Jimmystats: The Best Recruits Play

Submitted by Seth on February 3rd, 2015 at 10:45 AM

13655362573_ce1d0c81c8_z

Taco-ranked starters are far more likely than Glasgows [Fuller]

Every year, as college football recruiting becomes the only football thing left to pay attention to until spring, we are suddenly struck by an army of pundits so arrogantly attached to their "recruiting stars don't matter" narratives that they don't bother to care that math is against them.

Michigan typically gets taken to the woodshed in these articles for recurrently not matching recruiting expectations with on-field results. This discrepancy does exist beyond the normal J.T. Turners that everybody gets, and for various interrelated reasons: attrition spikes, spottily shoddy coaching, program instability, recruiting shortfalls. Anecdotally, there are examples I can point to, especially in the early aughts, when an otherwise two-star athlete was bumped to a three-star because Michigan offered. That explains less about how Wisconsin and Michigan State thrive on 2- and 3-stars, and more about how Michigan has recruited very few guys under a consensus 3-star.

However every time we find a new way to compare recruiting data to performance data, we consistently discover that recruiting stars handed out by the services correlate to better players. No, a 5-star isn't an instant superstar, but the 25-30 five-stars each season are consistently found to be about twice as likely to meet some performance metric (NFL draft, All-conference, team success, etc.) as the pool of 200-odd four-stars, who are consistently more likely to meet performance thresholds of the 400-odd three-stars, etc.

Today I present a new metric for proving it: starts.

rawdata
Example of raw data, via UM Bentley Library.

ALL the Starts

My project over Christmas was to take the data from Bentley's team pages (example at right), scrub the hell out of it, and produce a database of who started what years, at what positions, at what age, with what recruiting hype, etc.

HERE IS THE SPREADSHEET.

A few weeks back I released the initial results of my starts data. We noticed there were a lot of problems in that. I went back and did a lot of fixing, mostly just finding more weird errors in the Bentley pages I'd culled the data from, sometimes emailing the guys themselves to ask things like "Was there a game in 2001 that either you or B.J. didn't start?"

I think I've got it cleaned up now; at least the total number of starts for each season matches 22 players per game.

Recruiting By Starts

Starting in 1996 we start getting relatively uniform star rankings for recruits, though I had to translate Lemming rankings and such into stars (he had position rankings and national lists that line up with what we call recruits today). So I took the average of available star ratings of all players to appear on Michigan's Bentley rosters from the Class of 1996 through the Class of 2010, and put 'em against the number of starts generated. Guess what: recruiting actually matters.

Recruit Level 1996-2011
Recruits
1996-2014
Starts
Starts/
Player
5-stars 21 450 21.4
4.5-stars 28 462 16.5
4-stars 82 1215 14.8
3.5-stars 76 881 11.6
3-stars 82 669 8.2
2- or 2.5-stars 29 271 9.3
Walk-ons 217 97 0.4

Even with Michigan's notorious luck, the 5-stars were expected to give you about two seasons of starts, compared to the 8 or 9 games you'll get out of a 2- or 3-star. That is significant, and offers a bit more evidence toward the general statement about recruiting stars: the higher the star rating, the more likely he is to be a good college football player, though at best you're at 50-50.

As for walk-ons, I've linked to the list of the 217 guys in that time period who made the Bentley rosters and weren't special teamers, in case you doubt me. The Order of St. Kovacs have accomplished great things for Michigan, but turning up one of those guys anywhere other than fullback has been rare indeed.

Recruiting classessince1996

Best Classes

I'm going to try to use the starts data above to get predictive. The scatter plot of the 1996-2010 group was pretty linear so I'm just going to plug in a linear equation:

Expected Starts on Avg M Team = Stars x 5.30 - 6.35

And that gives us a reasonable expectation of Michigan starts to expect from a class based on their rankings:

Starts by Class 2vs Expectations by Star Ratings

click big makes

For the Class of 2011-2014 projections, I just guessed by hand, so those projections are going to be increasingly inaccurate once I'm predicting 2017 starters and whatnot.

The chart above has two stories to tell: 1) The strength of a recruiting class is strongly correlated to the value that class will produce in starters, and 2) the damage done by attrition to the 2005 and 2010 classes created ripple effects for several classes afterwards.

An Average Michigan Team:

By some quick averages I was able to get an average makeup of a starting 22. I took the average number of starts by experience (i.e. year in the program) for the classes of 1995-2010, adjusted those numbers for a 13-game schedule, then divided by 13 games to get an idea of what the starters ought to be against years of interest.

