National College FB Roster Rankings going into 2014

[golfballs]

it also doesn't differentiate by rankings by position which is even more important.

 

[LordVOLdemort]

Bama also has a lot of attrition. I dont know if it's just kids jumping ship that dont move up the roster, get recruited over, or what, but I cant tell you the number of 4 star kids (and some 5's) that were in their recruiting classes and not on their roster anymore.

Theyve got 106 (21 being squad or walk-on) on their roster, some guys going all the way back to 2010 class, avg 25 a class, that's 125 recruited - 85, so about 40 guys signed LOIs and arent on the roster now.

Im just guessing tho since i really didnt count as I went along.

 

[orangeluvr]

Hmmm. Very interesting premise, but the actual methodology is a bit crude. Averaging the team recruiting class rankings for the last five years doesn't account for all of the subsequent transfers, dismissals or early entrees that can deplete a class. A far more accurate approach would be something like adding up the sum number of Rivals stars or ESPN number grades for every roster.

That's refreshing, some sound reasoning on here. But don't get to carried away! :birgits_giggle:

 

[VolsnRavens]

I agree. I would like to see where we stand if one compares the players still on the teams rank. After all, this years team IS composed of those who remain from those classes and not those who have dropped off the team for one or more reasons. The way they have compared the teams is like comparing the total income of your ex spouses with the spouse you just married.

Ummm, how many ex spouses do you have to be comparing thier inc....wait. Hef, is this you?

 

[thebjd]

Why are people not pleased with Athlon's assessment? The days of being downtrodden are coming to an end. Is this battered fan syndrome?

 

[DERELICK_MY_VOLS]

It's good to be number 13.
Everyone else is afraid of it.

Oh yeah...then why is 6 afraid of 7?




Cuz 7-8-9.

 

[bamawriter]

Why are people not pleased with Athlon's assessment? The days of being downtrodden are coming to an end. Is this battered fan syndrome?

Speaking as a fan of the number 1 team on that list, someone who should be extremely pleased with the result, Athlon's "assessment" is based on extremely flawed methodology.

 

[thebjd]

Speaking as a fan of the number 1 team on that list, someone who should be extremely pleased with the result, Athlon's "assessment" is based on extremely flawed methodology.

As someone who hates Alabama football I am pleased to be on any positive list with a top twenty ranking of any kind. I don't know if hating Alabama has anything to do with being pleased but thought I would throw it in there.

 

[HuntlandVolinColo]

13 looks nice. We are at least a 5 this year; more if you count all recruits and not just the first 20.

But....one of those top 10 classes was a total, complete bust and that won't show up in the average rankings....nor will JUCOs.

 

[Sikerth]

Still sixth in the SEC.

 

[UTodd01]

Speaking as a fan of the number 1 team on that list, someone who should be extremely pleased with the result, Athlon's "assessment" is based on extremely flawed methodology.

Speaking for all Tennessee fans, nobody cares what you think. Now GTFO.

 

[bamawriter]

Speaking for all Tennessee fans, nobody cares what you think. Now GTFO.

Nothing says "I don't care" like an angry response.

 

[jimallvol]

That really surprises me that were in the top 25

 

[jimallvol]

Speaking as a fan of the number 1 team on that list, someone who should be extremely pleased with the result, Athlon's "assessment" is based on extremely flawed methodology.

If ur a bama fan why don't u go to their forums this is volnation!

 

[bamawriter]

If ur a bama fan why don't u go to their forums this is volnation!

Because I would miss out on witty, creative posts like yours.

 

[lincolncovol]

Interesting very very interesting

 

[daj2576]

This methodology, and others similar, are the beginnings of predictive models with results around the 70% range. In other words, if you rank all teams just by those numbers without accounting for any sort of attrition and/or experience, you could predict the outcome of 7 out of every 10 games (90% championship games).

Having done this and similar evaluations, I can tell you that attrition is largely similar across all Division 1 teams (yes there are a few outliers, and yes UT is potentially one of those). But, in general, teams lose players at roughly the same rate. That is, at a rate similar enough to produce results within a range of 70% +/-.

