football or any other endeavor. The idea is that high-SPARQ players generally tend to succeed, and low-SPARQ players generally struggle. Murray is far from a lock to succeed, but he has the requisite athleticism to do so, and is as such worth keeping an eye on.
Jordan Matthews, WR, Philadelphia, 119.4 pSPARQ
It was weird to hear pre-draft criticism of Matthews based around a lack of NFL athleticism. He’s not Calvin Johnson or Julio Jones, but his numbers are comfortably above-average and draw comparisons to two very successful receivers. Again, the simScore comps don’t mean that he’ll end up being as good as Hakeem Nicks or A.J. Green, but they do show that it probably isn’t a lack of athleticism that would hold him back from being a success in Philadelphia.
This particular player profile just tends to be successful. In addition to Nicks and Green, other high Matthews simScores include Larry Fitzgerald (83.5) and Alshon Jeffery (81.7).