Speed Score 2018
by Bryan Knowles
Breaking news from the NFL Scouting Combine: Saquon Barkley is pretty good.
Barkley entered the Combine already the consensus top running back. He's the top player on ESPN's player rankings, he's in the top 10 in every mock draft you can find, and he's been compared to Barry Sanders, LaDainian Tomlinson and Superman. That's set a pretty high bar for his Combine performance.
All Barkley did under that pressure was bench 29 reps (more than Joe Thomas), have a 41-inch vertical jump (higher than Julio Jones), put up a 1.54-second 10-yard split (quicker than DeSean Jackson) and turn in a 4.4 flat 40-yard dash (faster than Devin Hester). That was "only" the second-fastest time for running backs at the combine, but he did all that weighing 233. The result is a Speed Score of 124.3, the fourth-highest since they started electronic timing in 1999.
Created by Bill Barnwell and introduced in Pro Football Prospectus and ESPN Insider back in 2008, Speed Score is one of Football Outsiders' metrics for evaluating running back prospects. It's built on the simple idea that smaller backs tend to run faster than larger backs, so we should be more impressed by a 4.5-second 40-yard dash from a 220-pound back than the same clock reading from a 170-pound back. As such, Speed Score incorporates a back's official time in the 40-yard dash with his weight to produce a measure of his speed given his size using this formula:
The average running back who makes it to the NFL will have a Speed Score around 100, with most running back prospects falling between 85 and 110.
There's no such thing as a can't-miss prospect, but players who breach that 110 score threshold have phenomenal track records. We went back and looked at every running back prospect who ran at the combine between 1999 and 2013 and checked their average yards per season over the first five years of their career. The 110+ club was between 2.5 and 4 times more likely to hit each and every benchmark then the average running back.
Speed Score table! X% of players with a Speed Score of Y go on to average Z yards per season.
In short: Run really fast, be really big, or be Arian Foster. pic.twitter.com/PqdT7CWf0b
— Bryan Knowles (@BryKno) March 2, 2018
So, Barkley's time means he's guaranteed a successful career, right? After all, he's not just past the 110 mark, he's blasted over 120, into the stratosphere. Surely, all the previous members of the 120-club went on to great NFL success, right? Well, not so much.
|120 Speed Score Club||Year||Player||School||Weight||40 Time||Speed Score||Career Yards|
|2018||Saquon Barkley||Penn State||233||4.40||124.3||0|
|2005||Brandon Jacobs||Southern Illinois||267||4.56||123.5||5094|
|2004||Kevin Jones||Virginia Tech||227||4.38||123.4||3176|
|2009||Andre Brown||North Carolina State||224||4.37||122.8||876|
|2008||Chris Johnson||East Carolina||194||4.24||120.7||9651|
That's not exactly what you'd call a murder's row of elite running backs, there. Injuries have kept Keith Marshall from seeing the field and ended Mario Fannin's career before it began, while Knile Davis ended up behind Jamaal Charles' on the Chiefs depth chart. Chris Johnson was a very good running back for half a decade, but everything else would be a disappointing result for someone of Barkley's caliber.
Of course, none of that group was touted as highly as Barkley. Marshall, Fannin and Davis all had fewer than 1,900 NCAA rushing yards, so they could be considered "workout warriors" or a sort. Only Johnson and Jones were first-round picks, and none of the players listed paired their pure athletic traits with as much on-field success and production as Barkley has. Barkley put on an absolute show at the Combine, and expect his hype to hit a new stratosphere now.
Now the caveats: Speed Score measures speed in the context of strength and power. It doesn't measure agility, receiving ability or any of the other aspects related to the position. It does not claim that a larger player with a higher 40 time is somehow faster than a smaller player with a lower 40 time thanks to the power of exponentiation. Speed Score is useful because it's beneficial for a running back to be both fast and large.
Speed Score has a higher correlation with yards, carries and DYAR than 40-yard times alone, making it a better way to contextualize the performances at the Underwear Olympics and a better tool for finding valuable players later in the draft. It's also only part of our BackCAST projections, which combine these numbers with college production and will come out later this offseason.
