Travon Walker and the Case for the Combine

Georgia Bulldogs ER Travon Walker
Georgia Bulldogs ER Travon Walker
Photo: USA Today Sports Images

NFL Draft - With no top-shelf quarterback prospects available in this year's NFL draft, teams have understandably turned to the most valuable position on the defense: edge rusher. Michigan's Aidan Hutchinson is a strong candidate to be the No. 1 overall selection, and many analysts project that Oregon's Kayvon Thibodeaux and Georgia's Travon Walker will also be drafted in the top 10. Edge rusher is widely considered to be one of the strengths—if not the greatest strength—of this year's draft.

So will the 2022 NFL draft's edge rusher class live up to its reputation? One tool that could potentially help answer that question is SackSEER, Football Outsiders' system for projecting college edge rushers. SackSEER uses pre-draft workout data along with college statistics to project the NFL pass-rushing prospects of defensive ends and 3-4 outside linebackers.

SackSEER agrees with the conventional wisdom that this is a generally strong and deep draft for edge rushers. That said, contrary to some draft hype, there is a considerable degree of uncertainty as to just who is most likely to succeed. In SackSEER, Hutchinson, Thibodeaux, and Walker are all effectively tied, with only a small fraction of a sack separating their projections.

SackSEER is based on a statistical analysis of all edge rushers drafted in the years 1998 to 2019, and measures the following:

  • The edge rusher's projected draft position. Specifically, the rankings from ESPN's Scouts Inc.
  • An "explosion index" that measures the prospect's scores in the 40-yard dash, the vertical leap, and the broad jump in pre-draft workouts.
  • The prospect's score on the 3-cone drill.
  • A metric called "SRAM" which stands for "sack rate as modified." SRAM measures the prospect's per-game sack productivity, but with adjustments for factors such as early entry in the NFL draft and position switches during college.
  • The prospect's college passes defensed divided by college games played.

As always, Football Outsiders' 2022 NFL draft coverage is presented by Underdog Fantasy!

Underdog Fantasy

Below, we take a look at some of SackSEER's top prospects in the 2022 NFL draft, along with some similar prospects from previous drafts.

Travon Walker, Georgia

SackSEER Projection: 27.2 Sacks Through Fifth Season
Scouts Inc.: No. 8 Overall
Similar Historical Prospects: Ezekiel Ansah, Brian Orakpo

Travon Walker's draft stock has risen in recent weeks, and in SackSEER he manages to edge Aidan Hutchinson and Kayvon Thibodeaux by a small margin.

Walker's best SackSEER metrics come from his terrific combine. His workout numbers ranged from good to fantastic. His 4.51s 40-yard dash was great, especially for a 272-pound player. His 6.89s 3-cone drill is in the top 10% of drafted edge rusher times. His 35.5-inch vertical leap and 10-foot, 3-inch broad jump were also good.

Of course, what drags him down a bit—and why many NFL fans are likely skeptical of his high draft position—is his lack of college production. Walker had only 9.5 sacks in three college seasons. Is Walker another "workout warrior" whose lack of college production foreshadows a disappointing NFL career? Don't be so sure.

It may be counterintuitive, but workout scores are a more significant indicator of NFL success than college sack rate. Of course, you would prefer a prospect who had both production and workouts, but if you had to choose, historically you would be better off with the "workout warrior" than the productive player with bad workouts. In college, players such as Ezekiel Ansah and Clay Matthews had even fewer career sacks than Walker, and both turned out to be strong NFL players. Meanwhile, Jarvis Jones and Calvin Pace—strong college performers with poor workout numbers—had only a limited impact in the NFL.

Walker is a strong prospect. However, like the other members of this year's "big three," he is not a "generational" prospect, and his projection is weaker than the projection for recent top edge rusher prospects such as Myles Garrett and Chase Young.

Aidan Hutchinson, Michigan

SackSEER Projection: 26.9 Sacks Through Fifth Season
Scouts Inc.: No. 1 Overall
Similar Historical Prospects: Robert Quinn, Aldon Smith

Aidan Hutchinson is a prospect that SackSEER has difficulty figuring out. On the one hand, Hutchinson was terrific as a senior, recording 14 sacks in 14 games. However, SackSEER is normally skeptical of a prospect with only one good season, especially when that season was the player's senior season. In Hutchinson's case, of course, that gloss would be totally unfair. Hutchinson did not have a true junior year, as Michigan only played two games in a COVID-shortened season. You really would have liked to see what Hutchinson could have done as a junior. To Hutchinson's credit, he finished with a strong passes defensed rate, which does suggest professional success.