Similarly, the idea that inexperience is a big factor is over-played (notice I didn't say not-important, I said over-played). What is more important than experience is a base line of athletic ability. Many people tend to forget that top flight recruits are coming in with strength and conditioning programs that are superior to what recruits of even a decade, or half-decade ago, had. That means that younger players at many positions are much closer to having a direct impact than many believe.

Insofar as predicting absolute outcomes (W or L) talent is, by far, the superior component of any of the variables (location, coaching, weather, etc).

Many of your ideas about adjusting rosters for attrition, thus latent talent, will not increase this rate of prediction enough to make it worth your time, if you aren't being paid to do it (your time invested will approach the tens of hundreds of hours if you do it for every team, and the rate of prediction might only go up 5-10%).

Using this data as a base-line evaluation, there are ways to get huge jumps in the predictability of games, but they are so labor intensive as to be prohibitive to anyone who does not have a lot of free capital to invest. It isn't just as simple, or as intuitive, as finding the latent talent on the two deep and comparing an offense against a defense, you have to know exactly what sort talent creates statistical mismatches before you get substantial jumps in predictability enough to begin to use data to beat the spread.

If you want to see how well this sort of simple model correlates with actual wins and losses on the field, here is a chart illustrating the SEC last season. As you can see, not only did talent predict 67% of all SEC games played, but insofar as seasonal predictions only 4 of 14 teams ended exactly as talent predicted, BUT 9 of 14 teams (64%) ended the season within one game of their predictions, and 11 of 14 teams (79%) ended within two games of their predictions.



Tennessee, for instance, underperformed by two games but, 1) decreased the under-performance from Dooley's typical 3-4 game under-performance, to just 2, and; 2)still had a prediction rate of 10 of 12 games (83%) overall, or 75% in the SEC.

Here is how this list looks for the 2014 season:



Or, if you would like to see how all of UT's 2014 schedule ranks:

 

[daj2576]

We just signed a 30 plus class with a boatload of four stars. The talent level went up a lot. The talent is young but the roster has been significantly upgraded. I expect due to the size of this past class and the start Jones has for next year we will have legitimate top ten talent on the team. The future looks good if we can get through this season in decent shape. 2016-2017 should see us legitimately in the SEC race.

Here is the effect of the 2014 signing class on the projected two deep, compared to the same for 2013, using Rivals data.

A sizable jump in talent, indeed.



Further discussion / explanation:

If I Bleed Orange, I'm Bled Out: All SEC team predictions for 2014.

If I Bleed Orange, I'm Bled Out: Don't Curb Your Enthusiasm!

If I Bleed Orange, I'm Bled Out: Maybe next year (a season wrap).

 

[Exfix]

To show comparison to our rival Alabama (the only school I've had time to do so far) based on their current roster compared with composite rankings from 247 from each player's final hs grade.

Senior - two 5 stars, eight 4 stars, four 3 stars

Junior - two 5 stars, eleven 4 stars, four 3 stars

Soph - six 5 stars, fifteen 4 stars, fourteen 3 stars

Fresh - six 5 stars, fifteen 4 stars, five 3 stars

And based on my projected data format (see above)

our number is 400.5 and bama's is 526.5

Nice work

 

[HARTUPM]

I could not find if this was already posted so here it is. Thought this was very interesting. Vols 13th ranked roster going into 2014 season. SIAP

Ranking College Football's Rosters for 2014 | AthlonSports.com

We basically have had the same record as Ole Miss, but we have brought in more talented athletes over the last 4 years... Then the Dooley factor came into play and kinda equalized everything, and negated us having better players than a lot of other teams in the country.

 

[HARTUPM]

To show comparison to our rival Alabama (the only school I've had time to do so far) based on their current roster compared with composite rankings from 247 from each player's final hs grade.

