Why recruiting rankings DO matter...

[markshir]

I saw an earlier thread about how the recruiting rankings correlated to teams in the top 10 of the AP this year and it got me wondering if rankings (I used Rivals) really do matter, and if so, how much.

I took the recruiting rankings for each team from 3 years previous to the season I looked at the games for. For example, games played in 2006 were compared with recruiting rankings from 2004. By this time, JUCO players will have come and gone for the most part, but HS players will be either RS Sophs or Juniors, and heavily involved in their teams performance. They also won't have had a chance to leave for the NFL yet.

I compared every game played in the Big 12 the last three seasons, and this is how it turned out:

In 2004, the team with the higher recruiting ranking from three classes earlier went 37-12.

In 2005, the favored team went 31-18.

In 2006, 33-16.

Overall, the team with the higher recruiting ranking from three classes ago went 101-46, for a winning percentage of 68.7%. That number ignores every factor beyond Rivals rankings, including home field advantage and time of year the game is played.

I also did the same with the bowl games from 2006, assuming that interconference matchups (between teams that are often relatively evenly matched) could present some interesting scenarios and possibly show different winning percentages. While people are fond of pointing to Boise State and their low ranked classes in comparison to Oklahoma, what they are examining seems to be the exception rather than the rule. The games played out as followed, with the recruiting rankings following the teams, and the winner in bold type.

TCU - 55

Northern Illinois – 73

Oregon – 12
Brigham Young – 79


Rice – 111
Troy State – 73


East Carolina – 81
South Florida – 43


New Mexico – 87
San Jose St – 101


Utah – 61

Tulsa – 67

Hawaii – 96

Arizona St - 31

Middle Tenn. – 82
Central Michigan – 108


UCLA – 34
Florida St – 3


Alabama – 15
Oklahoma St – 37


Texas A&M – 13
California – 23


Rutgers – 52

Kansas State – 18

Kentucky – 45

Clemson – 53

Oregon State – 26

Missouri – 29

South Carolina – 35

Houston – 65

Texas Tech – 33

Minnesota – 58

Purdue – 20
Maryland - 17


Boston College – 24

Navy - 116

Texas – 10

Iowa – 38

Georgia – 6

Virginia Tech – 41

Nevada – 108
Miami FL - 4


Tennessee – 11
Penn State – 14


Auburn – 21
Nebraska – 27

Wisconsin – 39

Arkansas – 22

West Virginia – 47

Georgia Tech – 56

Southern Cal – 1

Michigan – 5

Oklahoma – 8
Boise State – 85


Wake Forest – 95
Louisville – 64


LSU – 2
Notre Dame – 32

Western Michigan – 73
Cincinnati – 80


Ohio – 101
Southern Miss – 67


Florida – 7
Ohio State – 9

Ultimately, if I counted correctly, the higher ranked team by Rivals went 21-10, for a winning percentage of 67.7%, in games that are typically set up to produce even matchups. I would suspect that most followers of college football would struggle to accurately predict close to 70% of the games correctly, either over the course of a Big 12 season or during a bowl season. While there are exceptions both in individuals players and teams, in the end, Rivals does a fairly good job of ranking the teams. Hopefully we are one of the exceptions in the next few years.

[darts180]

I'll put it to you this way, if we are going to go to war, would you rather have a cannon, or a sling shot. Now sometimes a sling shot can get that miracle shot in, but day in day out the cannon will get the job done.

[dornstar44]

How do you have time to do this? I'm impressed!

[Wesley]

I saw an earlier thread about how the recruiting rankings correlated to teams in the top 10 of the AP this year and it got me wondering if rankings (I used Rivals) really do matter, and if so, how much.

Ultimately, if I counted correctly, the higher ranked team by Rivals went 21-10, for a winning percentage of 67.7%, in games that are typically set up to produce even matchups. I would suspect that most followers of college football would struggle to accurately predict close to 70% of the games correctly, either over the course of a Big 12 season or during a bowl season. While there are exceptions both in individuals players and teams, in the end, Rivals does a fairly good job of ranking the teams. Hopefully we are one of the exceptions in the next few years.

You have a new way to bet/gamble on bowl games!

[Prone2Clone]

That's an interesting way to look at it...I do think the recruiting rankings, while not fool proof by any means, do have some value. The experts certainly don't nail every player and every class, but in the end, I would feel a lot better about seeing an Iowa State class in the top 30 than bottom 30.

[Wesley]

We would have to sign 3 and 4 star people to break into top 30. No no star and few 2 stars.

[iccyfan]

You have a new way to bet/gamble on bowl games!

Can you go back and see how the lower ranked recruiting classes did versus the spread?

Interesting research - let's hope ISU follows the Boise State anomoly...

[markshir]

Well, I'm not much of a gambler on sports (but I love poker), but I checked into the point spread thing...

Rivals only has spreads for some games, so I'm not sure if the games they don't have were considered even (which is my guess) or if Rivals just has incomplete information.

Anyways, here they are...

TCU (favored by rankings = F) easily covered a 12 point spread against N. Ill.

Central Michigan covered easily against Midd. Tenn. St. (F).

UCLA failed to cover against Florida State (F).

Oklahoma St. covered against Alabama (F).

California easily covered against Texas A&M (F).

South Carolina (F) covered against Houston.

Maryland (F) covered against Purdue.

Boston College (F) failed to cover against Navy.

Oklahoma (F) failed to cover against Boise State.

LSU (F) covered against Notre Dame.

Cincy (F) failed to cover against Western Mich.

Florida (F) didn't let OSU cover, obviously.

In the end, the favored teams by recruiting ranking in games that I had the spread on went 6-6. If I understand the spread right, I could have randomly selected the winners of the game and expected that result? So probably this isn't going to make you rich, but I was hoping for a minute...

[ISUFan22]

I'll put it to you this way, if we are going to go to war, would you rather have a cannon, or a sling shot. Now sometimes a sling shot can get that miracle shot in, but day in day out the cannon will get the job done.

Exactly.

And hopefully every now and then, we'll land a heat-seeking rocket launcher.

[CyCloned]

Ranking do matter, but I think it is less critical after you get past the first 25 to 30 classes. If ISU could move up from the 50-60 range to the 30-40 range there would be some improvement, but not a huge one. The chances of getting into the top 20 are about as good as your chances with a Powerball ticket. There are just too many established schools in states that have superior HS football talent.

[CyCy]

Following recruiting is a little like listening to the NFL draft. For every NFL team their picks are usually described as follows:
First round - Exactly the guy we wanted
Second round - Couldn't believe the guy was still available
Third round - A real sleeper

The good thing about recruiting or the draft is that everybody can feel great about their team because the actual proof won't come for months or years down the road.

The day that none of the signeees are described as "A real sleeper", "Hidden gem", "Undiscovered talent", or "Flew underneath the radar" is when the class is good.