Hutchinson's workouts were similarly mixed. Hutchinson's 6.73s 3-cone drill was terrific—it is better than the times of all but three edge rushers drafted since 1998. His explosion drills, however, were just average.

It all adds up to a projection that is certainly positive—there is more to like than not to like with Hutchinson. However, there are some weaknesses in his projection, which means that Hutchinson is a very good prospect rather than a great one.

Kayvon Thibodeaux, Oregon

SackSEER Projection: 26.6 Sacks Through Fifth Season
Scouts Inc.: No. 7 Overall
Similar Historical Prospects: Brian Burns, Ryan Kerrigan

Kayvon Thibodeaux is certainly athletic, but his workouts do not quite match his reputation. Thibodeaux had a great 4.58s 40-yard dash at the combine. However, he reserved the other workout drills for his pro day, and those numbers were less impressive. His 9-foot, 11-inch broad jump was lower than what you would expect from a player with a similar 40, and Thibodeaux did not perform a vertical leap. Thibodeaux also ran the 3-cone in 7.32 seconds, which is much, much slower than one would expect from a player with Thibodeaux's straight-line speed. Again, Thibodeaux's workouts were good, but they were far worse than other combine-busters of drafts past, such as Von Miller and Khalil Mack.

To Thibodeaux's credit, similar prospects who were hyped as athletic but disappointed in actual workouts—namely, Jason Pierre-Paul and Carlos Dunlap, whose workouts were far worse than Thibodeaux's—ultimately had an NFL career more consistent with their hype than their workout numbers. Thibodeaux could certainly follow a similar path, and he may just be one of those unusual prospects whose athleticism just does not translate into combine drills.

Thibodeaux's production paints a similarly uncertain picture. Again, Thibodeaux was good, but not quite great. Thibodeaux had 19 sacks in 30 games. That is much better than average but falls behind some of the great prospects such as Terrell Suggs and Julius Peppers.

Thibodeaux is an excellent prospect with a lot to his credit, and he is certainly worthy of a first-round pick. However, he is not quite a "can't miss" prospect, and no result—whether it be future star or bust—should really be surprising.

George Karlaftis, Purdue

SackSEER Projection: 25.4 Sacks Through Fifth Season
Scouts Inc.: No. 24 Overall
Similar Historical Prospects: T.J. Watt, Justin Smith

Karlfatis has a strong projection and is an interesting prospect because he has some unusual splits in his SackSEER numbers. First, Karlaftis performed very well in the two jump drills, recording a 38-inch vertical leap and a 10-foot, 1-inch broad jump. Typically, a player who performs well on the jumps also runs a quick 40-yard dash. However, Karlaftis did not. He ran a 4.78s 40-yard dash, which is below average.

A similar split occurs with Karlaftis' production. Typically, edge rushers become more productive as they advance in college, which is why SackSEER is more impressed with a 10-sack junior season than a 10-sack senior one. However, Karlaftis arguably performed best as a freshman. He recorded 7.5 sacks in 12 games as a freshman while only recording 4.5 sacks in 12 games as a junior. Karlaftis, however, did record four passes defensed as a senior and his projection benefits from a strong passes defensed rate overall.

Karlaftis, in many ways, is the perfect edge rusher for this draft, which has lots of potential but also includes a fair share of question marks.

David Ojabo, Michigan

SackSEER Projection: 25.1 Sacks Through Fifth Season
Scouts Inc.: No. 20 Overall
Similar Historical Prospects: Will Smith, Harold Landry

It is a true shame that David Ojabo tore his Achilles tendon during Michigan's pro day because he is an intriguing prospect with a lot to his credit. Ojabo showed great explosion at the NFL combine, recording a 4.55s 40-yard dash, a 35-inch vertical leap, and a 10-foot, 2-inch broad jump. Ojabo loses some points because he did not have a 3-cone time and his time in the short shuttle—a relatively slow 4.55 seconds—implies that his 3-cone would have been much worse than his explosion numbers suggest.

