Do stars matter...I think not!

[AndyDufresne]

For the love of God another "stars" thread...maybe I'll come at this another angle. Take the top 10 ranked teams in the country for example and look at their average ranked recruiting class from 06' to 09':

 

Alabama: 11, 10, 1, 1 for an average of 5.75

Texas: 5, 5, 14, 5 for an average of 7.25

TCU: 61st, 80th, 96th, 46th for an average of 70.75

Cinci: 102nd, 89th, 67th, 60th for an average of 79.5

Florida: 2, 1, 3, 11 for an average of 4.25

Boise St: 70th, 68th, 89th, 72nd for an average of 74.75

Oregon: 49th, 11th, 19th, 32nd for an average of 26

Ohio St: 12, 15, 4, 3 for an average of 8.5

Georgia Tech: 57th, 18th, 49th, 49th for an average of 43.25

Iowa: 40th, 28th, 53rd, 63rd for an average of 46

 

Now take the teams that are typically in the top 10 out of this group. Cinci hasn't been there for probably 40+ years and the team can be largely attributed to the job Kelly did with them. TCU usually has a decent year about every four years but doesn't ever make it in the top five like they did this year, IMO it can be chalked up to the year just flat coming together for them. It won't happen for a while after this year. Boise state is there almost every year bc they don't play anyone. Outside of Oregon the two biggest games they had this year were Fresno State and Nevada!! Are you kidding me? Enough said. Georgia Tech is a good team but they run a very unique system for today's college football and they do a very good job recruiting for their scheme much like NU did in their hay day. Iowa should have had at least three losses this year and should have been knocked outside the top 15. They still aren't a regular member of the top ten and just got plain lucky this year. The only teams in this group that consistently are in the top ten are Bama, average rank of 5.75, UT, average rank of 7.25, Florida, average rank of 4.25 and Ohio State, avearge rank of 8.5.

 

You are misrepresenting the facts to fit within your star-centric view. You mention that Iowa should have had 3 more losses, but don't mention that (based on the assumption that your only basis for this is a close game), Ohio State should have had 2 more (Navy and Iowa), Alabama should have had 2 (Tennessee and Auburn), Texas should have had 2 (Oklahoma and Nebraska), and Florida should have had 1 more (Arkansas).

 

You state that Bama is consistently in the top 10. They have been top 10 two of the last four years (including this year). So have TCU and Boise State.

 

With stars, people see what they want to see. There is plenty of evidence for both sides to point to in order to prove their point.

[zoogs]

No, I'm not sure there is. huskers1 didn't help his case by insisting to compensate for every single case, when in fact anomalies are just part of the game.

 

Stars are not 100% indicative of success. Fact. Not even close to 100%.

 

That is a far, far cry from "doesn't matter."

 

Year after year, in the Top 10, you are going to see teams with consistently strong recruiting. You'll have your exceptions to the rule, but these are exceptions. This is a bit of an unusual year where we see teams like Cincy and TCU go up to the very top.

 

But you look at the teams with high-star power recruiting, and you find a ton of teams that are consistent contenders most years. You look at teams with recruiting on par, in terms of stars, with Cincy, TCU, and the like, and you find a MINISCULE percentage of teams that are consistently contending.

 

Just because TCU did it with 3-stars and ND couldn't do it with 5-stars, does NOT mean stars don't matter. Stars aren't everything but they are a statistically significant indicator.

[AndyDufresne]

No, I'm not sure there is. huskers1 didn't help his case by insisting to compensate for every single case, when in fact anomalies are just part of the game.

 

Stars are not 100% indicative of success. Fact. Not even close to 100%.

 

That is a far, far cry from "doesn't matter."

 

Year after year, in the Top 10, you are going to see teams with consistently strong recruiting. You'll have your exceptions to the rule, but these are exceptions. This is a bit of an unusual year where we see teams like Cincy and TCU go up to the very top.

 

But you look at the teams with high-star power recruiting, and you find a ton of teams that are consistent contenders most years. You look at teams with recruiting on par, in terms of stars, with Cincy, TCU, and the like, and you find a MINISCULE percentage of teams that are consistently contending.

 

Just because TCU did it with 3-stars and ND couldn't do it with 5-stars, does NOT mean stars don't matter. Stars aren't everything but they are a statistically significant indicator.

 

How many exceptions can you have before it is a trend?

