Thursday, January 12, 2006

The Recruiting Lowdown

I started writing this post mid-way through the college football season and then I realized that it would probably be a better read if it included the entire season so I put it on the shelf for a while. It’s nothing revolutionary but it’s something worth looking at. Hopefully it’s an enjoyable read for you. My next post will likely be Monday or Tuesday. I am headed to London for the weekend where I will not be able to watch what will surely be an exciting weekend of football. My guess is that we’ll see the Colts vs. Pats and Seahawks vs. Panthers in the Conference Championships although, nothing would surprise me. I’ll keep you all in mind while I’m eating “fish and chips” and trying to “get left”.

The college football recruiting rankings are not 100% accurate by any means. With hundreds of thousands of high school football players, it would be nearly impossible to precisely measure which players are better than other players. However, most recruiting services are in agreement when it comes to which players are in the top 500 to 1,000 in any given year. Likewise, most recruiting services are in agreement in terms of “star” rankings, or how many stars a certain player is worth on a five star system. Sure, one service might give a player four stars while another service gives that same player five stars. However, when it comes to how many stars a certain player is worth, most services are in agreement. It gets very difficult when trying to rank these players in order in a top 100 or top 150-list. But, it’s not nearly as difficult to group these players into similar skill levels like in the star-system. Recruiting services rate players based on factors such as; statistics, physical attributes, performance against other top recruits, camp performances, scholarship offers, and film study. While no service will be able to perfectly rate players, I think these services give a pretty fair indication of which players are the best and which players are closer to average.

One of the most disappointing aspects of the Michigan football program is the consistent underachievement on the field compared to the caliber of recruiting class. Most Michigan fans have uttered the phrase, “Michigan should be competing for National Championships every year with the top five recruiting classes that they get every year.” Just like there’s a group out there that still finds underachievement to be soothing, there’s also a group out there that thinks that Michigan doesn’t get a top five recruiting class every year. So, I looked it up. The most notable recruiting service and the easiest to look up happens to be the Rivals recruiting service. Rivals has sortable records that date back to the 2002 recruiting classes. I would’ve liked to have had the 2001 recruiting numbers but Rivals doesn’t list them. Even still, all current players are covered with the exception of 5th-year seniors and Ohio St.’s kicker. The amount of 5th year seniors per team varies but they do not make up a great percentage of the players on a particular team. These rankings also include players that are not on the team for one reason or another. Since the number of players that have left any given football program probably isn’t a significant amount, I don’t think it makes a big difference over a four year period. Here are the top 20 college football programs in terms of recruiting over the last four years. I ranked these teams in terms of average rating per player. Over four years, a college team pretty much fills their entire roster so it makes sense to rank these teams based on average ranking per player rather than total points.


1). Texas------3.679012346
2). Oklahoma--3.677419355
3). Florida St.--3.652173913
4). Miami------3.612903226
5). Michigan--- 3.590361446
6). USC-------- 3.582417582
7). Tennessee-- 3.556701031
8). LSU-------- 3.548387097
9). Ohio St.-----3.469879518
10). Georgia----3.462365591
11). Florida-----3.455555556
12). Notre Dame3.185714286
13). Penn St.----3.155844156
14). Texas A&M-3.128712871
15). Virginia-----3.10989011
16). UCLA-------3.076086957
17). Maryland--- 3.02247191
18). Alabama----3.020408163
19). Auburn------2.990825688
20). B.C.---------2.987012987


Michigan is ranked #5 on the list of recruiting rankings over the last four years. Only, four teams are better. It is clear that Michigan is superior than 95% of all D-1 college football teams in terms of recruiting and caliber of athlete that makes up its current roster. Not surprisingly, Texas is number one on the list. But, just to emphasize how close Michigan is from being on the top of this list, here is the breakdown of recruits and total recruit points for Texas and Michigan over this time:

Texas # of recruits=81 total points= 298

Michigan # of recruits=83 total points=298

Clearly, these teams are essentially even in terms of recruiting strength. Two recruits separate these teams on a roster of 80+ players. Michigan doesn’t have a top five recruiting class every single season. If you’re judging which teams have top five recruiting classes by looking to see who finishes in the top five every season, then nobody has a top five recruiting class every year. However, if you’re comparing teams over time, which is vastly more important than single year rankings, then Michigan has a top five recruiting class year in and year out. Since Michigan doesn’t play Texas, Oklahoma, Miami and Florida St. on a regular basis, Michigan has the talent advantage over virtually every team it plays. Ohio St. has a very good average at 3.47 so I would never fault Michigan for losing to the Buckeyes based on recruiting. On average, Michigan brings in a more prolific recruiting class than Ohio St., but the differential is pretty small.