Experience Average 2014 2013 2012 2011 2008 1999 1997
5th Years 5 3 5 7 4 4 6 3
Senior / RS Jr 7 5 4 8 8 6 8 9
Junior / RS So 6 10 5 4 7 7 5 6
Soph / RS Fr 3 3 6 2 1 2 3 4
True Frosh 1 1 2 1 2 3 0 0
AVG starter age 3.55 3.27 3.18 3.82 3.50 3.27 3.77 3.50

By this the last two teams look extraordinarily young—about as young as the 2008 team or younger. The 2012 team by contrast seems like a wasted opportunity. FWIW I counted Devin, not Denard, as the quarterback, or it would have been even older. That fits the narrative: 2012 was a wasted opportunity, as a line with three 5th year seniors (two of whom were long-term productive starters) plus Lewan and Schofield was coached into one of the worst offensive lines in memory.

Comments

Seth

February 3rd, 2015 at 11:12 AM ^

That is simply a function of very few 5-star recruits, and that the star rankings are a measurement of likelihood of collegiate success, not pro potential.

Figure if half (a VERY high number compared to other star ratings) of all 5-star recruits make it to the NFL, and a quarter of them spend at least 3 years starting in the NFL (also a VERY high number). That means you get about 4 NFL starters out of the 5-star recruits per year. Over 10 years then you'll expect to accumulate about 40 NFL starters from the 5-star ranks. So each NFL team should have, on average, 1.25 former 5-star recruits. Given that distribution, it's no surprise at all that any two teams selected at random would have zero between them.

Indonacious

February 3rd, 2015 at 11:34 AM ^

One could actually make the argument that the 5* players who pan out enough to get drafted will be high first round draft picks (due to athletic gifts, visibility), and unlikely to end up on successful teams such as patriots and Seahawks who draft at the bottom of the first round.

Chitown Kev

February 3rd, 2015 at 11:04 AM ^

the Top 3 teams in the country and 4 top 10 teams. Still we played all of them but Alabama close. so with better coaching that's probably another 11-2 season in 2012. And Coach Harbaugh may feel as if he landed in San Francisco again. I wouldn't be surprised if we went 11-2 in 2015.

ak47

February 3rd, 2015 at 11:30 AM ^

Yeah 2012 was less of a wasted opportunity and more not having an elite coach necessary to navigate that schedule. The teams who have play 4 top 10 teams in one year and come away with less than 3 losses tend to be playing for national championships.  We got wrecked by Bama like pretty much everyone. Lost 13-6 at #2 Notre Dame who only lost to Bama in probably denards worst game ever, lost at Nebraska in the bellomy game which is also the worst loss of the season and then by less than a td at an undefeated osu and a top 10 south carolina in the bowl.

If we play anything other than the hardest schedule in the country that year we most certainly finish with 10+ wins.  That was more bad timing than a missed opportunity. 

Chitown Kev

February 3rd, 2015 at 11:58 AM ^

(for us old folks) a Penn State 1982 caliber. Michigan plays schedules like that from time to time (1991) and I prefer that we do that. Sure we might lose to a FSU (1991) or to an Alabama but if we win more of those type of schedules when we lose we probably get another 4- or 5-star player...or 2 or 3.

Gandalf the White

February 3rd, 2015 at 11:22 AM ^

beat others with his own recruits, regardless of their high school star rating.  Why, he could then turn around and beat his own team with the others' recruits, regardless of their high school star rating.

 

 

 

i hope...

DonAZ

February 3rd, 2015 at 11:40 AM ^

This is good stuff, Seth ... thanks for crunching the numbers!

I wonder ... is there a position or a position group where stars seem to be more predictive of performance?

DonAZ

February 3rd, 2015 at 12:20 PM ^

I'm going to guess ...

  • QBs are scrutinized a bit more and a 5-star QB must be really good,
  • OL has a tough time getting 5 stars because everyone knows they take time to mature,
  • RBs probably get 5-stars easier than other offensive positions
  • On the defensive side of the ball athleticism is key, and that can be seen more readily in a HS recruit, and
  • DBs get more earned 5-stars because their athleticism at that position is likely fairly evident in HS

I have nothing to go on in saying any of that ... just a hunch.

trueblueintexas

February 3rd, 2015 at 2:07 PM ^

I thought D-lineman were the most likely to be 5-stars. I checked the top 10 ranked recruits on 247 going back to 2010 and it appears this is true (at least at the very top of the rankings). 

2015 - 5

14 - 2

13 - 4

12 - 4

11 - 3

10 - 4

That is 36.6% of the top 10 being made up of D-lineman, with 4 of those 6 years being 40% or better.  

I included DT, SDE, and WDE. As I was scanning the top 10 looking for D-lineman.

OLB &S also seemed to surface quite a bit, although I did not actually count them. That would lead me to agree that stars are slanted towards defense.

1974

February 3rd, 2015 at 12:10 PM ^

I really enjoyed reading this. I have a thickheaded former colleague (of the STARZ DON'T MATTER variety) who is going to be getting a link very soon.

