Which Draft-Hopeful Running Backs Produced Like NFL Studs in College?
There's a lot of randomness to the NFL Draft. It's hard -- basically impossible -- to beat it consistently.
But that doesn't stop teams and analysts from finding things that can help increase the chance of hitting on a prospect.
I mean, think of it this way: if we know Toad and Yoshi win all the time in Mario Kart 64, then why would we select Bowser as our racing character?
If we know the vast majority of successful NFL running backs shared similar achievements within certain metrics at the college level, then why would we ignore it?
The main goal here is to look at successful NFL running backs and see if they shared similar production characteristics in college. Then, we can see how this year's draft class stacks up.
Naturally, the main issue in a study like this -- as is the case with a lot of college-to-NFL analysis -- is that it's hard to define and differentiate success strictly.
What makes a running back successful? His ability simply to gain yards on the ground? His receiving skills? His pass-blocking?
Rather than diving into that rabbit hole, it's easiest to keep things simple: a successful running back is one who's good enough to see a lot of touches in his offense all while successfully improving his team's chances of scoring.
That seems fair, right?
Because most running backs won't sniff 300 carries in a single season (unless they're good), anything above that mark means they've more than likely had more than one year of seeing opportunity in an NFL offense. That'll define the touches part of all of this: any running back with 300 or more touches during his NFL career.
(This is very scientific.)
To determine if the back has helped or hurt his team through the years, we can turn to Net Expected Points (or NEP, which you can read more about NEP in our glossary), which tells us how many real points a player adds or loses for his team.
More specifically, our Success Rate metric shows us how often a running back makes a successful, positive expected-point play for his team -- it's the percentage of carries made by a running back that positively impacts his team's scoring chances.
Through the years, the average Success Rate for a running back has hovered the 40% mark. In other words, 40% of rushes have been deemed a success through our expected points model.
So we'll stick with that number as our production parameter: any running back with a career Success Rate of 40% or greater.
Combining those two aspects -- 300 attempts and a 40%-plus Success Rate -- gives us 51 different FBS (FCS isn't included, which admittedly lowers the sample a tad) running backs since 2005. That's the sample we're working with.
How similar were those running back production profiles in college?
Now, it doesn't make a ton of sense to look at, say, the college Freshman years of these 51 backs. Instead, to find similarities among them, I only examined their final, right-before-the-draft college seasons.
So that's 51 seasons for 51 running backs. Some of them played more than others (Adrian Peterson was the only one to play fewer than 10 games), but this at least gives us something to work with.
Here's a snapshot of the running back sample's college averages in major categories:
|Attempts Per Game||19.18|
|Market Share (Attempts)||47.28%|
|Yards Per Carry||5.69|
|Market Share (Ru. Yards)||57.03%|
|Market Share (Ru. TDs)||52.49%|
|Receptions Per Game||1.86|
|Market Share (Receptions)||10.13%|
|Market Share (Rec. Yards)||7.40%|
It looks like successful NFL backs are successful in some way at the college level. Ground-breaking stuff, I know.
But it's not just with yardage and touchdown totals. It's with usage. Among the 51-player subset, only four backs had a sub-30% market share in attempts, with the lowest being Felix Jones' -- one of the subjectively worst NFL backs in the group -- 21.28% mark.
Another thing to note: pass-catching matters. Having a rate of 22.94 receptions is no joke when you consider the ordinary 100-attempt college back has averaged a little over 15 receptions per season since 2005.
Players of Concern
Let's translate this information over to the 2017 class, looking at players who've got some concerning production profiles.
(Players selected to analyze were taken from WalterFootball.com's top running back prospects.)
|Player||School||Games||Attempts||Attempts/G||Att Market Share|
|Curtis Samuel||Ohio State||13||97||7.46||16.64%|
|Tarean Folston||Notre Dame||10||77||7.70||17.58%|
|Justin Davis||Southern California||10||110||11.00||21.91%|
|Joseph Yearby||Miami (FL)||13||102||7.85||23.45%|
The most polarizing film-versus-numbers running back in this draft class is probably Alvin Kamara, a committee back for two years at Tennessee who is entering the draft after his Junior season. Because he wasn't the lead dog in his backfield, he saw just 19.92% of the team's touches this past year, averaging fewer than 10 attempts per game.
Meanwhile, many are comparing him to Jamaal Charles, and some think he's a first-round value. It's entirely possible that he becomes the next hyper-efficient, higher-volume runner at the NFL level, but he'd also become an exception to the rule. As noted above, none of the backs in the sample saw a rushing attempt market share lower than 21.28%. That includes Jamaal Charles, who had a 47.87% attempt market share during his last season at Texas.