I don’t know what to deem a “significant” difference in recruiting rating but it’s important to remember that a team with a 3.50 rating has a .50 advantage over a team with a 3.00 rating at every position on the field. For instance, Michigan has a rating of 3.59 and Minnesota has a rating of 2.44. This means that on average, Michigan has an advantage of 1.15 at every single position and every single roster spot. That’s above and beyond a significant advantage. I would say that a differential of .33 or more would be a fairly significant difference in recruiting ranking. There’s really no right answer but just remember that whatever the difference is will be the difference at every position. A difference of .33 might not seem that much but that means that every player on Team A has a .33 advantage over every player on Team B. That seems significant to me.

If all teams were coached by the exact same coaching staff, one could reasonably expect the team with the advantage in recruiting to beat the teams that don’t have the advantage in recruiting. Recruiting rankings are not so precise as to guarantee that a team with a rating of 3.59 will always beat a team with a rating of 3.50 even with the exact same coaching staff. Rivals could have an error of at least +/- .05 per player in terms of that players actual ability. If that’s the case, then a team with a 3.59 rating could be even with a team with a 3.50 rating and still be within a reasonable error. So, recruiting ratings aren’t the end all in terms of how good a college football team will be. Having said that, any time there is a significant difference in recruiting rankings, and the coaching staffs are the exact same, then the team with the significant advantage in recruiting should be the winner. Even with the same coaching staffs, I suppose it is possible for a lesser team to beat a better team due to injury, home-field advantage, breaking in of a new quarterback or a significant mismatch at one or two positions. However, barring injury, and the coaching staffs are the exact same, the team with a significant advantage in recruiting average over four years should win. As we all know, coaching staffs are not the exact same. Some are better than others. In order for a team with a significantly weaker recruiting class to beat a team with a significantly higher recruiting class, they will likely have to have the advantage in coaching.

Here is a list of top 20 teams on this list and the level of recruiting classes that they’ve lost to this season (2005-6):

1). Texas (3.68)

Losses= (0)

2). Oklahoma (3.68)

Losses= (4) Texas (3.68), UCLA (3.08), Texas Tech (2.60), TCU (2.39)

3). Florida St. (3.65)

Losses= (5) Florida (3.46), Penn St. (3.16), Virginia (3.11), NC State (2.92), Clemson (2.81)

4). Miami (3.61)

Losses= (3) Florida St. (3.65), LSU (3.55), Ga. Tech (2.51)

5). Michigan (3.59)

Losses= (5) Ohio St. (3.47), Notre Dame (3.19), Nebraska (2.98), Minnesota (2.44), Wisconsin (2.63)

6). USC (3.58)

Losses= (1) Texas (3.68)

7). Tennessee (3.56)

Losses= (6) Georgia (3.46), Florida (3.46), Notre Dame (3.19), Alabama (3.02), S. Carolina (2.96), Vanderbilt (2.21)

8). LSU (3.55)

Losses= (2) Tennessee (3.56), Georgia (3.46)

9). Ohio St. (3.47)

Losses= (2) Texas (3.68), Penn St. (3.16)

10). Georgia (3.46)

Losses= (3) Florida (3.46), Auburn (2.99), W. Virginia (2.41)

11). Florida (3.46)

Losses= (3) LSU (3.54), Alabama (3.02), S. Carolina (2.96)

12). Notre Dame (3.19)

Losses= (3) USC (3.58), Ohio St. (3.47), Michigan St. (2.78)

13). Penn St. (3.16)

Losses= (1) Michigan (3.59)

14). Texas A & M (3.13)

Losses= (6) Texas (3.68), Oklahoma (3.68), Colorado (2.90), Clemson (2.81), Texas Tech (2.60), Iowa St. (2.52)

15). Virginia (3.11)

