Statistics tell the story III

[daj2576]

Great numbers for Cincy! Hope it does mean something.

What recruiting rankings did you use? Scout, Rivals, some consensus rating?

Rivals, arbitrarily. I just prefer their interface and I know that they utilize JUCO's in their rankings.

 

[Voldemonium]

daj2576,
Nice intelligent post. One of my complaints regarding Dooley was that the talent didn't seem to improve over the course of his stay. And any areas of obvious talent improvement such as the O-line this year, were never reflected in the win column. Your research pretty much confirms this observation.

 

[daj2576]

daj2576,
Nice intelligent post. One of my complaints regarding Dooley was that the talent didn't seem to improve over the course of his stay. And any areas of obvious talent improvement such as the O-line this year, were never reflected in the win column. Your research pretty much confirms this observation.

I developed this algorithm mid-September this year. When I did an in-depth review of the SEC over the past half decade that is when my support for Dooley failed. When I saw that talent is a good predictor and our talent always fell short, even when trying to adjust for the massive failure that was our 2009 class, it was all inexplicable.

 

[FLVOL_79]

I developed this algorithm mid-September this year. When I did an in-depth review of the SEC over the past half decade that is when my support for Dooley failed. When I saw that talent is a good predictor and our talent always fell short, even when trying to adjust for the massive failure that was our 2009 class, it was all inexplicable.

Math outed Dooley's miserable coaching, excellent!

 

[daj2576]

Math outed Dooley's miserable coaching, excellent!

math works!

 

[theutvolunteers]

Simple you need top 10 classes every year to compete for the SEC.

Pretty much.

 

[daj2576]

Pretty much.

There are exceptions: South Carolina with Spurrier and Arkansas with Petrino are good examples.

 

[theutvolunteers]

Top shelf coaching like those you mentioned certainly make up ground.

 

[vol-un-tear]

For Vandy, you also need to factor in the amount of Redshirt Seniors they had on defense which Steve Martin made possible, but for which Franklin is taking credit.

Vandy is loaded with juniors. Franklin is coaching up Johnson's talent no doubt. Not many red shirt seniors though.

 

[daj2576]

I thought of another interesting way to look at things. The following is the BCS top 25 with the average recruiting ranking next to each team in parenthesis.

1. Notre Dame (16.25)
2. Alabama (2)
3. Florida (7)
4. Oregon (17.5)
5. Kansas State (69.75)
6. Stanford (18.25)
7. Georgia (9.5)
8. LSU (8)
9. Texas A&M (20.25)
10. South Carolina (18.25)
11. Oklahoma (11.25)
12. Florida State (6.25)
13. Oregon State (48)
14. Clemson (19.5)
15. Northern Illinois (86.5)
16. Nebraska (22.5)
17. UCLA (20)
18. Michigan (14)
19. Boise State (65.25)
20. Northwestern (70.5)
21. Louisville (48.75)
22. Utah State (103.75)
23. Texas (3.25)
24. San Jose State (96.25)
25. Kent State (81)

That means that 16 teams were in the BCS top 25 that were in the top 25 ranked only by average recruiting.

What teams were left out of the BCS who recruited in the top 25?

USC (4.25)
Auburn (10)
Ohio State (10.75)
Tennessee (12.25)
Miami, Fl (19)
California (23.25)
Ole Miss (23.75)
North Carolina (24.5)
Virginia Tech (25.25)

 

[tm3]

I tried to put my BJ question at your blog but apparently did not enter it correctly. Anyway thanks for the look over 3 years. Good news for UT fans is that BJ results were above predicted for 2 years, bad news is that cincy average recruiting rank slowly declined minus Kelly.

Your last post re top 25 is interesting also. If UT was going to settle for less than a top tier coach I would have preferred hiring the dude from San Jose st and loading up the staff with first rate recruiters like zook and rodney garner (but, but, he will never leave GA -- ha ha ha). IMO the death blow for BJ in the SEC is going to be inability to compete with the big boys in recruiting wars.

 

[daj2576]

I tried to put my BJ question at your blog but apparently did not enter it correctly. Anyway thanks for the look over 3 years. Good news for UT fans is that BJ results were above predicted for 2 years, bad news is that cincy average recruiting rank slowly declined minus Kelly.

Your last post re top 25 is interesting also. If UT was going to settle for less than a top tier coach I would have preferred hiring the dude from San Jose st and loading up the staff with first rate recruiters like zook and rodney garner (but, but, he will never leave GA -- ha ha ha). IMO the death blow for BJ in the SEC is going to be inability to compete with the big boys in recruiting wars.