Ojabo was also extremely productive early in his career, a good sign for success. Ojabo had 11 sacks in 13 games when just a sophomore at Michigan. However, early production is not a guarantee of professional success, as there have been prospects such as George Selvie, who put up strong sack numbers as sophomores and then faded over their final two years. A skeptic might also suggest that Ojabo benefited tremendously from having possible first overall pick Aidan Hutchinson drawing attention on the other side of the field. That said, Ojabo also made a few plays other than sacks and finished with a good passes defensed rate.

Ojabo's injury makes his likely draft position highly uncertain. A team willing to wait a year for Ojabo to recover could end up getting great value if he turns out as good as some of his SackSEER numbers suggest.

Jermaine Johnson II, Florida State

SackSEER Projection: 23.7 Sacks Through Fifth Season
Scouts Inc.: No. 10 Overall
Similar Historical Prospects: Bradley Chubb, Kamerion Wimbley

Jermaine Johnson is a very similar prospect to Travon Walker. Johnson's combine was excellent, although not quite as good as Walker's. Johnson ran a quick 4.58s 40-yard dash and had a great broad jump at 10 feet, 5 inches. Johnson did disappoint a bit at the vertical leap, gaining only 32 inches.

The biggest difference between Walker and Johnson is that Johnson did not run the 3-cone. Walker had an extremely quick 6.89s 3-cone. SackSEER thinks that Johnson's 3-cone would have been approximately 7.17 seconds based on Johnson's other drills.

Johnson was more productive than Walker, notching 11.5 sacks as a senior and recording a slightly better passes defensed rate. However, because workouts are a stronger indicator of edge rusher success than sack production, Walker's workouts put him on top. Overall, despite being SackSEER's sixth-ranked prospect, Walker is fewer than four sacks away from Walker, SackSEER's top-rated prospect. Johnson's proximity to Walker demonstrates just how closely SackSEER rates the top edge rushers in this year's draft, and how a smart team might be able to snag a promising player at virtually any point in the first round.

Drake Jackson, USC

SackSEER Projection: 20.9 Sacks Through Fifth Season
Scouts Inc.: No. 52 Overall
Similar Historical Prospects: Jerry Hughes, Rashan Gary

Drake Jackson is a bit unusual in that he performed every combine drill except for the 40-yard dash. That said, Jackson was excellent in the drills that he did perform in pre-draft workouts: he notched a 36.5-inch vertical leap and a 10-foot, 7-inch broad jump at the combine and then a 7.09s 3-cone at his pro day. Jackson had very consistent production at USC, recording nearly the same sack total in each of his three seasons—approximately one sack every two games. He also finished with four PBUs, which gives him a better-than-average passes defensed rate.

Overall, Jackson is a solid prospect and could be great value where he is currently rated, which is somewhere in the second round—and his SackSEER numbers are not far off from the top prospects. Indeed, if you did not adjust for expected draft position, Jackson's projection would be comparable to that of the big three.

Boye Mafe, Minnesota

SackSEER Projection: 19.5 Sacks Through Fifth Season
Scouts Inc.: No. 46 Overall
Similar Historical Prospects: Cliff Avril, Chandler Jones

Overall, Boye Mafe performed better than any prospect in this draft in the "explosion" drills. Mafe recorded a 4.53s 40-yard dash, a 38-inch vertical leap, and a 10-foot, 5-inch broad jump, which are all great numbers. However, he fared less well on the agility drills, and his 3-cone time of 7.24 seconds was much slower than you would expect for an edge rusher who can run a sub-4.6 40.

Mafe's projection also suffers a bit from lack of production. Mafe only recorded three passes defensed in 31 games, which translates into a below-average passes defensed rate. Mafe's sack numbers were just slightly above average for a drafted edge rusher—Mafe recorded 7.0 sacks in 12 games in 2021 and 4.5 sacks in six games in 2020.

Like many edge rushers in this draft, Mafe's projection suggests some flaws, but his strong workout numbers suggest that he could have a tantalizing upside.

