clock menu more-arrow no yes

Filed under:

Alabama Football Recruiting 2016: SPARQ

New, comments

Doing nerdy stuff again. If you were around last year for this discussion, feel free to tune me out.

Reinhold Matay-USA TODAY Sports

It's that time of year again where all of the college football focus is on the decisions of a few 17-18 year old kids, twitter becomes the most reputable source in commitment scandals, and CB and I run our mouths way too much.

In the NFL draft, players are typically ranked by big boards and top-100 lists from "experts," and later by how high they are drafted. Similarly, high school players moving up into college are ranked by many different recruiting services, and are given "star" grades. The most elite players are 5 stars, and it goes down from there. However, the NFL also has a the Combine, where the athletes are put through different athletic tests to give teams pure numbers on how athletic the player is. Sometimes, players with weak college resumes can stand out to teams with a strong performance, or a highly rated player can really hurt his draft stock with a poor performance.

Recently, the recruiting scene has developed its own combine: Nike's The Opening. In 2003, Pete Carroll and some other of those west coast people began working with Nike to develop a new method to evaluate and rank high school athletes based solely on their athleticism. Eventually, they decided on four core drills that most accurately reflect a player's athleticism and explosiveness. The forty yard dash measures both acceleration and a player's top speed. The 20-yard shuttle measures a player's short-area quickness and acceleration. The vertical jump measures the player's explosiveness in his lower body (and obviously, how high he can jump). And the kneeling powerball toss measures his upper-body explosiveness.

These four measurements, along with the player's weight, were combined into a formula that isn't made known to the public (although it has been back-calculated through regression by some very ambitious people), and a single number for each player is spit out, known as a SPARQ score. The higher the score, the more athletic the player is, with heavier players getting higher scores than lighter players with the same testing measurements.

As an example, here are the test results of a player from Alabama's 2015 class, Deionte Thompson.

Last Name First Name Position State Stars National Rank Position Rank Height Weight 40-Yard Dash 20-Yard Shuttle Vertical Jump Power Throw SPARQ
Thompson Deionte S TX 4 44 2 6010 174 4.77 3.98 35.8 37.0


*Note: for the "height" number, the first digit refers to the feet, the next two digits are the number of inches, and the last digit is the 10th of an inch. Ex: 6035 would be 6 feet and 3 and 1/2 inches tall

As you can see, Deionte Thompson did not run a good 40-yard dash for a safety, nor is he very heavy at all. However, his shuttle time is well into the "elite" levels, and his powerball and vertical jump are also both well above average for a safety. So, despite the poor 40 time (what is usually the end-all be-all of measurements for the media), Thompson was given a very impressive SPARQ score of 107.

"What constitutes a good score," you ask? Well, that is where my boredom a couple of summers ago came in handy. I started to notice that even the most average of safeties or running backs would have higher SPARQ scores than the most athletic offensive linemen. So, I decided to find the average SPARQ score of a college-bound high schooler for each position. From there, using some fancy stats magic, I was able to give each player a Z-score, which is a measure of how much a player is better or worse than the average on a bell curve.

This allowed the SPARQ scores to be normalized for each position, and I feel that a Z-score represents how truly athletic a player is, even compared to other positions. Here is an example to demonstrate this:

Last Name First Name Position State Stars National Rank Position Rank Height Weight 40-Yard Dash 20-Yard Shuttle Vertical Jump Power Throw SPARQ Z-Score
Thompson Deionte S TX 4 44 2 6010 174 4.77 3.98 35.8 37.0 107.40 1.39
Cotton Lester OG AL 4 54 4 6040 325 5.34 5.00 21.5 44.0 81.30 1.46

You can see that Thompson tested better in every category than Lester Cotton except the powerball, and wound up with a much higher SPARQ score. However, that is expected for a safety. If your offensive linemen are faster than your safeties, then you've got a problem (or a track team for an offensive line... that could be good... maybe). So, when compared to how much that individual player stands out from the others at his position, you can see that Thompson and Cotton are actually very similarly athletic, and Cotton's 1.46 is also very slightly higher than Thompson's 1.39.

When using Z-scores, it's best to remember that they were calculated based on a bell curve. Therefore, a Z-score of 0 means that player is perfectly average (compared to all college athletes at his position), and is more athletic than exactly 50% of other athletes. Anywhere between -1 and 1 is typically considered just average.

At a Z-score of 1, that player is more athletic than about 84% of all college athletes at his position. From here, each increase in a Z-score is much more impressive than an increase while within the -1 to 1 range. By the time a Z-score reaches 2, that player is more athletic than around 97% of athletes. These are players such as Minkah Fitzpatrick, Tony Brown, Shaun Hamilton, Cam Robinson, and Derrick Henry.

Obtaining a z-score of three is near impossible. That would mean that the player is more athletic than upwards of 99% of athletes, and basically has to be one of those guys that just breaks the scale on every single test. The closest any Alabama player has gotten to that was O.J. Howard, who received a Z-score of 2.80. That's incredible.

From here on out, you'll hear me mentioning Z-scores and SPARQ when I'm talking about the new recruits committing (or hopefully committing) to the Tide. I will make sure to link back to this article anytime I mention them in the future. As always, if you have any questions about how I did the stats magic, or don't trust my methods, send me an e-mail (just click on my SBNation profile).