Alabama Football Recruiting 2017: SPARQ Primer

When spending your time lurking the interwebs and getting in arguments with anonymous strangers over trivial things such as sports, politics, and existential ponderings, you’ll inevitably have an opinion such as “Wow! Alabama’s football team is so much more athletic than Auburn!”

To which BarnerBoi991 will reply, “u r biased. Alabama sux and is cheaterz. Auburn is still a yung team. W8 till next year.”

And then what? You can try to argue, but what proof do you really have that can’t be refuted by such a profound argument as the one above?

Well, that’s where SPARQ comes in.*

*and for those of you who aren’t masochists who browse the comment section on aye-ell dot com, it’s also pretty helpful for getting a good idea of the future freshmen of Alabama and how they compare to others

Over a decade ago, Nike partnered with some prominent names in football like Pete Carroll to develop a series of athletic tests that represented an athlete’s total explosiveness and sheer athleticism, then combined them in a formula weighted by the player’s size and muscle groups that are more useful in the sport of football in order to get a single numeric value, known as SPARQ.

An example from Alabama’s 2016 recruiting class:

*note: for height measurements, I use a system of four digits: the first is the number of feet, the next two digits are the inches, and the fourth digit is for fractions of an inch. So a height of 6015 would be 6 feet and 1 and a half inches tall.

Nigel Knott was easily the most athletic recruit in Alabama’s 2016 class. a 4.38 forty yard dash is impressive for sure, as is the 3.94 20-yard shuttle time. Even more so is that a 174-pound cornerback managed a 39.5 power throw, which is further than many linemen. But the most ridiculous feat is his 46.5-inch vertical jump.

As such, Knott ended up with a SPARQ score of 147.96, which is really blowing the roof off the place.

“But Brent,” you may ask, “how do I know what a good or bad SPARQ score is? It just seems arbitrary!”

Allow me to nerd for a moment:

That’s where some statistics work comes in. Have you ever heard of standard deviations? Maybe? No? Well what about percentiles and bell curves? In any data set, there will be a distribution of high scores and low scores, with most falling somewhere in the middle.

Think about grades in school: There are a few people that make A’s, more that make B’s, a big chunk that makes C’s, a few less that make D’s, and those last few that make F’s. Depending on teachers, school systems, or demographics, that may shift one way or another, but you’ll still see that “bell curve” where there are few people at the extremes and most are in the middle.

A standard deviation, then, is a number calculated that represents how spread out, or varying, that data set is. If the entire class made C’s, except for one person making an A and one making an F, then the standard deviation would be really small. On the other hand, a set of data with 20 students making A’s and F’s would have a really large standard deviation.

So how does this relate to SPARQ? Doesn’t it already measure how athletic a player is?

Think about it this way: If I asked you who was a more impressive athlete (not football skills, but in pure athleticism) between former offensive tackle Cam Robinson and former defensive back Maurice Smith, who would you say?

If you answered Cam Robinson, you would be correct, and pretty much everyone in the media would back you up.

Except SPARQ, that is. Robinson tested at 97.50 while Smith ended up with a 106.89. When you start looking at large groups of players, you begin to notice that the smaller, quicker defensive backs and receivers get routinely higher scores than even the most impressive offensive linemen.

So, a couple of years ago, I set out to find the standard deviation of SPARQ scores for each position group. With some spreadsheet and math wizardry, I put together a number that represents how many deviations from the standard (the average) that a player is compared to others of his position. This number is called many different names by the stats community, but I like the term “Z-score.”

Basically, it normalizes the SPARQ between different positions so that you can compare just how impressive of an athlete each different player is relative to his position.

Here are three players from the Tide’s 2016. Nigel Knott’s Z-score was a 2.93, Jonah Williams a 1.50, and T.J. Simmons a 0.04.

A score of 0 would mean that the player is a perfectly average college athlete, or more athletic than 50% of all college football players. A negative value means he is less athletic than average and a positive is more athletic.

A Z-score of 1.0 means that he is more athletic than about 84% of all college football players. A 2.0 means that he is more athletic than 97.7%. And a 3.0 means he is more athletic than 99.9%.

So, back to our previous three players: T.J. Simmons is barely above average in terms of athleticism. For an Alabama player, that’s about as low as it gets. We rarely see any recruits in the below average category here.

Jonah Williams, on the other hand is an exceptionally elite athlete, residing in the 93rd percentile.

Nigel Knott, on the other hand, is just otherworldly. His Z-score of 2.93 pretty much means he’s going to be one of, if not THE most athletic players in all of college football. In fact, only Speedy Noil and Donovan Peoples-Jones have tested higher than him in the years since 2013.

In the next few weeks, recruiting coverage is really going to ramp up here at RBR. Throughout that time, I’ll be constantly referencing SPARQ and Z-scores in my player evaluations.

Athleticism is a good starting place for analyzing players, but it’s only part of what makes a great football team. They must also have football skill, awareness, and discipline in order to be successful.

But being faster than everyone on the other team never hurts.