Potential Sleeper

Sam Williams, Mississippi

SackSEER Projection: 12.1 Sacks Through Fifth Season
Scouts Inc.: No. 139 Overall
Similar Historical Prospects: Maxx Crosby, Mark Anderson

This year, SackSEER may differ slightly with the conventional wisdom as to whether Player X should be ranked over Player Y but agrees with the general area where draftniks rate each edge rusher. The biggest exception, however, is Sam Williams, and the closer you look at his numbers, the more you wonder why he is flying under the radar.

Williams has a profile that reads like a high pick. He ran the 40-yard dash in 4.46 seconds at 261 pounds. His vertical leap was a bit disappointing at 32.5 inches, but that sticks out because he performed a 12-foot, 3-inch broad jump, which is very good. He rounded out his combine performance with a 7.03s 3-cone at his pro day, suggesting that he has good lateral movement to go with his explosion.

Williams also had 12.5 sacks in 13 games on an SEC team last year. His production was just OK before last year. Overall, his production stacks up just fine against the other edge rushers available. Although it is a good draft for edge rushers generally, this year's class is a bit weak in sack rate. Williams' career sack rate is no worse than Travon Walker's, Aidan Hutchinson's, Jermaine Johnson's, or that of many other highly-rated prospects. Williams' worst metric is his passes defensed, but his workout numbers make up for it.

The quintessential SackSEER sleeper is a consistently productive small school edge rusher with lots of passes defensed. Historically, NFL teams tend to underrate productive edge rushers from small schools, overrate unproductive edge rushers from large programs, and completely ignore pass defensed rate. Interestingly, Williams checks none of those boxes, and instead fits the profile of a player who would typically command a high pick. However, for some reason, he seems to be slotted in the fourth round.

The following table lists the SackSEER projections and ratings of all of the 2022 edge rusher prospects who received invites to the NFL combine. SackSEER projection considers forecast draft position; SackSEER rating represents the player's position among all historical edge rusher prospects in our database, regardless of projected draft position.

2022 SackSEER Projections
Edge Rusher College Proj.
Round
Explosion
Index
SRAM PD/Rate Sack
Proj.
Rating
Travon Walker Georgia 1 1.3 0.38 0.138 27.2 89.7%
Aidan Hutchinson Michigan 1 0.3 0.50 0.278 26.9 84.5%
Kayvon Thibodeaux Oregon 1 0.9 0.63 0.233 26.6 86.0%
George Karlaftis Purdue 1 0.6 0.52 0.231 25.4 81.9%
David Ojabo Michigan 1 1.1 0.59 0.214 25.1 78.4%
Jermaine Johnson II Florida State 1 0.9 0.50 0.143 23.7 70.0%
Drake Jackson USC 2 1.2 0.49 0.222 20.9 84.9%
Boye Mafe Minnesota 2 1.6 0.46 0.097 19.5 83.0%
DeAngelo Malone Western Kentucky 2 0.9 0.62 0.140 18.5 73.1%
Arnold Ebiketie Penn State 2 1.3 0.48 0.036 17.8 72.0%
Nik Bonitto Oklahoma 2-3 1.0 0.50 0.222 17.7 76.7%
Amaré Barno Virginia Tech 3 2.3 0.39 0.143 16.5 81.5%
Sam Williams Ole Miss 4-5 1.2 0.51 0.057 12.1 81.7%
Cameron Thomas San Diego State 3 -0.1 0.63 0.061 12.0 56.9%
Myjai Sanders Cincinnati 3 0.4 0.33 0.286 10.2 10.6%
Kingsley Enagbare South Carolina 2-3 0.0 0.40 0.103 10.1 17.0%
Dominique Robinson Miami (OH) 3-4 1.0 0.32 0.037 9.9 37.1%
Isaiah Thomas Oklahoma 5-6 0.0 0.52 0.194 8.7 64.0%
Jeffrey Gunter Coastal Carolina 5 0.7 0.35 0.106 7.7 42.7%
Tyreke Smith Ohio State 4 -0.2 0.26 0.167 6.9 12.5%
Micheal Clemons Texas A&M 4 -0.2 0.37 0.063 6.7 21.3%
Alex Wright UAB UDFA -0.1 0.41 0.161 3.6 45.3%
Joshua Paschal Kentucky UDFA 0.7 0.31 0.047 3.1 42.7%
Tyree Johnson Texas A&M 6-7 0.4 0.43 0.050 3.1 23.3%
Christopher Allen Alabama 7-UDFA 0.6 0.45 0.000 1.0 17.2%
Myron Tagovailoa-Amosa Notre Dame UDFA -0.6 0.15 0.053 0.0 0.0%
Tre Williams Arkansas 6 -2.2 0.33 0.000 0.0 0.0%
Zachary Carter Florida UDFA -1.6 0.01 0.000 0.0 0.0%