Curtis Samuel is another dude who could be drafted early-ish, but his situation is different than Kamara's. Whereas a team drafting Kamara is looking for a Charles-like back, Samuel profiles more as a Percy Harvin type: he caught just 23 fewer passes this season (74) as he saw rushes (97). He's listed as a running back, but it's unlikely that he'll be a traditional one at the NFL level -- he's actually listed as a wide receiver for the NFL Combine.
Rushing Production Concerns
|Player||School||Ru. Yards||Ru. Yds Per Carry||Ru. Yds Market Share|
|Tarean Folston||Notre Dame||334||4.34||17.04%|
|Justin Davis||Southern California||607||5.52||23.22%|
|Curtis Samuel||Ohio State||771||7.95||24.18%|
|Joseph Yearby||Miami (FL)||608||5.96||30.82%|
|Marlon Mack||South Florida||1,187||6.82||31.96%|
|Elijah Hood||North Carolina||858||5.92||45.25%|
The table above depicts draft hopefuls who saw fewer than 50% of their team's rushing production this past year. (Due to Leonard Fournette's injury, I removed him from the list. Trolls would undoubtedly hit my mentions if he was on here.) Keep in mind, the average "successful" NFL back had a 57.03% mark in the rushing yard market share department.
The reason to look at production from this angle -- as opposed to raw rushing yards -- is pretty obvious: some teams run more than others do. Corey Clements, for instance, rushed for 1,375 yards this past season. He also played for Wisconsin, though, who ran the ball 658 times in 2016, the sixth-most at the FBS level.
Of the successful NFL backs in our sample who saw sub-50% rushing yard market shares -- 19 of them -- the majority were able to combat the lack of production with high efficiency. In fact, only 6 of these 19 NFL backs had yards per carry averages lower than 5.00 in their final college season. That's good news for players like Joe Mixon, Boom Williams, and Marlon Mack, among others.
|Player||School||Games||Receptions||Rec/G||Rec Market Share|
|Jamaal Williams||Brigham Young||10||7||0.70||2.86%|
|Tarean Folston||Notre Dame||10||8||0.80||3.57%|
|Joseph Yearby||Miami (FL)||13||10||0.77||3.80%|
|Justin Davis||Southern California||10||14||1.40||4.61%|
D'Onta Foreman was one of the best pure runners in the country in 2016, tallying more than 2,000 yards on the ground. And while it was revealed recently that he played 2016 with a broken hand, you've got to be concerned about his receiving totals. He ended up with just 2.90% of Texas' receptions this past year, when the average successful NFL running back saw a reception market share of 10.13% during his final college season.
And Brian Hill, who could end up being a mid-round pick, is in the same boat. Hill ended up catching just 0.57 balls per game this past year, which was the sixth-lowest rate among 300-plus attempt FBS runners since 2005. His season, to be blunt, was Andre Williams-esque. That's not a good thing.
Fitting the Mold
Now, despite the red flags within this 2017 draft class, there are plenty of backs who fit the mold of a successful NFL runner.
The table below shows players who'll enter the draft after seeing at least average marks -- you can reference the successful running back table above to find the exact numbers -- in attempt, rushing yard, reception, and receiving yard market shares. (Note: For Leonard Fournette, his 2015 season was used.)
|Player||Year||School||Att. MS||Ru. Yds MS||Rec. MS||Rec. Yds MS|
|Leonard Fournette||2015||Louisiana State||59.41%||63.37%||12.75%||11.69%|
|Donnel Pumphrey||2016||San Diego State||54.96%||57.95%||17.42%||11.42%|
|Aaron Jones||2016||Texas-El Paso||56.27%||79.47%||14.36%||10.42%|
|Matt Dayes||2016||NC State||49.40%||57.44%||12.17%||7.89%|
|Dalvin Cook||2016||Florida State||55.60%||67.11%||13.52%||14.21%|
|Jeremy McNichols||2016||Boise State||63.69%||75.35%||14.62%||12.22%|
The sheer number of backs hitting these production parameters tells us just how impressive the class is.
It's nice to see players like Leonard Fournette, Dalvin Cook, and Christian McCaffrey make this list. The majority of analyst draft boards you'll find will have those three atop their prospect rankings, which means numbers and film are matching up. That's a win.
And that's really the case for the most part here, aside from someone like Donnel Pumphrey, who's got a small frame and is coming from a smaller program. The NFL Combine itself will weed out some of these players, too.
Overall, the biggest misses versus the consensus look to be the aforementioned Kamara and Mixon, players who split backfield touches during their final college seasons. Or maybe Foreman, who didn't show much receiving ability at the collegiate level. But as I've mentioned, the fact that they're not on this final list doesn't mean they won't have success at the NFL level.
It just means probability isn't in their favor.