Losses= (5) Miami (3.61), Maryland (3.02), Boston College (2.99), Va. Tech (2.90), N. Carolina (2.70)

16). UCLA (3.08)

Losses= (2) USC (3.58), Arizona (2.76)

17). Maryland (3.02)

Losses= (5) Florida St. (3.65), NC State (2.92), Va. Tech (2.90), Clemson (2.81), W. Virginia (2.41)

18). Alabama (3.02)

Losses= (2) LSU (3.55), Auburn (2.99)

19). Auburn (2.99)

Losses= (3) LSU (3.55), Wisconsin (2.63), Georgia Tech (2.51)

20). Boston College (2.99)

Losses= (3) Florida St. (3.65), Va. Tech (2.90), N. Carolina (2.70)


There are upsets in college football. To expect every team with better recruits and athletes to win every game against teams with inferior talent is unreasonable. However, it is obvious that teams with very good recruiting classes don’t lose very often to teams with significantly lesser recruiting classes.

Here is a list of the top 20 teams in recruiting over the last four years and the average rank of recruiting class of the teams they’ve lost to this season:

Texas=no losses
Oklahoma=2.94
Florida St.=3.09
Miami= 3.24
Michigan=2.94
USC=3.68
Tennessee=3.05
LSU=3.51
Ohio St. =3.42
Georgia= 2.95
Florida=3.17
Notre Dame=3.28
Penn St.=3.59
Texas A&M=3.03
Virginia=3.04
UCLA=3.17
Maryland=2.94
Alabama=3.27
Auburn=2.90
BC=3.08

Of all the teams with a recruiting rating above 3.00, Michigan is the worst on the list. Michigan’s five losses came to teams with an average recruiting ranking of 2.94. Michigan is the only team in the top 13 that lost three games to teams with a recruiting rating of less than 3.00. Michigan’s recruiting ranking is 3.59. That means that Michigan had an average advantage of .65 at every position over the teams it lost to this season. When asked if Michigan’s season was disappointing, Michigan defender Pierre Woods said that Tennessee’s season was a disappointment, not Michigan’s. Tennessee’s average loss came to teams with a recruiting ranking of 3.05. That doesn’t justify Tennessee’s six loss season but in terms of how bad their losses look, Michigan’s were clearly worse. Oklahoma was also a big disappointment this season having had an advantage of .74 in its losses. However, Oklahoma doesn’t perennially lose to teams with inferior talent. Michigan has been the major culprit over the last few years and they continued that trend this season. That doesn’t excuse Oklahoma from underachievement this season but it’s important to remember that even the best programs have a down season. One of the factors that I mentioned above that could allow a team to lose to a noticeably inferior team talent-wise is the breaking in of a new quarterback. Oklahoma was clearly hindered by the graduation of six-year quarterback Jason White. Nonetheless, Oklahoma clearly underachieved in 2005-6 compared to the talent on the roster.

There have been four “surprise” teams in college football this year with the unexpected emergence of Notre Dame, UCLA, Alabama and Penn St. These schools have been relegated to mid-tier programs recently but the recruiting rankings over the last four years put these four schools in the top 16. The success of these four schools may have been a surprise based on recent play on the field but it certainly wasn’t a surprise if you look at each team’s recruiting over the last four years.

In looking at the losses that the top 20 recruiting programs sustained this year on the field, certain lesser teams keep popping up. TCU, Texas Tech, Minnesota, Wisconsin, W. Virginia, and Ga. Tech all had ratings of 2.63 or less yet they all beat a top 20 recruiting program and made it to a bowl game. The coaches of these teams clearly made “something” out of nothing as they generally were the less athletic team on the field. Gary Patterson, Mike Leach, Glen Mason, Barry Alvarez, Rich Rodriguez, and Chan Gailey were among the more accomplished coaches this year. Leach, Mason, and Alvarez have successful teams virtually every year despite having a significant disadvantage in recruiting. Even more impressive may be the success that Northwestern, Louisville, and Fresno St. had this season. None of those three schools had a four year recruiting rating of more than 2.34. Randy Walker, Bobby Petrino, and Pat Hill clearly achieved more than their rosters would suggest. These coaches are clearly more efficient than the Michigan coaching staff. I often wonder how Michigan’s coaching staff would fare at a school with less talent. If they can’t beat lesser teams at Michigan on a regular basis, then how would they beat better teams with a considerably less athletic roster? It’s important to remember that wins and losses aren’t the only measure of success in college football. A coach at Texas Tech will never be able to recruit the same players as Texas. As a result, seven wins at Texas Tech is considerably more impressive than seven wins at Texas. The same can be said for Minnesota. Seven wins at Minnesota is considerably more impressive than seven wins at Michigan.