Barkley wasn't the only running back who put up big numbers on Friday, however. Four running backs ended up topping 110, the most since the loaded 2012 class had seven. Kalen Ballage of Arizona State, Rashaad Penny of San Diego State and Darrius Guice of LSU also cracked the 110 barrier.
Guice might well be the second-highest touted running back of the class, even coming off of an injury-plagued 2017 season. His Combine performance looked more like the Guice of 2016, when he led the SEC in rushing despite only starting six games, backing up Leonard Fournette. If he's back up to top speed, he's going to be a wrecking ball in the pros.
Penny led FBS in rushing yards in 2017, but was criticized for not being explosive enough to be a truly great back. A 4.46-second 40, third-fastest of the day, might turn some heads, and has him pretty firmly planted as a second-day pick.
Ballage is the odd one out here. He tied the FBS record with an eight-touchdown day in 2016, with most of those coming from direct snaps out of the Wildcat in the red zone. He can return kicks, catch passes out of the backfield and can line up as a slot receiver. His athleticism has just never translated to the field as one might hope; he displayed poor vision and decisiveness at the college level. Add his combine performance to his potential versatility, however, and someone might well find a third-day steal with Ballage.
Further down the list, we see Nick Chubb beating out Sony Michel in the battle of the Georgia running backs, 108.8 to 100.7; neither of that is a bad performance by any means, but Michel probably did the worst out of the tippy-top running backs – unless you include USC's Ronald Jones in that group. Jones came up lame with an injured hamstring midway through his first 40-yard attempt and only managed a 4.65-second 40, for an injury-asterisked Speed Score of 87.7.
Of course, a low Speed Score doesn't mean you're necessarily doomed. Alvin Kamara put up a 99.0 last year (to go along with a huge SPARQ score and great performances in the drills) and Kareem Hunt only managed a 94.8 after an unusually (and, it turns out, uncharacteristically) slow 4.62 in the 40. They both turned out alright.
The all-time low for a productive running back is Ahmad Bradshaw's 87.7 back in 2007, and only two backs today managed a lower score than that: the aforementioned injured Jones and West Virginia's Justin Crawford, whose 87.2 probably erased the slim chances he had of getting drafted.
Friday was all about Saquon Barkley, however. Add a great college career to one of the most impressive days we've ever seen at the combine, and the hype train is about to go into overdrive. All aboard.
|2018 Speed Scores||Player||School||Weight||40 Time||Speed Score|
|Saquon Barkley||Penn State||233||4.40||124.3|
|Kalen Ballage||Arizona State||228||4.46||115.2|
|Rashaad Penny||San Diego State||220||4.46||111.2|
|Nyheim Hines||N.C. State||198||4.38||107.6|
|Ryan Nall||Oregon State||232||4.58||105.5|
|Jarvion Franklin||Western Michigan||225||4.63||97.9|
|Donnie Ernsberger||Western Michigan||255||4.78||97.7|
|Roc Thomas||Jacksonville State||198||4.56||91.6|
|Demario Richard||Arizona State||218||4.70||89.4|
|Jordan Wilkins||Ole Miss||216||4.71||87.8|
|Justin Crawford||West Virginia||202||4.64||87.2|
Did not run: Josh Adams, Nick Bawden, Kerryon Johnson, John Kelly, Jeff Wilson
21 comments, Last at 11 Mar 2018, 4:43pm
#20 by JimZipCode // Mar 08, 2018 - 2:41am
I know this is just anecdata, but Ray Rice had a pretty low speed score in 2008. That's because he was teeny-tiny: his 40 time was actually quite good (4.42). Led the league in yards-from-scrimmage in 2011; was close to the lead in '09 and '10.
I don't have any suggestion on what to do with that info. The article already mentions that speed score can't say anything about receiving chops or elusiveness. I guess Rice is just an outlier. But it's interesting.
#11 by wabuffo // Mar 05, 2018 - 3:36pm
...and try to find a film clip of someone running him down from behind.
