Maybe talent isn't the silver bullet ... | Syracusefan.com

Maybe talent isn't the silver bullet ...

Quazzum69

Stable Genius
Joined
Oct 27, 2012
Messages
3,921
Like
6,107
Here is a little plot of the average recruiting rank of the previous three seasons (according to ) and number of wins for Syracuse, Pitt, Maryland, BC and Rutgers from 2005-2013. For example, the recruiting rank from 2006,2007 and 2008 would be averaged and paired with the number of wins in 2008.

upload_2014-9-28_12-4-40.png


The square orange dots are GROB years and the triangle orange dots are Marrone years. GROB had much better recruiting than Marrone with much different results. In each of their four years, GROB had an average recruiting rank of 51 while Marrone had an 85.

I also used the current year (2008 rank with 2008 season) and two year average. There appears to be no relationship of recruiting rank (a pretty good indicator of "talent") with number of wins. I have no idea how reliable Rivals is but it seems like the best indicator of talent without getting too involved.

Maybe I should try recruiting rank with yards gained ...
 
Nobody said that coaching doesn't matter.
 
Here is a little plot of the average recruiting rank of the previous three seasons (according to ) and number of wins for Syracuse, Pitt, Maryland, BC and Rutgers from 2005-2013. For example, the recruiting rank from 2006,2007 and 2008 would be averaged and paired with the number of wins in 2008.

View attachment 26268

The square orange dots are GROB years and the triangle orange dots are Marrone years. GROB had much better recruiting than Marrone with much different results. In each of their four years, GROB had an average recruiting rank of 51 while Marrone had an 85.

I also used the current year (2008 rank with 2008 season) and two year average. There appears to be no relationship of recruiting rank (a pretty good indicator of "talent") with number of wins. I have no idea how reliable Rivals is but it seems like the best indicator of talent without getting too involved.

Maybe I should try recruiting rank with yards gained ...

Is it possible that finding "no relationship" between recruiting ranking/talent and wins in your data was confounded by how poor GRob was as a HC?

This issue has been evaluated many times using a much larger sample size and the correlation is undeniable between talent (recruiting rankings) and number of wins.

2008-12_Recruiting-Head-to-Head_Chart.jpg


"On the final count, the higher-ranked team according to the recruiting rankings won almost exactly two-thirds of the time (66.4 percent of the time, to be exact), and every "class" as a whole had a winning record against every class ranked below it every single year. (The only exception, if it even qualifies, came last year, when "two-star" teams finished one game below .500 in head-to-head collisions with "one-star" teams. Elsewhere, the hierarchy held across every line.) The gap on the field also widened with the gap in the recruiting scores: At the extremes, "one-star" recruiting teams managed a grand total of six wins over "four-star" and "five-star" recruiters in 59 tries. Where a small handful of teams defied their rankings, none managed to do so as part of a larger group.

Which is, again, about as accurate as we can realistically expect from a system designed to predict an uncertain future."

http://www.cbssports.com/collegefootball/eye-on-college-football/21641769
 
The quality of the "ranking" of recruiting classes while Marrone was here is highly suspect. Lack of/quality of coverage, recruiting strategy, overall discounting of players going to SU. Those rankings don't take into account transfers and late additions.

All you have to see is the evolution of the roster, how fast it turned over to know that Marrone's era overall recruiting was better.
 
Here is a little plot of the average recruiting rank of the previous three seasons (according to ) and number of wins for Syracuse, Pitt, Maryland, BC and Rutgers from 2005-2013. For example, the recruiting rank from 2006,2007 and 2008 would be averaged and paired with the number of wins in 2008.

View attachment 26268

The square orange dots are GROB years and the triangle orange dots are Marrone years. GROB had much better recruiting than Marrone with much different results. In each of their four years, GROB had an average recruiting rank of 51 while Marrone had an 85.

I also used the current year (2008 rank with 2008 season) and two year average. There appears to be no relationship of recruiting rank (a pretty good indicator of "talent") with number of wins. I have no idea how reliable Rivals is but it seems like the best indicator of talent without getting too involved.

Maybe I should try recruiting rank with yards gained ...

Interesting graph, thanks for the work. Would probably be better served with NFL draft picks per current roster rather than star rankings.
 
It could also be that GRob's top three recruits, from a rating standpoint--Lavar Lobdell, Averin Collier, and Doug Hogue--combined to contribute virtually nil offensively. So, adding 4 star players like that inflated class rankings, but had no impact on the field.

Hogue obviously excelled after being switched to defense, but that was after the coaching change.
 
Interesting graph, thanks for the work. Would probably be better served with NFL draft picks per current roster rather than star rankings.

Give points for star rating, for All-Conference and All-America selections, pro draft picks and guys who made the pro bowl to get a good view of overall talent. Of course, that would take a while...
 
Interesting graph. Might give a little insight into why Marrone had some success here and why Shafer is struggling some now. Talent matters like many have said.

I've seen a lot of posters say we just don't have the talent a ND does. Yet every year when we get a bunch of 2-3 star players, people say we get screwed and our recruits are great. In converse, when a ND signs a bunch of 4 stars, people say those players are only 4 stars because ND recruited them. Talent matters.
 
CIL said:
Interesting graph, thanks for the work. Would probably be better served with NFL draft picks per current roster rather than star rankings.

I value draft picks over star rankings coming out of high school but that only assesses the top end talent on the roster.
 
