Coaching vs. Recruiting Stars

Donate To GopherHole






The article says that they took into account the last 5 years of finishes in Massey ratings, and then for the recruiting rankings in a given year they used the average Rivals ratings of all 5 recruiting classes that might be playing in that year, that is, freshmen through redshirt seniors. So, if I'm understanding it correctly, this would include Kill's 2011, 2012, and 2013 recruiting classes, all of Brewster's classes, and the last couple of Mason's, with these classes spread over two 3-9 seasons, two 6-7 seasons, and last year's 8-5.
 

Interesting. I'd like to see more years included but this shows what seems to be empirical on the field: helmet programs are overrated. Correct me if I'm wrong but only Alabama outperformed their recruiting rankings, and just.

Rosters full of primary donnas. SEC teams almost entirely below the line. Hmm.
 

This analysis is interesting, but flawed from the get-go because they didn't give more weight to upperclassmen. Their rationale for not doing so is specious at best.
 

maroonedinCA said:
Except, of course, the article, using numeric evidence, says they do.
Click to expand...

Okay, they do. Bigger, faster, more accomplished players should result in better teams. Uh, except when they don't, which usually creates problems (see Michigan) far more difficult to identify and resolve than an apparent lack of talent. After all, it was our own Herb Brooks that taught the world a lesson in team building.
 



I've changed my mind. Stars are all that matter. :banghead::banghead::banghead:
 

maxwellsmart said:
Okay, they do. Bigger, faster, more accomplished players should result in better teams. Uh, except when they don't, which usually creates problems (see Michigan) far more difficult to identify and resolve than an apparent lack of talent. After all, it was our own Herb Brooks that taught the world a lesson in team building.
Click to expand...

You don't know what statistically significant means, do you?
 


maxwellsmart said:
Okay, they do. Bigger, faster, more accomplished players should result in better teams. Uh, except when they don't, which usually creates problems (see Michigan) far more difficult to identify and resolve than an apparent lack of talent. After all, it was our own Herb Brooks that taught the world a lesson in team building.
Click to expand...

So because you're able to cite a team that doesn't support the correlation it means it's totally invalid?

"There was a significant correlation (r=0.77) between recruiting and on-field rankings. The mean difference between recruiting and on-field rankings was 16.6, and the median difference was 14.8. Arizona and Nebraska had the smallest discrepancies with a 0.2 difference, while Navy had the largest with a 57.4 difference. Recruiting obviously isn't an exact science and has its flaws. But given there's more than 120 teams in FBS college football, these rankings give a decent idea of where teams will sit in upcoming years."

http://mathbits.com/MathBits/TISection/Statistics2/correlation.htm

A correlation greater than 0.8 is generally described as strong, whereas a correlation
less than 0.5 is generally described as weak. These values can vary based upon the
"type" of data being examined. A study utilizing scientific data may require a stronger
correlation than a study using social science data.

As for the idea that helmet schools are overrated, you guys realize it's pretty hard to significantly outperform your recruiting rank when that rank is in the top 10, right?
 



What makes this data look worse for the coaching at UMN vs. the recruiting is that the class of 2008 is factored in each of the previous 5 year recruiting class averages, which we were ranked 17th by rivals. This made it look like we recruit better than Wisconsin has in the past, which is flat out not true. It's more of an indictment on Tim Brewster and his horrid coaching than on Minnesota as a program. If we want to see what is a better indicator for us at the U is to look at Northern Illinois and their over-achievement vs. recruiting rankings.
 

Run...stats are the answer to everything.
 

RunGopherRun said:
What makes this data look worse for the coaching at UMN vs. the recruiting is that the class of 2008 is factored in each of the previous 5 year recruiting class averages, which we were ranked 17th by rivals. This made it look like we recruit better than Wisconsin has in the past, which is flat out not true. It's more of an indictment on Tim Brewster and his horrid coaching than on Minnesota as a program. If we want to see what is a better indicator for us at the U is to look at Northern Illinois and their over-achievement vs. recruiting rankings.
Click to expand...

Agreed. Plus, remember that the 2008 class was rated 17th, but almost none of those top recruits both ended up on campus and became starters; for various reasons (e.g., Maresh, Hill). In effect, the rating for that class ended much lower if you didn't include those who failed to make the team.

The data probably should have been created using "adjusted" recruiting rankings based on players that actually ended up making the teams, which I believe was not the case. That said, the data do accurately portray the fact that Minnesota performed very poorly on the field given their recruiting rankings in the late '00s.
 

cncmin said:
Agreed. Plus, remember that the 2008 class was rated 17th, but almost none of those top recruits both ended up on campus and became starters; for various reasons (e.g., Maresh, Hill). In effect, the rating for that class ended much lower if you didn't include those who failed to make the team.

The data probably should have been created using "adjusted" recruiting rankings based on players that actually ended up making the teams, which I believe was not the case. That said, the data do accurately portray the fact that Minnesota performed very poorly on the field given their recruiting rankings in the late '00s.
Click to expand...

Is data more important than stats?
 


justthefacts said:
So because you're able to cite a team that doesn't support the correlation it means it's totally invalid?

"There was a significant correlation (r=0.77) between recruiting and on-field rankings. The mean difference between recruiting and on-field rankings was 16.6, and the median difference was 14.8. Arizona and Nebraska had the smallest discrepancies with a 0.2 difference, while Navy had the largest with a 57.4 difference. Recruiting obviously isn't an exact science and has its flaws. But given there's more than 120 teams in FBS college football, these rankings give a decent idea of where teams will sit in upcoming years."

http://mathbits.com/MathBits/TISection/Statistics2/correlation.htm

A correlation greater than 0.8 is generally described as strong, whereas a correlation
less than 0.5 is generally described as weak. These values can vary based upon the
"type" of data being examined. A study utilizing scientific data may require a stronger
correlation than a study using social science data.

As for the idea that helmet schools are overrated, you guys realize it's pretty hard to significantly outperform your recruiting rank when that rank is in the top 10, right?
Click to expand...

Only one person mentioned "totally invalid" and it wasn't me. Not sure how to express that concept via bean counting, but there it is.
 


This isn't a rigorous statistical analysis, and they aren't trying to make it one. Just a nice graphic with some descriptive statistics to generate some interesting questions.
 

justthefacts said:
As for the idea that helmet schools are overrated, you guys realize it's pretty hard to significantly outperform your recruiting rank when that rank is in the top 10, right?
Click to expand...

Glancing at the chart, most of the underperforming helmet schools did not have average rankings in the top 10.

Furthermore, the preseason rankings bias the rankings all season. The preseason rankings are equal parts speculation and perception. I would say this study, even with its flaws, is pretty damning for certain programs over that time span. They didn't include upperclassmen for any of the teams so that isn't really a strong factor. It's just a snapshot though.

In aggregate, recruiting rankings matter but it is really refreshing to see more evidence they are not all-important.
 

The problem with statistics is that it's hard to come up with an empirical measure for subjective qualities: heart, desire, willingness to accept coaching, willingness to put the team's goals ahead of personal goals - those are some of the hard-to-quantify qualities that go into a winning program. All the 4* and 5* recruits in the world don't mean squat if you don't have the intangibles to go with them.

having said that, it would be fun to see what Kill and his staff could do with some real blue-chip players - providing they could recruit blue-chip players who are willing to play within the system. The lower-recruited guys with a chip on their shoulders seem to be more of the type of player Kill likes to have.
 

You must log in or register to reply here.



Top Bottom