Senior - two 5 stars, eight 4 stars, four 3 stars

Junior - two 5 stars, eleven 4 stars, four 3 stars

Soph - six 5 stars, fifteen 4 stars, fourteen 3 stars

Fresh - six 5 stars, fifteen 4 stars, five 3 stars

And based on my projected data format (see above)

our number is 400.5 and bama's is 526.5

Is it possible you are double counting some players or counting players no longer in the program? Those are some big classes, I counted 92 players.

 

[BruiserVol]

This methodology, and others similar, are the beginnings of predictive models with results around the 70% range. In other words, if you rank all teams just by those numbers without accounting for any sort of attrition and/or experience, you could predict the outcome of 7 out of every 10 games (90% championship games).

Having done this and similar evaluations, I can tell you that attrition is largely similar across all Division 1 teams (yes there are a few outliers, and yes UT is potentially one of those). But, in general, teams lose players at roughly the same rate. That is, at a rate similar enough to produce results within a range of 70% +/-.

Similarly, the idea that inexperience is a big factor is over-played (notice I didn't say not-important, I said over-played). What is more important than experience is a base line of athletic ability. Many people tend to forget that top flight recruits are coming in with strength and conditioning programs that are superior to what recruits of even a decade, or half-decade ago, had. That means that younger players at many positions are much closer to having a direct impact than many believe.

Insofar as predicting absolute outcomes (W or L) talent is, by far, the superior component of any of the variables (location, coaching, weather, etc).

Many of your ideas about adjusting rosters for attrition, thus latent talent, will not increase this rate of prediction enough to make it worth your time, if you aren't being paid to do it (your time invested will approach the tens of hundreds of hours if you do it for every team, and the rate of prediction might only go up 5-10%).

Using this data as a base-line evaluation, there are ways to get huge jumps in the predictability of games, but they are so labor intensive as to be prohibitive to anyone who does not have a lot of free capital to invest. It isn't just as simple, or as intuitive, as finding the latent talent on the two deep and comparing an offense against a defense, you have to know exactly what sort talent creates statistical mismatches before you get substantial jumps in predictability enough to begin to use data to beat the spread.

If you want to see how well this sort of simple model correlates with actual wins and losses on the field, here is a chart illustrating the SEC last season. As you can see, not only did talent predict 67% of all SEC games played, but insofar as seasonal predictions only 4 of 14 teams ended exactly as talent predicted, BUT 9 of 14 teams (64%) ended the season within one game of their predictions, and 11 of 14 teams (79%) ended within two games of their predictions.

View attachment 75054

Tennessee, for instance, underperformed by two games but, 1) decreased the under-performance from Dooley's typical 3-4 game under-performance, to just 2, and; 2)still had a prediction rate of 10 of 12 games (83%) overall, or 75% in the SEC.

Here is how this list looks for the 2014 season:

View attachment 75055

Or, if you would like to see how all of UT's 2014 schedule ranks:

View attachment 75059

Are you telling me that out of all that fancy statistics we are suppose to finish 8-4 (5-3)? I'd take it! Lol.

 

[bamawriter]

Is it possible you are double counting some players or counting players no longer in the program? Those are some big classes, I counted 92 players.

That's counting every player in each of those classes. It doesn't take attrition into account.

 

[daj2576]

Are you telling me that out of all that fancy statistics we are suppose to finish 8-4 (5-3)? I'd take it! Lol.

The summary would be thus: We have an 80% chance of finishing within 2 games of 8-4. Or, another way to look at it is that we have a 70% chance of winning every game we play against teams with a lower recruiting average, however, that means that the likelihood of us winning all 8 games we should win is only about 6%. Or to look at the inverse, of the 4 games that talent says we should lose, we only have a 24% chance of losing them all.

But, here is the kicker that I didn't discuss: Over his 7 year coaching tenure, Jones has a history of over performing, in relation to talent, by an average of 3 games a year (including the -2 we saw last year).

 

[LordVOLdemort]

Is it possible you are double counting some players or counting players no longer in the program? Those are some big classes, I counted 92 players.

Maybe, but I went right off the roster marked the names off as I counted them. they have 106 ppl on their roster