 

2009

3. TCU

4. Cincinnati

6. Boise State

7. Oregon

9. Georgia Tech

10. Iowa

 

2008

2. Utah

7. TCU

10. Oregon

 

2007

4. Missouri

6. West Virginia

7. Kansas

9. Virginia Tech

10. Boston College

 

2006

5. Boise State

6. Louisville

7. Wisconsin

10. West Virginia

 

So are 45% of the top 10 for the last four years not being among the recruiting elite an exception? Is a 55% success rate for the recruiting services' team rankings "the rule"?

[caveman99]

No, I'm not sure there is. huskers1 didn't help his case by insisting to compensate for every single case, when in fact anomalies are just part of the game.

 

Stars are not 100% indicative of success. Fact. Not even close to 100%.

 

That is a far, far cry from "doesn't matter."

 

Year after year, in the Top 10, you are going to see teams with consistently strong recruiting. You'll have your exceptions to the rule, but these are exceptions. This is a bit of an unusual year where we see teams like Cincy and TCU go up to the very top.

 

But you look at the teams with high-star power recruiting, and you find a ton of teams that are consistent contenders most years. You look at teams with recruiting on par, in terms of stars, with Cincy, TCU, and the like, and you find a MINISCULE percentage of teams that are consistently contending.

 

Just because TCU did it with 3-stars and ND couldn't do it with 5-stars, does NOT mean stars don't matter. Stars aren't everything but they are a statistically significant indicator.

 

How many exceptions can you have before it is a trend?

 

2009

3. TCU

4. Cincinnati

6. Boise State

7. Oregon

9. Georgia Tech

10. Iowa

 

2008

2. Utah

7. TCU

10. Oregon

 

2007

4. Missouri

6. West Virginia

7. Kansas

9. Virginia Tech

10. Boston College

 

2006

5. Boise State

6. Louisville

7. Wisconsin

10. West Virginia

 

So are 45% of the top 10 for the last four years not being among the recruiting elite an exception? Is a 55% success rate for the recruiting services' team rankings "the rule"?

Good post, saved me the research. Recruiting does in deed matter, but those Top 10's show that Coaching is a bigger factor IMO. Having good to great recruiting does matter, the % chance of a 4 or 5 star being All Conference and/or All American is much higher than a 2 or 3 star. There have been multiple stories and posts to show such. However an All Conference Player doesn't make a great team, nor does a few of them. Heck all you have to do is look at OU this year or MU the last 2 years. This is the point that most star obsessed gazers forget, but Andy's post illustrates that flaw in their thinking. Great Coaching is more important than recruiting, but both are needed to be really great.

[huskers1]

Ok Boise State plays one hard game a year and if they get by that then they are almost a sure fire undefeated team bc the hardest teams they play are the likes of nevada and fresno. The win against OU did wonders for their program bc that made them a mainstay inside the pre season top 10-15 and bc they don't play even an average schedule they either go undefeated or have one loss and just chip their way up the polls. I keep trying to emphasize CONSISTENTLY, CONSISTENTLY, CONSISTENTLY in the top 10. Yes of course there are always going to be teams (mizzou, ku, BC, utah) that are in the top 10 every year that don't have decent recruiting classes but what keeps them there are top 25 recruiting classes year in and year out. Again, of course there are the excpeptions like TCU or West Virginia for two examples that run a system that lends to not needing the best athletes due to the system that they run. They need a dynamic playmaker here and there and it works for them. I agree a big factor is coaching, I never once said that didn't matter. But even the best coaches will fail without talent. Look at the consistent top 10 teams and who are they? USC had a down year but they've gone to a BCS bowl the last what 9 years? Florida, Texas, OU, Ohio State, LSU, Georgia, Penn State for a while now. I'd like to see what their average classes are. Sure there are going to be outstanding two and three star players emerging as stars, it happens every year. Same goes for five and four stars being flops, that happens every year. But to compete on a CONSISTENT basis for a NC recruiting "stars" does matter.

[zoogs]

So are 45% of the top 10 for the last four years not being among the recruiting elite an exception? Is a 55% success rate for the recruiting services' team rankings "the rule"?

 

It's all in how you look at it.

 

Sure, 45% is a lot. Let's say that year-in, year-out, only half the Top 10 consists of the "recruiting elite". That's reasonable; college football is extremely competitive and it's impossible to sustain at the top forever. There will always be openings and always be willing teams jostling for those spots. So let's just say 50-50 is the way it is.

 

Now: among the "recruiting elite" category, what percentage of these teams end up in the Top 10?

 

Now: among the "lower-tier recruiting" category, what percentage of these teams end up Top 10?