Interesting graph. Might give a little insight into why Marrone had some success here and why Shafer is struggling some now. Talent matters like many have said.

I've seen a lot of posters say we just don't have the talent a ND does. Yet every year when we get a bunch of 2-3 star players, people say we get screwed and our recruits are great. In converse, when a ND signs a bunch of 4 stars, people say those players are only 4 stars because ND recruited them. Talent matters.

The average fan/poster (in all sports, for all teams) is an idiot. They always believe their team is better than they really are, they believe that everyone else is out to get them whether it's power rankings, recruiting rankings, draft rankings, officials or something else.

In general I think that we have a few recruits who are underrated by star services, and teams like and have some kids who are overrated. However the majority tend to be about right over the life of their careers.
 
The average fan/poster (in all sports, for all teams) is an idiot. They always believe their team is better than they really are, they believe that everyone else is out to get them whether it's power rankings, recruiting rankings, draft rankings, officials or something else.

In general I think that we have a few recruits who are underrated by star services, and teams like and have some kids who are overrated. However the majority tend to be about right over the life of their careers.

And the solution to all the team's issues is to can the head coach and/or the coordinators.

If we just had the right head coach and coordinators, we would never lose.
 
Is it possible that finding "no relationship" between recruiting ranking/talent and wins in your data was confounded by how poor GRob was as a HC?

This issue has been evaluated many times using a much larger sample size and the correlation is undeniable between talent (recruiting rankings) and number of wins.

2008-12_Recruiting-Head-to-Head_Chart.jpg


"On the final count, the higher-ranked team according to the recruiting rankings won almost exactly two-thirds of the time (66.4 percent of the time, to be exact), and every "class" as a whole had a winning record against every class ranked below it every single year. (The only exception, if it even qualifies, came last year, when "two-star" teams finished one game below .500 in head-to-head collisions with "one-star" teams. Elsewhere, the hierarchy held across every line.) The gap on the field also widened with the gap in the recruiting scores: At the extremes, "one-star" recruiting teams managed a grand total of six wins over "four-star" and "five-star" recruiters in 59 tries. Where a small handful of teams defied their rankings, none managed to do so as part of a larger group.

Which is, again, about as accurate as we can realistically expect from a system designed to predict an uncertain future."

http://www.cbssports.com/collegefootball/eye-on-college-football/21641769

Most of
Is it possible that finding "no relationship" between recruiting ranking/talent and wins in your data was confounded by how poor GRob was as a HC?

This issue has been evaluated many times using a much larger sample size and the correlation is undeniable between talent (recruiting rankings) and number of wins.

2008-12_Recruiting-Head-to-Head_Chart.jpg


"On the final count, the higher-ranked team according to the recruiting rankings won almost exactly two-thirds of the time (66.4 percent of the time, to be exact), and every "class" as a whole had a winning record against every class ranked below it every single year. (The only exception, if it even qualifies, came last year, when "two-star" teams finished one game below .500 in head-to-head collisions with "one-star" teams. Elsewhere, the hierarchy held across every line.) The gap on the field also widened with the gap in the recruiting scores: At the extremes, "one-star" recruiting teams managed a grand total of six wins over "four-star" and "five-star" recruiters in 59 tries. Where a small handful of teams defied their rankings, none managed to do so as part of a larger group.

Which is, again, about as accurate as we can realistically expect from a system designed to predict an uncertain future."

http://www.cbssports.com/collegefootball/eye-on-college-football/21641769

What is the rationale for the point cutoffs for the star rankings? Why are "five star" teams above 10,000 points and "four star" 6000-9999? The article gives no explanation of why the teams are grouped like that. Maybe they used some type of clustering method but they don't specify.

For all three years combined, the differences in winning percentage between stars are not even significantly different for a lot of these things.

For example, five star teams don't have statistically different winning percentages against 4 or 3 star teams (.663 with 96 games and .667 with 93 games). Four star teams do not have a statistically significant difference between 3 or 2 star teams (.64 with 111 games and .654 with 78 games). For 2 star v. 1 star: is .533 significantly different than random with only 75 games? I doubt it.

My theory is that talent only makes a difference when you have freak athletes. A team like Syracuse is never going to have the players that "five star" teams get. When you get to the level of FSU, Alabama, Auburn those athletes can make up for coaching deficiencies and make huge plays.

Unless you are at the very top or very bottom, talent level is less important than other factors (like coaching) to winning (so against teams like Maryland, Pitt, BC, Rutgers, etc.).
 
Interesting graph. Might give a little insight into why Marrone had some success here and why Shafer is struggling some now. Talent matters like many have said.

I've seen a lot of posters say we just don't have the talent a ND does. Yet every year when we get a bunch of 2-3 star players, people say we get screwed and our recruits are great. In converse, when a ND signs a bunch of 4 stars, people say those players are only 4 stars because ND recruited them. Talent matters.

Marrone had four of the five WORST three year ranking averages (Shafer has the other, which should tell you how bad Marrone's recruiting was) of these teams and years but still got average results. GROB had decent rankings but failed miserably. It used to be coaching that was the problem now it's the talent?

Notre Dame played for the championship two years ago and are on another level - a level that SU should not be expected to attain. We should be compared to BC, Pitt, Rutgers right now.
 

Forum statistics

Threads
170,327
Messages
4,885,180
Members
5,991
Latest member
CStalks14

Online statistics

Members online
199
Guests online
1,086
Total visitors
1,285


...
Top Bottom