If my numbers are correct, it is actually the opposite. Jones recruiting numbers increased the overall average of Cincy every year he was there (that is why I put them in bold, so you could see them creeping up the list). It appears to me that Jones recruiting improved the average almost 14 spots while he was there. Considering that is a 4 year average, that impresses me.

I have read several articles that suggest (and I admit I have not researched this to verify) that Jones inherited a roster mess at Cincy. By these accounts, the simple fact that he won what he should have with the talent before attrition is a huge plus, at least to me.

 

[tm3]

Yes I see that I interpreted it backwards from what you posted. My bad, thanks for correcting.

UT fans would hope that it was a roster mess as a root cause in going from 12-1 to 4-8.

 

[Weee'rre Baaccckk]

That is interesting work, and I appreciate you compiling the information. But, I don't think that performing within a 2 game window is statistically helpful really. I mean, it's not much of a measure of reliability to be within 2 games in an 8 game conference schedule.

Agreed. Irrelevant information/stats

 

[daj2576]

Agreed. Irrelevant information/stats

Interesting interpretation. Just out of curiosity, please enlighten me about what your interpretation of "relevant" might be?

One of the best definitions of relevance comes from the "Rules of Evidence" as promulgated by the federal court system and the State of Tennessee which if memory serves states something to the effect of: information is relevant if it has any tendency to make the existence of any fact that is of consequence to the determination of the action more probable or less probable than it would be without the evidence.

 

[daj2576]

Yes I see that I interpreted it backwards from what you posted. My bad, thanks for correcting.

UT fans would hope that it was a roster mess as a root cause in going from 12-1 to 4-8.

Again, I haven't independently verified that information. I have heard "pundits" discuss the issue and have read articles that alluded to him stepping into a roster that suffered a great deal of attrition after Kelly left.

 

[Rockytopia]

It might help..

2012 stats;

Rank:

UT ypg surrendered, #14 in SEC

The worst since before those stats were first recorded in 1950.

UT ppg surrendered, #14 in SEC
Worst in UT history since the 1890s

UT Number of sacks; #14 in SEC

Again, I haven't independently verified that information. I have heard "pundits" discuss the issue and have read articles that alluded to him stepping into a roster that suffered a great deal of attrition after Kelly left.

Have you read 'Bringing down the house?'

The thing about recruiting statistics, although pertinent, those are somewhat subjective and not entirely scientific.

I've never seen a rating given to player developement.

Some players come in from HS as 'projects' and end up being all pro in the NFL.

Some come in as the second coming and never pan out.

Not to try to diminish your good work with the recruiting rankings stats, but I do submit that player developement goes hand in hand with those stats.

 

[tm3]

True that player development plays a role.

However there is data in another post around here noting that each year there are about 20 5* players. About 16% of them make it to the NFL.

The majority of NFL players were rated 3*. But, there are hundreds of them and only about 2% make it to the NFL.

So while not a sure thing the odds are much, much higher that any random 5* player will make NFL over any random 3* player.

There is no substitute for horsepower.

 

[orangebloodgmc]

True that player development plays a role.

However there is data in another post around here noting that each year there are about 20 5* players. About 16% of them make it to the NFL.

The majority of NFL players were rated 3*. But, there are hundreds of them and only about 2% make it to the NFL.

So while not a sure thing the odds are much, much higher that any random 5* player will make NFL over any random 3* player.

There is no substitute for horsepower.

There was a lot of interesting info packed into that one post. :question:

 

[photovol]

True that player development plays a role.

However there is data in another post around here noting that each year there are about 20 5* players. About 16% of them make it to the NFL.

The majority of NFL players were rated 3*. But, there are hundreds of them and only about 2% make it to the NFL.

So while not a sure thing the odds are much, much higher that any random 5* player will make NFL over any random 3* player.

There is no substitute for horsepower.

or displacement

 

[daj2576]

2012 stats;

Rank:

UT ypg surrendered, #14 in SEC

The worst since before those stats were first recorded in 1950.

UT ppg surrendered, #14 in SEC
Worst in UT history since the 1890s

UT Number of sacks; #14 in SEC







Have you read 'Bringing down the house?'

The thing about recruiting statistics, although pertinent, those are somewhat subjective and not entirely scientific.

I've never seen a rating given to player developement.