 

TREMENDOUS difference. That is why if a school averages 3 stars or less in recruiting, and still ends up in the Top 10, it's an exception to the rule. Because the vast, vast majority of schools that average that amount of stars, don't. That is why if you recruit at a close to 4-star average level, you have reason to be optimistic. Because schools that recruit at this level have a much higher shot at Top 10 status than those that don't.

 

It's like the NFL draft argument. You can look at some drafts and say, wow, check out how many 3-stars, 2-stars, or no-stars are being drafted in the first round. Stars must not be indicative of talent. Then you flip it around and see that (just throwing numbers out there), say, 10% of five star recruits make it, and 1% of (3/2/0) stars.

 

There will always be plenty of exceptions to the rule, but it doesn't mean they still aren't exceptions. You will never be able to look at any given school that averages 2.8 stars recruiting and say, "I expect them to be major contenders." You will never be able to look at any given school that averages 3.8 stars recruiting and reasonably say, "I expect them to barely make bowl games."

[AndyDufresne]

So using our current 2010 Rivals ranking of 29th, our 5 year class average for 2006-2010 is 24th. If I told you that using historical data, I predict that in 2010 our final ranking will be somewhere between 4th and 44th, would you consider this a good predictor of future success? Because in a multi-year study, BCS teams were found to rank an average of 20 places higher or 20 placess lower than their 5 year Rivals class average. To draw any meaningful predictions based on recruiting rankings, I would think most would like to see a tighter correlation between the two.

[zoogs]

Not too bad. With our recruiting, based only on that alone, you would probably expect a more outside shot at #1 (undefeated) but also that we'd fall safely within the Top 50 (say, no worse than 7-5).

 

I mean, to narrow it down too much further would just be crystal ball. Stars are hardly everything, after all. But it gives you a good sense of the bounds you'd expect a team to fall in. Solid recruiting teams should be safely in the top half, and poor recruiting teams should be in the bottom half. I think that's a fairly clear delineation for the most part and shows that stars do mean something. Now, whether a team stocked with talent goes 11-1 or 8-4 in a year...it could be a matter of a couple plays.

 

By the way, the "+/- 20 year thing" sounds more like bounds than averages. Is there a figure for the actual average? That almost sounds like the statistical mean is not too far off from the rivals ranking.

[AndyDufresne]

Not too bad. With our recruiting, based only on that alone, you would probably expect a more outside shot at #1 (undefeated) but also that we'd fall safely within the Top 50 (say, no worse than 7-5).

 

I mean, to narrow it down too much further would just be crystal ball. Stars are hardly everything, after all. But it gives you a good sense of the bounds you'd expect a team to fall in. Solid recruiting teams should be safely in the top half, and poor recruiting teams should be in the bottom half. I think that's a fairly clear delineation for the most part and shows that stars do mean something. Now, whether a team stocked with talent goes 11-1 or 8-4 in a year...it could be a matter of a couple plays.

 

By the way, the "+/- 20 year thing" sounds more like bounds than averages. Is there a figure for the actual average? That almost sounds like the statistical mean is not too far off from the rivals ranking.

 

That is the average. The bounds would be much higher. For instance, in the study Cincinnati outperformed it's recruiting class average by about 75 places.

[Ohio Pete]

I think when we've looked at it in years past, we've found a weak correlation between recruiting rankings and top 25 rankings. Plenty of other things go into on the field success, recruiting is just one of them.

[zoogs]

(oops, I think I misunderstood): you are saying the average difference is 20? The way you put it (between 4th and 44th) made it sounds like a standard deviation range or something.

[dylan]

I suspect that if you take every team, add up their Rivals class scores from 2005-2008 inclusive, and rank them, you will find the Top 10 or Top 20 contributes a greater % of the current Top 25 than #'s 21-30, 31-40, and so on.

 

i suspect you're right about this, but you must also consider the contributing factor that teams that win games tend to get the benefit of the doubt with the ranking of their recruits. recruiting obviously matters, and recruiting rankings have a loose correlation with actual talent. the question is how loose that correlation is.

 

i find it interesting though that the on field results seem to be questionable even with the top recruiting schools. schools that by all accounts are just nailing it in getting the top recruits. if you can't predict success for even those top schools, with their acclaimed kids, what does that say about their rankings of the rest of the schools, with recruits that are really a crap shoot?

[hskrpwr13]

Here's an easy way to look at; best to worst:

 

1) good coaching and perceived good talent

2) good coaching and perceived mid level talent or worse

3) poorer coaching and perceived good talent

4) poorer coaching and perceived mid level latent or worse