I tried thinking about a way to measure this from game play stats. That is to say, a way to measure breakaway speed for RBs. I thought of looking for the number of running plays where a RB ran more than 50 yards for a TD. That's half the field for some defensive player to try to bring down a RB after he's broken through. You see it all the time, where a RB gets caught from behind or runs out of gas.
Unfortunately - the football reference game play stats go back only to 1994. But the leaders for 1994-2017 were:
1) Barry Sanders 12 TDs, 1630 Rush Att. .74 per 100 rush attempts.
2) Napoleon Kaufman 6 TDs, 978 Rush Att. .61 per 100 rush attempts.
3) Robert Smith 8 TDs, 1411 Rush Att. .57 per 100 rush attempts.
4) Adrian Peterson 14 TDs,2574 Rush Att. .54 per 100 rush attempts.
5) Chris Johnson 11 TDs, 2163 Rush Att. .51 per 100 rush attempts.
Obviously, the early part of Barry Sanders career is not part of the data set.
I did some manual boxscore by boxscore looks for other names. Eric Dickerson (.26) and Tony Dorsett (.32) did not score very highly.
Bo Jackson 5 TDs, 515 Rush Att. .97 per 100 rush attempts.
I think this reinforces Bo Jackson's break-away speed, although this is not a perfect way to measure this, obviously.
FWIW - good discussion!
#13 by Aaron Brooks G… // Mar 05, 2018 - 6:08pm
The problem is that Barry got caught from behind all the time. He had amazing acceleration but his top-end wasn't world-class. DBs could reel him in after 50-70 yards.
Granted, Bo famously got caught from behind, too.
#16 by Bobman // Mar 06, 2018 - 12:54am
My memories of Sanders often have him tackled from the SIDE because he jitterbugged around the backfield for an eternity, looking fore that hole to explode through. Not a Lions fan (or a loins fan, to RJ) but it used to frustrate the hell out of me. Then at the end of the game he'd have a 5.0 ypc. I'd mentally erase those 4-5 backfield tackles and his positive ypc was considerably higher.
eh, I guess you take the bad with the good...
#18 by Aaron Brooks G… // Mar 06, 2018 - 9:41am
This is what I'm talking about, Huge acceleration and burst through the hole, but he needs to break two late tackles because DBs caught up to him.
God. Imagine him on a vintage Chargers or Chiefs team, or what he could do on the Patriots. He was unstoppable in open space. I'm glad I got to see his career, but man, I wish he didn't get stuck on the Lions.
#7 by ChrisS // Mar 05, 2018 - 1:02pm
The NY Times had an article about the limited usefulness of this data https://www.nytimes.com/2018/01/04/sports/football/nfl-speed-leonard-fournette.html?rref=collection%2Fsectioncollection%2Fsports&action=click&contentCollection=sports®ion=rank&module=package&version=highlights&contentPlacement=3&pgtype=sectionfront
lightly edited quote: "At the 2017 combine, 53 players ran the 40-yard dash faster than Louisiana State’s Leonard Fournette (4.51 seconds). Fournette has actually clocked the N.F.L.’s fastest speed on offense this season, according to Sportradar. Against the Pittsburgh Steelers in Week 5, Fournette showed why game speed is more important than combine speed. Fournette found a hole on the left side and broke through to the open field. From there, it was a foot race against Steelers safety Sean Davis (4.46 in the 40-yard dash), who was closing fast from the right side. But Fournette found an extra gear, reaching a top speed of 22.05 m.p.h. to pull away and score a 90-yard touchdown."
#15 by Bobman // Mar 06, 2018 - 12:51am
Agreed. And one reason I like the derogatory nickname the underwear olympics. Put everybody in full football pants, at least.
Never timed him, but my oldest actually seems to run faster in pads than shorts. Maybe that's because it actually counts in a game and not on a track. Probably why he wears his football girdle in lacrosse--makes him faster psychologically.
#19 by Pat // Mar 07, 2018 - 1:13pm
Yeah, um, that paragraph is nonsense.
1: Fournette was already in stride when Sean Davis was starting to run.
