clock menu more-arrow no yes

Filed under:

Roundtable: Unifying the un-unifiable in recruiting

New, 25 comments

What makes an impact when — and how should we look at ratings that are measuring different things for different prospects?

COLLEGE FOOTBALL: OCT 16 Washington Practice Photo by Jeff Halstead/Icon Sportswire via Getty Images

With little to go on for practice reports other than players’ and coaches’ thoughts to the media (and the occasional message board rumor), Dawg fans don’t have the information out there that usually becomes known over the course of one of Washington’s fall camps.

But one thing’s been brought up frequently anyway: the Huskies’ true freshman receivers are, apparently, good at football — and Rome Odunze in particular has appeared to make a splash early. Then, with Christian Caple’s piece about Odunze out in The Athletic today, it prompted some of the UWDP writers to have an impromptu discussion on stuff. Which turned into the following...

Enjoy!


Max Vrooman: Caple’s piece on Odunze today talks to his track coach who is convinced he would’ve set the Nevada state 200 meter if they’d been allowed to compete this past spring. He was close as a junior.

His junior year times in the 100 and 200 were about 3-5% off of the Pac-12 all-time records.

Andrew Berg: In a situation like that, how do you suppose someone like Rome gets underrated coming out of high school? He was #227 nationally and the 40th WR. The second he steps on campus, he leapfrogs McMillan, who was 67/11. They both had high-level competition. They’re both big and fast. Why was the consensus that McMillan was clearly better, but when they lineup next to each other, Rome is immediately better?

Maybe this is an argument for the offer list being a better predictor than the consensus rating. Despite being outside the top 200, Rome had offers from EVERYBODY — Alabama, Oklahoma, Florida, Georgia, Ohio State...

None of this is to knock McMillan. I think he’s going to be excellent.

Max: I would say 1) The margins at that level aren’t all that big between #50 and #200. 2 ) 10 months of development can be a long time at that age (especially if Rome handled training better with the pandemic). 3) It could be Rome is a better fit for what UW needs from them right this second and in another offense McMillan would be ahead on the depth chart and 4) It could be as simple as Rome is quicker to learn a playbook.

The difference in their offer sheets. Only offered Rome: Stanford, Texas, BYU, San Diego State, Texas Tech, Michigan, Miami, Penn State, Duke, Michigan St, NC State, Minnesota, West Virginia, Northwestern, Kansas.

Only offered Jalen: Wazzu, Oregon St, LSU, Fresno St, Ole Miss, Kansas St.

Offered both: Everyone else in P12, Tennessee, Oklahoma, Bama, Notre Dame.

Gabey Lucas: Also it’s worth noting — and I tried to emphasize this in my impact true frosh piece — that the methodology for ratings don’t tell you a huge piece of it. And a huge part of that is “How good will/could you be in college?” vs “How high is your NFL potential?” vs “How good are you now?”

The 247 guys like to say they’re evaluating based on NFL potential or whatever the fuck, but A) that’s not as useful for understanding, ya know, college potential (although obviously they’re related) and B) when you actually break this down on a case by case basis, you can see that’s not consistently what they’re doing.

That’s why I think a system that ranks floor vs ceiling would be much more helpful — cuz part of floor is your physical readiness to play ASAP, which was huge when looking at Rome vs Jalen.

Like, their skillsets are both super high and Jalen’s is likely higher in the long run hence his higher rating, but even just looking at them you knew Rome was closer to contributing.

Andrew: That’s interesting because so much of McMillan’s praise was for things like route running and technique, which are seen to be “ready to play” characteristics.

But those are all good points, and the part about what the ratings are actually rating is something I definitely overlooked.

Gabey: Totally.

But there’s all those which are big in readiness to play, and then there’s just are you physically ready? You can tell in McMillan’s case, gaining 5 to 10 pounds would be significant. But the factors that matter obviously differ by player.

Andrew: These types of case studies are fascinating to me. I love when a player who wasn’t rated as super elite immediately breaks out at a major program. Wasn’t Justin jefferson similar? Like a high 3* recruit who was awesome right away? Maybe I’m thinking of someone else.

Jeff Gorman: I think recruiting people love speed. Rome is obviously very fast but McMillan I think has that extra gear that few possess. I think at every skill position they drool over it because it’s impossible to stop a receiver like that if they do the other things right too. A la John Ross.

Max: Justin Jefferson had a 0.7982 composite rating. That’s essentially a walk-on for 90% of power programs.

Gabey: Shit dawg.

Max: Aaron Fuller is our lowest rated receiver of the Chris Petersen era at 0.8311.

