First we look at the college production of a specific position group. For the safeties, we'll be using a metric based on Production Points (the metric is explained in detail in the post on linebackers), which looks at the key stats for safeties and weights them with a point system that gives you a single number that shows how many Production Points a player averaged per game.
Then we match that with a metric for athleticism, SPARQ, which is a single number designed to summarize a player's athleticism. The number is calculated with a proprietary formula that incorporates player weight, bench press, broad jump, vertical jump, forty-yard dash, ten-yard split, short shuttle and 3-cone drill (details here).
Once we have those two metrics, we combine them to see which draft prospects emerge as the most productive AND most athletic.
Let's kick things off with the Production Points for this year's safety class (all the caveats and provisos from the previous posts still apply). The next table features 20 safeties, all with a draftable grade according to the CBS Sports big board and sorted by their rank on the same board from April 17. The table is sortable so you can see who ranks where for each category (just click on the blue column headers):
|Production Points for Safeties in 2017 NFL Draft|
|FS||7||Malik Hooker||Ohio State||6-1||206||1||84||5.5||0||0.5||4||7||0||25||6.9|
|SS||63||Josh Jones||NC State||6-1||220||2||172||6.0||0||1.5||10||4||2||26||10.3|
|SS||110||Justin Evans||Texas A&M||6-0||199||3-4||165||6.0||0||0.0||11||5||0||25||10.2|
|FS||128||Rayshawn Jenkins||Miami (Fla.)||6-1||214||4||128||7.0||2||1.5||11||5||0||26||8.9|
|SS||151||Xavier Woods||Louisiana Tech||5-11||197||4-5||145||14.0||1||3.0||9||8||3||27||11.2|
|FS||168||John Johnson||Boston College||6-1||205||5||140||4.0||1||1.5||12||6||3||25||10.5|
|SS||218||Lorenzo Jerome||Saint Francis (PA)||5-10||204||6||117||11.0||3||5.0||12||9||0||20||14.0|
|FS||234||Jordan Sterns||Oklahoma State||5-11||198||6-7||209||7.0||0||0.0||8||5||1||25||11.9|
For this safety class, a Production Score of 10 is average. Going by what we've seen from other position groups using this metric, a number above 12.0 suggests very high college productivity, and anything above 14.0 is exceptional.
Having said that, we can't look at Production Points in isolation, but need to add some context to these numbers. We know that the stats for small-school players tend to be overinflated because of the quality of the opponents played. Also, we need to factor in the type of role each player had in college.
Some players played both corner and safety during their college careers. Others, like Jabrill Peppers, played more of a hybrid linebacker/safety role, others yet again spent more time in deep center field than in the box - all these things affect a player's college production. That's why you want to see players with a track record of big defensive plays against the pass and against the run.
Note also that players like Malik Hooker and Jadar Johnson only became starters in their final college season, and using their last two college seasons to calculate production points may not do them justice. Looking at just their final season, both Hooker (12.4 points) and Johnson (9.5) have much better scores.
Overall, the standout in terms of Production Points here is Lorenzo Jerome, whose 14.0 points beats all other safeties, but that comes with the small-school caveat.
Jordan Sterns and Obi Melifonwu follow with scores very close to the 12-point threshold. Sterns was very productive at Oklahoma State, but his lack of athleticism probably limits his potential at the next level, hence his late-round grade. The imposing, 6'4" Melifonwu is exactly what you are looking for as an enforcer in the backfield, but questions about his coverage ability as a deep safety keep him from the top of the draft, though he'd be an excellent addition for a team drafting at the bottom of the first round looking for a downhill tackler playing near the line of scrimmage.
There's an interesting, nine-player strong group of players next on the list who all have 10 points or more and have all demonstrated that they can play the ball well, and these nine players are distributed through almost all rounds of the draft.
Moving on to SPARQ, our list of safeties is reduced to 13 players. Unfortunately, multiple top prospects including Malik Hooker, Jabrill Peppers, Desmond King, Justin Evans, and others do not have scores. Melifonwu doesn't have a score on 3sigmaathlete.com, but we know that he had a better SPARQ score than any other safety, so I estimated his score for the graph below. The graph provides a visual representation of what happens when we plot Production Points against the SPARQ score for 2017 safety class.
Going clockwise from the top left of the graph, the C quadrant features players with a strong record of production at the college level, but that have questions regarding their athletic ability. The A quadrant (top right) shows the players most likely to succeed at the NFL level; they have a strong track record of production and combine that with the necessary athleticism to allow them to compete at the NFL level. The B quadrant (bottom right) shows superior athletes whose college production has been sub par, leaving scouts to question why this might be the case. The D quadrant (bottom left) is a nasty place for a prospect to find himself; it's where the guys sit whose college production and athletic markers are both below those of their peers.
With seven draftable payers missing from the graph above, it's hard to evaluate the overall quality of this safety draft class. This class offers four players with above average athleticism (Quadrants A+B), two more (Xavier Woods and John Johnson) sit right on the line to the A-quadrant.
The Cowboys invited five safeties for official pre-draft visits, the four players marked in blue above, as well as Texas A&M's Justin Evans, so there's definitely some interest in the position. The only question is, will they be able to get a talent (probably in the mid- to late rounds) that is better than what they already have on the roster? If they look at some of the A-quadrant players, the odds of that happening will likely increase.
Again, there's not a lot of historic SPARQ data around, but these are the numbers I could get my hands on for safeties.
For most other positions, we've seen a stronger correlation between production and athleticism when looking at successful NFL veterans. This may be due to the unique situation of the position as outlined above, may be the result of the limited number of historic players available, or it may indicate that the model simply does not work as well for safeties.