Some players come in from HS as 'projects' and end up being all pro in the NFL.

Some come in as the second coming and never pan out.

Not to try to diminish your good work with the recruiting rankings stats, but I do submit that player developement goes hand in hand with those stats.

You're right. There are certainly many intangibles. I think using a four year average tends to mitigate both the booms and the busts, plus it tends to flatten out the impact of leaving early for the draft and other things.

Basically with every school being calculated as a four year average, the algorithm tends to assume that players are developed or not roughly at the same rate across the NCAA. Is that a bad assumption? Could be. I bet that if there are outliers that blow up that average they are fewer than the norm. That is just a gut feeling.

Good input though. I don't mind if we disagree, in fact that is the best way to sharpen an analysis.

 

[thebjd]

You're right. There are certainly many intangibles. I think using a four year average tends to mitigate both the booms and the busts, plus it tends to flatten out the impact of leaving early for the draft and other things.

Basically with every school being calculated as a four year average, the algorithm tends to assume that players are developed or not roughly at the same rate across the NCAA. Is that a bad assumption? Could be. I bet that if there are outliers that blow up that average they are fewer than the norm. That is just a gut feeling.

Good input though. I don't mind if we disagree, in fact that is the best way to sharpen an analysis.

Wouldn't player development show up in those that consistently perform above or below their predicted position? This is an interesting analysis. Thanks.

 

[daj2576]

Wouldn't player development show up in those that consistently perform above or below their predicted position? This is an interesting analysis. Thanks.

Very well could be, which would go towards my hypothesis that the best coaches are the ones who tend to over-perform the talent based evaluation. As I showed previously, in the SEC there is a strong correlation between talent and wins. Exceeding those expectations is truly what makes guys like Petrino and Spurrier stand out. Who knows if that means they are just better developers of talent or are just out scheming their competition. That evaluation also falls short of determining why guys like Miles and Saban are so good because as their recruiting is so strong, it is hard to exceed their predicted rankings.

 

[SSVol]

Great work on this. Very telling, IMO.

It has always been my contention that programs are built when players are coached to a higher level than as recruited. Recruiting success follows on-field success. So...coach the team to 10 spots above their talent...recruiting follows...rinse and repeat. Obviously that doesn't account for basic recruiting ability...some coaches are just better at it than others. But relatively speaking, for a given level of recruiting ability, success on the field will increase recruiting.

So I think the formula becomes something like Program Desirability * Coaching * Recruiting = Long Term Results. With a desirable program, coaches that elevate the players they have, and strong recruiting, you...um...Rise To The Top. Sorry...

Spurrier, Petrino, Kelly, and apparently CBJ, are examples of coaches that elevate their players. Given that we are a desirable program with a staff that elevates players, we *are* going to improve. Time will tell whether we continue to recruit well or not, but assuming even moderate recruiting success, I think we get there pretty soon. I find this really encouraging.

 

[daj2576]

Great work on this. Very telling, IMO.

It has always been my contention that programs are built when players are coached to a higher level than as recruited. Recruiting success follows on-field success. So...coach the team to 10 spots above their talent...recruiting follows...rinse and repeat. Obviously that doesn't account for basic recruiting ability...some coaches are just better at it than others. But relatively speaking, for a given level of recruiting ability, success on the field will increase recruiting.

So I think the formula becomes something like Program Desirability * Coaching * Recruiting = Long Term Results. With a desirable program, coaches that elevate the players they have, and strong recruiting, you...um...Rise To The Top. Sorry...

Spurrier, Petrino, Kelly, and apparently CBJ, are examples of coaches that elevate their players. Given that we are a desirable program with a staff that elevates players, we *are* going to improve. Time will tell whether we continue to recruit well or not, but assuming even moderate recruiting success, I think we get there pretty soon. I find this really encouraging.

The interesting thing to note is that Petrino only averaged recruiting in the mid thirties. If memory serves that was well within the bottom rung of the SEC over his tenure. What he did at Arkansas is nothing short of phenomenal. I will say that I do not see any data to suggest that he could break into the top tier of the SEC until he could increase his recruiting. In other words, I am pretty certain that he had topped out at Arkansas without drastically improving his "star" power. That is all just presumption based on gut feeling.

Like you, I am excited to see what Jones can do with the talent that he has already on the roster. I would like to see him bring in a very solid class this year, but that looks like an almost impossible task given where we sit currently. If he could miraculously close with a top 15 class, anyone who wears orange should be elated.