It would be interesting to take a look at these numbers after every season just to see which teams make the most of their talent and which teams underachieve. The biggest potential drawback of this analysis is the accuracy (or perceived accuracy) of the recruiting services. Whereas college football actually has wins and losses that success can be based on, the recruiting services use subjectivity to rate each recruit. There is no right answer for a recruit’s ranking. It’s possible for certain recruiting services to overrate players from their own region or overrate simply because that player received interest from a major school. As a result, evidence of a team underachieving or overachieving based on the ratings of their recruits over the previous five seasons would likely not hold water. It would be too easy for a coach to say, “just because a recruiting service (Rivals etc.) says that a player is worth five stars doesn’t mean that the player really is that good.” Rating a recruit is not an exact science. However, over five years, I would guess that the recruiting rankings would be accurate enough that the teams with the best rating over five years should, in reality, be the best teams in college football. There might be a difference of a star here and there but over five years, I think the recruiting services end up with a pretty accurate depiction of the talent on a college football roster. However, I doubt any school would ever put any weight on its success on the field compared to its success in recruiting. That would force coaches at major college football programs to look at who they’re beating and who they’re losing to rather than just looking at the overall win total. A coach looks much better if he can simply say, “I won nine games” instead of having to justify not winning eleven games when the talent differential dictated eleven wins.

Five-year recruiting rankings can be helpful in a number of ways. They can indict programs that do less with more. They can highlight programs that win with less. They can give an indication of which teams will break through like Alabama, UCLA, Notre Dame, and Penn St. this season. Since recruiting rankings are not given a whole lot of credence, the extent of this type of analysis will likely end at the fan level.

5 comments:

Anonymous said...

Great article on the recruiting classes. Some lunk heads say that ranking recruiting classes is an inexact science and you can't predict what a team would be like with that data. It seems like the only program having a problem matching success with talent ratings is Michigan. The teams with talent do better than teams with not as much talent--except Michigan. Is there any chance that something good will come out of this years crash?

Anonymous said...

I'd be curious to know how recruit rankings translate into future NFL success. There are a ton of wolverines in the pros, not so many gophers or wildcats (of the Northwestern variety that is). Yet every year we all know that either team is a real threat to beat Michigan. I'd like to train a chimp throw darts. Obviously, that would be fun in its own right, but if you also arrange the playbook in a dartboard pattern, lloyd would have a natural heir.

Jake said...

Former Michigan players in the NFL has always been a calling card of the Michigan football program but I haven't looked at it that way in quite some time. Recently, Michigan has been more known for having a) underperforming players in the NFL and b) players thriving in the NFL who were otherwise marginal at Michigan. That speaks for a few things. First, I think that speaks for Michigan's inability to help a play move from a good high school recruit to a player that can dominate at the next level. Second, I think that speaks for UM's inability to recognize how good some players are on their roster. For instance, Cato June could've been a dominating linebacker at UM but UM kept him in the secondary where he was below average.

I do think that this year's crash will have an effect in the long term. You won't see Lloyd or anyone else change. They've had ten years to change and they haven't done a thing. This season may encourage Bill Martin to emphasize quality coaching when the change comes. I do know one thing, big time recruits will stop coming to UM if things don't change. I mean, if you're a high school recruit outside of Michigan, why on Earth would you go to UM over ND or Ohio St.? Sadly, I think the near future is going to be more of the same. We'll keep getting our fill of 10-3 seasons.

How much does anyone want to bet that Matt Gutierez and Max Martin become legitimate college football players?

Good points, take care Jake.

Lombaowski said...

Great info here!

Anonymous said...

I would have loved a 10-3 season this year. It is a little gut-wretching that the best we can hope for is 10-3 or 9-3, depending games played. That is supposed to be a down year, not an acceptable or average year. I really hate seeing them play so slothly when you know they are capable of so mych more.

 

Powered by Blogger