As much as it pains me to say so, this evidence makes Sparty's talent evaluation (or whatever it is they're doing up there) seem more impressive. They've sent a lots of 3s and 2s to the NFL. (By the way, the colleague I mentioned earlier also believes that MSU's success is all about Dantonio "taking over the state.")

M-GO-Beek

February 3rd, 2015 at 12:14 PM ^

Seth, thanks for the info, very interesting.

2 thoughts:

1) Is it possible that at different programs there would be different results suggesting rankings are less important.  For example, at Alabama under Saban where 5 stars are more of the norm than at other programs, would this still hold up since presumbaly the competition would be greater from one 5 star vs. another, etc.

2) More important to the predictive modeling, does a coaches ability to develop talent alter the relative value of a recruits ranking?  Obvious example here would be to look at Harbaugh's Stanford teams and see if higher stars still means more playing time, or is he more adept at finding the diamond in the rough and turning the player into something useful.

Seth

February 3rd, 2015 at 3:52 PM ^

The thing about using starts is it's dependent on the level of competition, since somebody is going to use those starts regardless. Ferrara earned starts on Michigans offensive line in 2008; he wouldn't again.

At Alabam I think we would get a very good sample using starts since they've been playing at a consistently high level for a long time now.

M-GO-Beek

February 3rd, 2015 at 12:14 PM ^

Seth, thanks for the info, very interesting.

2 thoughts:

1) Is it possible that at different programs there would be different results suggesting rankings are less important.  For example, at Alabama under Saban where 5 stars are more of the norm than at other programs, would this still hold up since presumbaly the competition would be greater from one 5 star vs. another, etc.

2) More important to the predictive modeling, does a coaches ability to develop talent alter the relative value of a recruits ranking?  Obvious example here would be to look at Harbaugh's Stanford teams and see if higher stars still means more playing time, or is he more adept at finding the diamond in the rough and turning the player into something useful.

Wolverinefan84

February 3rd, 2015 at 1:40 PM ^

I've always wondered how much stars truly mattered in terms of actual starts at the college level. Many of my close friends go to MSU and I have to restrain from banging my head against a wall during the "stars don't matter" arguments where we Michigan fans are constantly reminded of the Will Campbells of the past. I've always countered saying we just had 2 coaching staffs that either 1.) unsuccessfully tried to change Michigan football into the opposite of how it played the past 100 years, or 2.) painfully underdeveloped high-caliber recruits and couldn't win against quality competition/coaches. Also include that neither of our prior 2 coaches were able to last long enough for their first recruiting classes to get to upperclassmen level. However now, like most Michigan fans, I'm cautiously optimistic that we finally have a coach that will stay at least 7-8+ years so we might finally see our recruiting classes live up to their potential. Again, great research Seth. I like the way this program is headed!

saveferris

February 3rd, 2015 at 2:41 PM ^

Many of my close friends go to MSU and I have to restrain from banging my head against a wall during the "stars don't matter" arguments where we Michigan fans are constantly reminded of the Will Campbells of the past.

Remind your MSU friends of the following fact with regards to the 4 teams represented in this years college football playoff:

Rivals Rankings from 2011-2014:
Ohio State - 11, 4, 2, 3 (Avg: 5)
Oregon - 9, 16, 22, 26 (Avg: 18.25)
Florida State - 2, 6, 10, 4 (Avg: 5.5)
Alabama - 1, 1, 1, 1 (Avg: 1)

Stars matter, but you may have to help them with the math to get your point across.

white_pony_rocks

February 3rd, 2015 at 2:37 PM ^

I know that using percentages to compare groups is usually the way to go, but maybe in this case it isn't. let's make the assumption before hand that our coaching staff is the very best at talent evaluation, then if we look at the counts of 5 star players who contribute vs. the 2-3 start players who contribute, we probably see that if you can are the absolute best at identifying talent then you don't have to worry so much about having multiple 5 stars per class because there are enough underrated 2-3 stars to make up a damn good team, plus the 4 stars you'll inevitably get some of

UMFootballCrazy

February 3rd, 2015 at 4:22 PM ^

Moderately interesting...nothing truly predictive, though. It seems that would require you to compare the average star rating of all your starters and how experienced they are with all the other teams and where those teams finished in terms of end of season rankings. This would give you an expectation of where you should finish and if you under or over perform your ranking.

Seth

February 4th, 2015 at 6:12 PM ^

Great catch; thank you. I should have known that since he's from Troy and we played against them. He was a really tall lanky guy in high school--they listed him at 6'4/245 as a recruit, but was up to 260 by the time he got to Michigan.

He doesn't change the stats much; he had two starts at Michigan. Strangely he's not in the 2000 team photo or listed on the 2000 team, though my 2000 edition of The Wolverine preview includes him. Do you know what happened to him?

Thanks again. 200 points to you.