2: Davis had farther to go, since he was off to the side. Even if Davis had only a yard farther to go and they were both moving at their average 40-yard speeds, he couldn't've caught Fournette.
3: Davis and Fournette both had 1.86 seconds between their 40 yard and 20 yard dashes. They both had pretty much *exactly the same* top speed. There was no way Davis was ever going to be able to catch him, and he knew, which is why he stopped running at full speed.
4: Most damningly, they're comparing *peak speed* to 40 yard time. Which is totally different.
It's not like Fournette's 4.51 as a RB was *slow*. Especially for a RB. And his 20 yard-to-40 yard time (1.86 seconds) translates to (wait for it...) *22 mph*. Which is... pretty much exactly the same as most of the other top running backs in the draft. And *especially* for a 240 pound RB, that's a very good top speed.
And of course, the hilarious thing about this is... Fournette, who hit the fastest peak speed in the NFL last year... had the highest Speed Score at the combine in his year.
#5 by Will Allen // Mar 05, 2018 - 11:24am
No hand timed 40s should be taken at face value. Era-adjusted, it is hard to top Jim Brown for a speed and power combination. 60 years ago, a 232 pound running back was an absolute giant, and try to find a film clip of someone running him down from behind.
#12 by Travis // Mar 05, 2018 - 3:53pm
Thorpe had to have been among the fastest men in the world in his time, and someone nicknamed "Bronko" probably was not.
FWIW, Thorpe ran an 11.2 decathlon 100 meters at the 1912 Olympics while presumably holding something back for/tired from the 15 other events in which he'd compete. The 100-meter gold medalist ran 10.8 in the final.
#17 by Will Allen // Mar 06, 2018 - 6:57am
A team comprised of 11 Jim Thorpes, versus a team of 11 Bronco Nagurskis, would probably be the ultimate game of the two way football era, although you can't ignore the likes of Sammy Baugh and Don Hutson, and the underrated Chuck Bednarik was playing on both sides, at an All Pro level, in the late 50s.
#14 by Bobman // Mar 06, 2018 - 12:47am
Speaking of RB size/speed, I have a friend, a retired MD who was captain of the Harvard D like 45 years ago. He's about 6-0 or 6-1 and weighs I'd guess about 200. I asked if he played safety. "No," he said, half-amused, half offended. "DE."
"Holy cow, were you a lot heavier in your playing days?"
"No, but the other guys were smaller too. But then Ed Marinaro came along. Nobody on my team ever saw a 240 lb RB and man could he fly. He beat the hell out of us."
A little checking indicates Marinaro probably weighed closer to 215, but I guess when your standard was a 5-11/185 lb guy, that extra 30 lbs coming at you that fast makes your day a whole lot longer.
All hail the speed score.
Where does Dickerson fit in? He was big.
#2 by wabuffo // Mar 04, 2018 - 12:08pm
If we had official data going back decades, I wonder which RB would've had the highest Speed Score in their athletic prime?
My guess would be Bo Jackson. He was rumored to have run a 4.16 in the 40 at one of the scouting events with a playing weight of 227lb according to Pro Football-Reference.
That would give him a Speed Score of 151.6.
Even if his 40-time was cut back a bit, he would still be at the top of this list.
#4 by Aaron Brooks G… // Mar 05, 2018 - 8:58am
4.13, but that was on an indoor track. The combine is run on turf.
Other possibles: Michael Turner
Jay Hinton (who?)
Ahman Green apparently ran a 4.17 40 in a pro day, although his combine numbers were much more pedestrian.
The best comp would probably be Herschel Walker. As big and a more accomplished track star, he had better 55m and 100m times than Jackson did.
He was less fluid and less all-around versatile (although that's arguable, given his bobsled and MMA exploits), but seemed to have every bit the burst. 55m isn't the 40y (it's less pure acceleration, but not yet pure top-end), but it's similar. Their 55-100m splits suggest Walker might have had a touch more top end, too.
The other two guys in the discussion are OJ and Curtis Dickey.
If you open it to other positions, Mandarich had a 141 based on 315 pounds and a 4.6 40. Chemically-enhanced, though.