Tracking Tom Cruises
Let's just get this out of the way. What I'm about to do here is just about the dumbest thing I've ever done. I'm going to analyze my own analysis. I'm going to go over a bunch of data and then tell you what I think that data means, and that data was produced by.. me. Which means that if I wanted to build a narrative, all I'd have to do is shade the data towards my narrative.
I'll try to explain. Say I didn't like Brad Underwood as the basketball coach. I try not to have any opinions on any of that beyond "he deserves more time to continue to build the program" or "he has been given enough time and failed to build a program" (Underwood is obviously "A" right now), but whatever. If I didn't like Underwood, I could give Luke Goode 1 Tom Cruise and then give Brandon Lieb 0.5 Tom Cruises and then I could say "the last two Underwood recruits averaged 0.75 Cruises does Underwood even have a clue what he's doing?" Change the data, build the narrative, point to the "data" and say "see, this proves I'm right".
So where I'm headed with this post - looking at my own Tom Cruise rankings for football recruits the last five years and using that as a flashing warning sign - could easily be struck down by simply saying "Robert, you're coming up with the ratings, so you're artificially deflating this class in order to make your point". And there's no real way to argue against that except to say that I try to rate each player as fairly as I can. Sometimes, that's taking a 4-star QB like CJ Dixon and giving him 2 Cruises. Sometimes that's taking a 0-star unknown like Sydney Brown and giving him 3.75 Cruises (I bring that one up too much). Most of the time, when I do that, I get accused of building a narrative - "you're just a Lovie lover so you're inflating Brown's rating to make Lovie look good" or, well, just see the first comment last week when I gave Theodore Lockley 1.5 Cruises for the flipside.
I point those out simply to say that I give these ratings my best effort. I'm going to be wrong a fair bit (Dominic Thieman, bumped from 2 Cruises to 3.5 Cruises on Signing Day because he led St. Thomas Aquinas in receiving his senior season and I thought that deserved a big rating bump), but the whole goal behind this was to try to get more accurate ratings for Illini recruits. I got tired of Rivals giving every Illini recruit a 5.5 rating and 247 auto-assigning an 83 for everyone (does anyone even try to watch film or is that just reserved for players going to big-time schools?), so in 2013, I decided I'd do my own ratings, starting halfway through the 2014 recruiting class (meaning Nick Allegretti and Mikey Dudek missed getting Tom Cruise ratings by a few months).
Last year was the first year where every scholarship player on the roster had a rating. So it's the first time I began to play with the data. Someone had sent me a list of all Tom Cruise ratings in a spreadsheet, so I played around with it a little bit and started including the ratings when I did The 90 Illini and such. My feeling the first time I went through the data: I can see why Rivals and 247 simply play it safe with ratings. Hedging is BY FAR the easiest way to do this. There's no payoff to attempting to identify a Bobby Roundtree and giving him an appropriate rating. If he hits, there's no praise. If he misses, you were just boosting the grades to fluff the coach.
That's why no one else does this. It is absolutely lose-lose. It's just so much easier to take a third party rating and say "I don't agree with that - I think he's much better" but assign no value to how much better. Hedge, hedge some more, and then profit.
I went through with it (and will continue doing it) because it still bothers me that no one is willing to paint an accurate picture of Illini recruiting classes. The best way to do it would be to rank all Big Ten classes and see how we compare, but doing 300-some player evaluations every year would likely take me six months. So I'm left to evaluate the Illini recruits, compare them to other Illini recruiting classes, and see if we're boosting our overall talent level or lowering it.
And as the recruiting sites continue to rate classes without including transfers, I feel like school-centric recruiting rankings are even more important. You see all kinds of preseason rankings formulas which include 247's composite class rankings, and it still blows my mind that many of these numbers-centric guys continue to rely on rankings that don't include transfers at all. The rankings come up with a number averaging four recruiting classes, and they plug that number into the system, and that number for the Illini acts like Brandon Peters, Luke Ford, Josh Imatorbhebhe, Milo Eifler, and the other transfers don't exist.
Yeah I can feel that I'm on a tangent now but I need to keep going.
As you know, I swear by Bill Connelly's SP+ ratings. I quote it more than I quote any other data. The NERDstat revolution. I'm a shadow of my former self.
But it still blow my mind that he uses only "class ranking" for this data point. Here's what he says about the recruiting class ranking part of his SP+ formula:
2. Recent recruiting. After determining how much of last year's team is being brought back, the next step is to determine the caliber of the players who will be filling in the missing returning production. To do that, I use a weighted mix of recent recruiting rankings. They primarily pull from the past two recruiting classes, but I have begun to incorporate older classes as well, to a lesser degree. No matter what the "stars don't matter" crowd will tell you, recruiting rankings are extremely predictive, and these projections are more effective because of them.