Comments

22 comments, Last at 27 Apr 2022, 10:42am

1 Hutchinson

Hutchinson did not have a true junior year, as Michigan only played two games in a COVID-shortened season.

Michigan played six games in 2020; Hutchinson broke his right leg early in the third one. (Some sources have Hutchinson only playing two games, but that's because he did not make a tackle in the third.)

3 Given U-M's performance in…

In reply to by Travis

Given U-M's performance in 2020, it's fair to say they only played two games. They appeared for six games (not including their refusal to show up to play Ohio State), but only played two.

4 Like I said elsewhere

There's a small gap in athleticism but a much larger one in production.

The choices aren't athleticism and no production vs no athleticism and production, it's athleticism and no production vs athlecism AND production. 

If you're picking Travon ED1 you're probably placing too much faith in your coaches. 

5 Or you just need a 5-tech.

These guys definitely aren't interchangeable. Hutch did most of his work standing up, often from wide 9. Lots of two-lane rushes where he's given the freedom to vacate a gap. Walker played much closer to the ball, hand in the dirt, more deliberate lane assignments. Had to control blockers more often rather than just getting to shoot gaps. Was often asked to take the bad end of a stunt.

It's a big gap in production to overlook, but I think the context of how each player was used and their development levels (Walker was a true junior last year) explain a lot of that.

7 Ehhhhhh. At the end of the day

They all played the most at outside the tackles the last couple years and that's where they'll play at the next level. They're being picked high so there's no need for bad teams to worry about scheme so much.

Same age and experience as Kayvon and slightly younger than Hutch. 

11 They're being picked high so…

They're being picked high so there's no need for bad teams to worry about scheme so much.

 

And that's how bad teams stay bad.  Hiring new coaches who smash cylinders into square holes, because the highly prospect was highly rated.  

That said, I share your mistrust of Walker; he had the seventh best pressure rate of the Georgia front seven.  And almost all of them blew up the combine, except for the guy who made the most plays, Dean.  Perhaps Walker will make more plays when he's no longer stuck on the NCAA version of the 1985 Bears.  Or maybe he'll keep not getting pressures and sacks.  I think I'd go with Hutchinson or Thibodeaux, especially since the main knock on Thibodeaux is that he didn't run a 3 cone.

13 They're the ones that bring square holes.

When they can buy (literally, no salary cap for coaches) whatever shape holes they want. Scheme to your players or you'll never find what you're looking for. 

Kayvon ran a 7.23s 3 cone at his pro day. Still a 9.63 RAS. The only thing he didn't do is the vert. Even if we assumed it was 0(th percentile, like 24.5in, unlikely based on other testing), it'd still be an 8.68, aka in the green. 

15 Scheming to your players is…

Scheming to your players is great in theory, but coaches have schemes they know and work for them, and then there are schemes they do not know and would mess up anyway.  Hutchinson and Thibodeaux have some scheme flexibility, but if a team wants to put Evan Neal in a zone blocking system, it's not going to work out that well, and would be a waste of Neal's skills.   So if Neal is by far the best player available at your pick, perhaps trading down to some team that plays power football is a better idea.

17 When you write "The best do…

When you write "The best do it", you mean Belichick, because nobody else is that flexible.  Even his acolytes discard decent players because they don't fit their system; Eric Mangini changed the Jets to a 3-4 and ended up trading former Defensive Rookie of the Year Jonathan Vilma to New Orleans a couple years later.  Tomlin is probably the only other coach I can think of with the flexibility you are talking about; he never installed the Tampa 2 in Pittsburgh, and the Steelers are all the better for it.  But the Steelers kept Dick LeBeau to run the defense.  Belichick's defensive system is very flexible; unfortunately for many teams Belichick's assistants don't seem to be.

