Here are the 247 recruiting rankings-- the most used and relied upon ones (team rankings, including both recruits and transfers-- which is super important given the era we live in in which transfers are a non-trivial part of virtually EVERYONE's plans): https://247sports.com/Season/2023-Football/CompositeTeamRankings/
1. The numbers are not based on an average of ratings. They are based on total points. Thus Texas is 3, but that is largely because they brought in more players. Ohio State is number three based on average recruit quality, but number 5 based on the total points method.
2). Those who say "recruiting rankings have been proven to be accurate based on NFL outcomes" are right but only kind of right. It's true that when a service rates a kid as a 5-star player they are pretty regularly right-- about 61% get drafted. But a) that's also pretty prone to error-- wrong 39% of the time--given that b) they only rate 32 kids five star a year and it's easier to be on average right when you only have to be right a few times. 4 star player accuracy drops a ton-- 23% of four-star kids get drafted. 5.1% of 3 star kids get drafted. Under 3 stars only .51% get drafted.
All of this is well and good. But recruiting *rankings* are not necessarily showing you things on a relevant scale and don't map straightforwardly onto star ratings. No study I'm aware of has or could show a meaningful difference between outcomes of a recruit rated 91.77 and one rated 91.03, or even 90.36. All are 4 star ratings. Yet 91.77 is the average rating of LSU's class-- ranked #4. 91.03 is the average rating of Florida's class, ranked #13. 90.36 is the average rating of Florida State's class, ranked 16. Is there a meaningful difference between the recruiting classes of these schools? Probably not, realistically. Nothing you are going to really see on the field, or be able to separate from the impacts of coaching, nutrition, strength and conditioning, playcalling, momentum/mindset, etc., etc.
As you go down the ratings, the bands get bigger and less tied to the end team or "class" rankings, although the early bands are pretty competitive since there are not too many five star recruits. To give you an idea, just create a series of recruiting bands in which the average player rating is within 1.
Here's what it looks like:
94+ = #1 Alabama
93-94 = #2 Georgia & #5 Ohio State
92-93 = #3 Texas
91-92= #4 LSU, #6 Oklahoma, #7 Miami, #8 USC, #11 ND, #12 A&M, #13 Florida
90-91= #9 Oregon, #10 Tennessee, #14 Penn State, #15 Clemson, #16 Florida State
89-90= #17 South Carolina, #19 Michigan, #18 Auburn, #20 Ole Miss
88-89= #22 TCU, #24 UCLA, #25 Utah, #26 Arkansas, #27 Kentucky, #30 Washington, #31 Louisville
87-88= #21 Colorado, #23 Michigan State, #28 UNC, #29 Nebraska, #32 Maryland, #33 Mississippi State, #34 Baylor, #35 Texas Tech, #37 Wisconsin, #39 Virginia Tech, #44 Missouri, #45 NC State, #46 Minnesota, #50 Pitt, #55 UCF, #59 Stanford
86-87 = #36 Az State, #38 Kansas State, #40 Indiana, #41 West Virginia, #42 SMU, #43 Ok. State, #47 Iowa, #48 Illinois, #49 Arizona, #51 Oregon State, #52 Iowa State, #53 BYU, #54 Georgia Tech, #56 Northwestern, #57 Cal, #60 Boston College, #62 Wake Forest, #64 Houston, #65 Cincinnati, #66 Kansas, #67 Purdue
The difference in average recruit ranking between #1 Alabama at 94.69 and #2 Georgia at 93.92 (.77) is 3x as big as the difference between #21 Colorado (87.96) and #55 UCF (87.72) (difference = .24).
Recruiting is important and it matters a ton. But recruiting measured solely in a forced ranking system using total points, not average player rating, is misleading. It treats differences that are not research-validated (the difference between two recruits, both rated four stars, for example) and treats them as though they are quite meaningful. Even the people that are making the ratings don't claim that their "ratings" are accurate to two decimal places. That's absolutely absurd. Even within 1-1.5 in terms of rating, most of them don't claim to be discerning anything in particular, and there is no validation for those differences, aside from certain thresholds related to star rankings: 87.02 versus 87.99 are essentially the same player, with the same basic chance of being excellent, as are 87.99 and 88.99, but the recruiting rankings treat them very very differently.