I totally agree - recruiting rankings are extremely predictive. But with the explosion of the transfer portal - 1,105 players transferred last year - how can the data he enters into his formula come from recruiting class rankings which ignore transfers? The 2019 Illini class was ranked 53rd in the composite rankings. If you add Peters, Ford, Imatorbhebhe, Betiku, Sidney, Petitbon, and Chase Brown, that ranking is going to go up. I did that exercise last month where I re-ranked each player (dropping them to more accurate rankings based on them transferring "down" to Illinois) and then used the class calculator to see where the class would rank. The result for the 2019 class:
2019 original point total and ranking: 184.37, 53rd nationally
2019 point total and ranking once I added the seven transfers to the class (plus Witherspoon): 222.29, 27th nationally
If you missed that post and you're rolling your eyes at that, I urge you to go read it. I stand by my math. I only left Ford and Bhebhe as four-stars, dropped Peters to borderline 3-star/4-star, and dropped everyone else to three star recruits. And even with that, when running the numbers, the class jumps 26 spots in the class rankings. The only real hole there, as noted in the article: it doesn't include the transfers for all of the other schools either, which means the Illinois class ranking wouldn't jump to 27th, it would jump to 29th or 31st or whatever. Nobody brought in eight transfers like Illinois, but others did bring in several. That's the whole point here - every one of those transfers remains un-counted in the class rankings that are used for these predictive preseason rankings. For the 2020 season, it pretends like 15 of the 85 players on the Illinois roster don't exist.
I'm way, WAY off track here. I already ranted about that and here I am ranting again. But I just don't get why transfers continue to be ignored. Seems to me that better numbers could have seen last year's 6-win season coming.
Back to Tom Cruises and the point of this post. A second person sent me a spreadsheet of Tom Cruises, and with apologies to the other guy who sent me the spreadsheet last year, A) he organized it by position and by recruiting class, and B) he attended the Michigan State game with me. So I'm rolling with this spreadsheet from now on.
At some point I might make it public, but for now, I'll just organize all the data here. What can we learn from my OBVIOUSLY BIASED rankings of the recruits over the last five seasons?
First, by class. Please note that this DOES include transfers, unlike, well, you know...
2016 (Cubit) class
Average Tom Cruise rating: 2.304
Top three recruits: Dele Harding 3.5 Tom Cruises, Dom Thieman 3.5, Tymir Oliver/Stanley Green/Zarrian Holcombe 3
Bottom three recruits: Brandon Jones 1 , Eddie Fish/Ayo Shogbonyo/Evan Jones/Griffin Palmer 1.5
Average Tom Cruise rating: 2.730
Top three recruits: Owen Carney 4.25 Tom Cruises, Ricky Smalling 4, Lere Oladipo 4
Bottom three recruits: Five players rated 1.5: Ra'Von Bonner, Jamal Woods, Marc Mondesir, Isaiah Gay, Bennett Williams
Average Tom Cruise rating: 2.563
Top three recruits: Calvin Avery 4.5 Tom Cruises, Verdis Brown 4.25, Sydney Brown 3.75
Bottom three recruits: Four players rated 1.5: Jacob Hollins, Julian Pearl, Carlos Sandy, Edwin Carter
Average Tom Cruise rating: 2.820
Top three recruits: Marquez Beason 5 Tom Cruises, Isaiah Williams 5, Shammond Cooper 4
Bottom three recruits: Five recruits rated 1.5: Casey Washington, Dalevon Campbell, Evan Kirts, Josh Plohr, Tarique Barnes
Average Tom Cruise rating: 2.650
Top three recruits: Reggie Love 4.25 Tom Cruises, Deuce Spann 4, TreSean Smith 4
Bottom three recruits: Chinedu Odeogu 0.5, Desmond Dan 1, Taz Nicholson/Anthony Shipton/Phifer Griffin 1.5
2021 class so far
Average Tom Cruise rating: 2.475
Top three recruits: Brody Wisecarver 3.5, Patrick Bryant 3.25, Snook Edwards/Trevor Moffitt 3
Bottom three recruits: Prince Green/Theodore Lockley 1.5, Samari Collier/Dylan Rosiek 2
Many things to cover here. Perhaps a list of things to remember.
1. Remember, this is me evaluating me.
Please keep that in mind as we go through this. I'm both judge and jury. I should really use third-party player evaluations for this, but, uh, I have some issues with said third-party player evaluations. This is simply one person - an Illinois fan - sitting down and evaluating every player through the same (unquestionably tinted) lens.
But since these number don't relate to anything (Tom Cruises? Really?), well, the only bias we have to worry about is whether I undersold the Cubit class (I don't think I did) or if I'm overselling Lovie's classes (I don't think I am besides maybe not caring enough about the fact that transfers are gone in 1-2 years and we don't have much high school talent built up behind them). If I wanted to oversell a class, 2018 would have been the class to do it, and I don't think I boosted those ratings at all. That was when I first started taking heat for it (very specifically when I rated Daniel Barker with 3 Cruises), but I feel pretty good about how history will tell the story of that spat.