Jeff: Maybe scout if you’d like a Jordan Chin.... only has speed, but would be really good if a bunch of other things fall into place. I mean maybe scouts saw JJ like Chin.

Max: Sorry, missed Nik Little who was in the 0.78 range.

Jeff: And JJ obviously panned out. But I don’t know. I just think recruiting people over-emphasize speed.

Max: Husky WRs from highest to lowest composite rating since 2014: Spiker, McMillan, Nacua, Osborne, Odunze, Bynum, Jones, Renfro, Culp (listed as WR in HS), Cook, T Davis, Baccellia, Pettis, Chin, Dickey/Lenius, Fuller, Little.

And the top-7 names on that list are essentially our depth chart with Jordan Chin up there possibly just because of the injuries to Spiker/Nacua.

Kirk DeGrasse: This is all instructive in remembering that while stars are important in a macro sense, don’t get TOO caught up in them in a micro sense.

Gabey: I think it’s less that, Kirk — although tangentially that can be the case at times — and more that if you look at the ratings for these guys, they’re not all being measured the same way.

That if you’re looking at ratings without any of the context surrounding what they’re measuring in each individual case, you’re not really understanding what you’re looking at.

Kirk: But the end result being Jalen was rated higher, and at the moment at least it looks like Rome is more ready to contribute.

Gabey: The one thing I do think is helpful about 247 is that, below their player profiles, they sometimes do include a more in-depth rating of each trait, but their methodology of what’s being measured is inconsistent.

Kirk: I guess my issue is that too many people look at ratings from 247, etc. and think of it like a Madden video game, like this rating actually is a 100% accurate representation of their ability. I think we all know that’s not true, there’s a lot of margin for error, but the extent of that range of possible error is greater than a lot of recruiting addicts will admit.

Gabey: Yeah that first part I agree with completely. And I think that it comes from the fact that since there’s such a diversity of skillsets, one player’s rating could be based on a bunch of different traits than another player’s rating, but then the guys evaluating them have to come up with a way of unifying those ratings which are based on different things.

I think that’s the most concise way I can put my thoughts I suppose... Just the amount of variables involved means you can’t control for most things, so it’s really not scientific.

Kirk: And to be clear, I’m not blaming the evaluators at 247 and the rest. It’s a damn hard job what they’re doing. When you look at how often NFL teams make bad draft picks and how many guys drafted in low rounds — or not drafted at all — carve out big careers, it’s clear that this kind of talent projection is tough.

My use case that drives me crazy is rando fan screaming “Why are we going so hard after X who’s only an 87.5 when there’s still Y out there who is a 90.3?!?” Well, you have to give the coaches some benefit of the doubt that they know what they’re looking for and what works in their system and their culture.

I try not to sweat smaller ratings differences.

Gabey: Totally same, I’m not trying to shit on them at all. The whole concept of what they’re trying to do makes it pretty much impossible to unify. Like take Gary Bryant vs Rome Odunze for example (just two guys from last year that popped into my head) — it’d be next to impossible to accurately measure both of them on the same scale really, because even though they’re the same position, what are you measuring in both of them?

Kirk: Exactly. They play different positions when it comes down to it. You’re not going to use them the same way.

Gabey: I’m not as familiar with Bryant as Rome, obviously, but with the former you’re evaluating him on nuts-o speed and whatever other traits he has, whereas with Rome you’re looking at route-running, physicality, speed to size ratio... And then they’re expected to rate both of those as a unified metric.

Kirk: You’d think the rating would be a distillation of how effective the services think that player will be, but as you point out 247 has recently adjusted their system to try to predict NFL success. Or more specifically, they want their stars to line up as close as possible with draft results.

Gabey: Point being, when you look at what traits were being evaluated for Rome and Nacua vs, say, Spiker or Osborne, you’re looking at traits that are more predisposed towards being able to make an impact early — even if a guy like Spiker (or maybe McMillan, although we don’t know how much he’ll have an early impact), might have had traits being evaluated that potentially could be even more impactful later.

Kirk: Right. And of course we’re getting waaaaaaay ahead of ourselves if we’re assuming the early tidbits we’ve heard about the receiver room mean that Rome will have a bigger impact this season than Jalen. He very well could, but it’s still early, and plenty of time for things to change.

Gabey: Lol totally. Just that looking at them, the reports mesh with what make sense.


Anyhoo, that’s all. Now, back to speculating about the quarterback competition and all the other stuff we don’t know about this season. Hooray!

Do good things, don’t do bad things, and bow down to Washington.