That PFF article you linked to only talks about college schemes; the skill sets of the players is much more important.  Orlando Brown played in an Air Raid offense at Oklahoma, but he fell in the draft due to a terrible combine, which implied he did not have the quickness to play in a zone blocking scheme.  Instead he got drafted by a power blocking team, the Ravens, and worked out great.  

The other issue is that there are 10 other guys on the field.  Should a coach change the system for one guy if it fits everyone else, or perhaps take someone else who does fit your scheme.  If you can trade down and get more picks, then you are a well run team.

18 I dont understand how the OBJ pick fits your argument

It'd fit mine more lol.

Speaking of Baltimore we do know Ronnie Stanley is a weird fit for their running team? 

I wouldn't worry about Neal, if anything this thread strengthens my belief in him. The article talks about how misleading college schemes are. Wirfs completely flipped responsibilities from schemes going from Iowa to TB. You'll have to explain further why Neal would be bad.

And at the end of the day, Travon is not that different from others where scheme should be talked about and be the deciding factor that high. If he's first you're betting a whole lot in your coaching over Georgias.

6 I got 12 to 1 odds KT goes number 2 in draft

He's now the favorite at around two or three to one to go to, team doesn't matter. I'm a huge Lions fan and the money aside I hope they get Hutchinson. 

Detroit general manager Brad Holmes came from Rams and he seems to really like pac-12 players. He took Sewell from Oregon last year in the first round, a defensive tackle in the second round from the pac-12 Washington and his best rookie pick was probably Saint Brown who also played in PAC 12.

The reason I pulled the trigger though is Holmes seems to be all about production and I thought Walker was the one who didn't live up to the other two and but now this article proves otherwise so who knows

 

8 Amaré Barno

That explosion index sure jumps off the chart! Seems worth a 3rd round pick to see if he can be coached up into a top tier rusher.

9 Are college sack numbers…

Are college sack numbers down in general? Given that all the QB seem to be RPO heavy, are schools just going with systems that make it harder for the QB to be sacked?

12 Ezekiel Ansah just started…

Ezekiel Ansah just started playing football in college, so he's not a very comforting example for the idea that athleticism + limited college production is fine. Ansah is like Jimmy Graham - a good NFL player who wasn't that good yet in college because he was just getting started. Whereas Walker was already a touted DL recruiting entering college, and still didn't produce much in college.

Clay Matthews played off-ball linebacker until his last season at USC, when he switched into the "elephant" hybrid role where he mixed it up between rushing the passer & dropping into coverage. So his limited sacks came on a fairly limited number of opportunities. I don't think that's true of Walker, who has a low sack rate & a low pressure rate (per pass rushing snap) along with low sack totals.

14 RE: Karlaftis

The Purdue received a LOT of attention from opposing offenses.  He rarely had a one on one opportunity after his freshman season.  Being a WI person I call it the “Krumrie” effect.  He was a d lineman for the Badgers who after a great sophomore season the following seasons faced double teams.  Still made All Big10 but pro scouts saw fewer splash plays so gave him bad grades. Ended up being a 10th round pick.  Went to several Pro Bowls before breaking his leg in SB against Niners 

19 Example from a Purdue blog

In reply to by big10freak

"This is the perfect distillation of Purdue's pass rush so far this year.  [Karlaftis] is giving 3 guys all they can handle, while the other 3 guys rushing the passer make no progress."

https://www.hammerandrails.com/2021/10/13/22724902/purdue-football-pass-rush-needed-karlaftis-iowa-brohm-lambert

20 Is the Upper Body irrelevant?

Its interesting that bench press and other upper body measures don't seem to fit into the predictive power of your metric. Is that because there is no correlation between upper body strength and NFL pass rush success? Is it all in the leg power?

21 Is the Upper Body irrelevant?

Its interesting that bench press and other upper body measures don't seem to fit into the predictive power of your metric. Is that because there is no correlation between upper body strength and NFL pass rush success? Is it all in the leg power?

22 Usually not much correlation.

Also longer arms are usually an impediment. Mostly about stamina too.

For me, as long as bench percentile is the inverse of arm length (say 75th percentile arm length therefore 25th percentile bench) and/or just matches their weight percentile (assuming it's all muscle) it's fine and shouldn't be worried about much.