2. Remember, 2021 needs to be his best class.
It will make up 30% of his future rosters. We were in a bowl last season and have momentum to sell. The Smith Performance Center now exists. We've been over this a lot, but it needs to be mentioned again: this is the class. Smallest class in the Big Ten the last two years (so few seniors graduating); likely the largest class in the Big Ten this year.
That, and the fact that the successful rebuilds I've studied (see the 2017 and 2018 IlliniBoard Football Previews) all had one common trait after the inevitable CLICK season: a boost in recruiting based on that click. There's almost always a dip after the first surge of a rebuilt program, and a recruiting hot-streak can help offset that dip.
How are we doing so far? Tom Cruise-wise, it's his worst class. Yes, there's only 10 recruits, and when you add transfers next spring, there might be as many as 30 new players in this class. But so far, when taking the average of all 10 recruits, it's the lowest average rating I've handed out.
Not Great, Bob. Still room to grow (which, yes, means there's still room for me to inflate the next 15 ratings and artificially boost this class), but for now, not only is it not his best class, by my ratings, so far, it's his worst.
3. Remember, the 2020 class was not in any way the "88th-best recruiting class".
We've been over this, but that's what the word "remember" is for. Remember, the composite rankings are not set up to accurately compare a class of 25 with a class of 13. Add eight walkons to those composite rankings - players who would never, ever see the field here, so eight warm bodies - and the class jumps 23 spots in the class rankings.
And there's this little thing where the class rankings don't include transfers. So it acts like Brian Hightower and TreSean Smith don't exist. AND, of those 13 high school players, seven of them (Love, Frenchie, Newton, Davis, Riggins, Spann, and Tyler) have offer lists on par with Iowa and Wisconsin recruits. I'll probably never not be upset with how the 2020 class is viewed.
4. Remember, this is still nowhere close to what we thought Lovie would do.
We can't really forget that. The day Lovie was hired, I really thought we were walking into a Mack Brown-at-UNC situation. His name would dominate Chicago (and Texas, and Florida) and we'd hop right to Zook-like classes again. Maybe even better. Somewhere, another John L. Smith will appear to accuse us of cheating.
It hasn't been that. At all. It's been an eval-heavy, we-like-our-guys, we-don't-need-in-state-players transfer-fest. The staff completely ignored the Beckman players (and most of the Cubit class) and started over like they were rebuilding a team after the death penalty. Recruiting was like Jerry Kill, not Mack Brown, and "we'll find our guys and build our kind of program" became the thing.
I feel like I need to make that point because most all of us started there. "We hired Lovie Smith? He's going to own recruiting with his NFL name." Nope. Herm Edwards has done more with his NFL name than Lovie. I mean, Jeff Brohm has done more with his XFL name than Lovie. It just hasn't been that kind of rebuild at all.
5. Remember, that's still maybe OK?
I've written about this several times recently, but great classes don't mean successful rebuilds and bad classes don't mean failed ones. Texas fans have been celebrating their great classes for 10 years and it hasn't gone anywhere. Nebraska had TWENTY four-stars on their roster last year and couldn't get to a bowl in year two of a hotshot coach. Florida State? 41 four-stars and five stars and they've gone 18-19 the last three seasons. Chip Kelly has had 20+ four-stars both seasons and went 7-17. Average recruit ranking at USC? 90.42 (AVERAGE recruit is a four-star, equal to Georgia and LSU). Average recruit at Utah? 84.44 (equal to Colorado and Vandy).
Yes, recruiting rankings generally hold. The playoff will probably be Alabama, Clemson, Ohio State, and Oklahoma simply because of recruiting. When you're trying to maintain your spot in college football, you have to recruit.
But when building a program, it doesn't require high-end recruiting. It certainly helps, but even high-end rebuilds with lots of four-stars will flop. You can get there just with a bunch of experience. That was basically the 2019 IlliniBoard football preview: 89 of the last 96 P5 coaches hired got to a bowl game. It's pretty easy to do. Play the kids, once you get them to 20+ starts in your system, they'll win enough to get you to a bowl. It's really hard to not win at least six games by your fourth year.
It's what you do after that. You brought the program up, you've built a culture, and your players know the expectations. As I wrote a few weeks ago, Lovie has improved the roster at (in my estimation), 7 of the 10 positions. Some positions (like cornerback) have gone from a 3.5 out of 10 to an 8.5. It's better than it was when Bill Cubit handed him the keys. And we just went to a bowl.
So the very next step is to level-up again. Improve this roster at 7 of the 10 positions. I think there's enough talent + experience to win 8+ games this season, but then those players graduate and it's time to rebuild the rebuild (something Zook nor Turner could do). The biggest step there? Turning these wins into better recruits.
So far, Tom Cruise-wise at least, it doesn't look like it's happening.