The SEC v. Recruiting Averages

[daj2576]

Back again...

I thought you guys might like to see how recruiting evaluations are predicting the performance of all of the SEC teams (see chart below).

This chart shows the SEC broken down into divisions (West on top, East on bottom). Each team is contained within it's own rectangle. The yellow highlighted team is the subject team for that rectangle. The schedule is ranked by talent averages. If a team is white, the yellow (subject) team has not played them yet. If a team is green, the yellow team beat them. If the team is red, the yellow team lost to them.

As a general rule, the teams above the yellow team should be red, and the teams below green. This is because (say it with me) roughly 70% of the games are predicted by talent averages.

Here are some things that jump out to me in SEC play.

1) Bama, Arkansas, Miss. State, Texas A&M, Kentucky, SCAR, LSU and Tennessee have performed in the SEC exactly as talent would predict. LSU has lost games to UGA and Ole Miss that they should have won, but won games against Florida and Auburn that they should have lost. That is a wash and insofar as this evaluation is concerned, means that LSU is tracking along the arc of talent predictions.

2) Teams exceeding expectations: Ole Miss (+1), Missouri (+2) and Vandy (+1). Let's be honest here, both Missouri and Vandy have benefited from Tennessee's game of attrition against UGA. Tennessee's game against UF might have also changed the Gator's season trajectory with the injury to Driskel. You're welcome Mizzou and Vandy. We will see you in a few weeks.

3) Teams under-performing: Auburn (-1), Florida (-2), and Georgia (-1). With Florida's issues, Georgia's Richt is likely to win that game and could finish the season as usual (talent +0). Muschamp, on the other hand, has a history of being the SEC's biggest under-performer (now that Chizik and Dooley are gone). I don't expect him to be at UF in 2015.

4) First year coaches' performance: Arkansas (+0), Auburn (-1), Kentucky (+0) and Tennessee (+0). A closer look at those numbers reveals that both Arkansas and Kentucky have bottom rung talent in their division, meaning that they could win no games and meet expectations (it's tough to fall below the floor). That is different than Tennessee's Jones who is not only meeting expectations but who has had to beat a ranked team to do so. Jones will probably have to win against a ranked Mizzou to end the season as a +0 coach. Math says we beat them, just as math said that we beat SCAR.

The best news is that we have definitive proof that Jones can drive the Ferrari he has been given. We haven't had a coach who could do that since sometime in the early 2000's. Wait until next season when Jones' system and players begin to affirm his history of being a talent +2 to +3 game coach.

Sleep soundly Vol nation. We are still right where we should be. We are in striking distance of a 7-5 season and an upset or two isn't out of the question.

As the season progresses, I will keep it updated here--> https://docs.google.com/spreadsheet/ccc?key=0AkwyQgwl-hyfdEpCdTBVRldVaS1HT3Vnczh4SFdEZFE&usp=sharing

Also, it is Bama hate week and here is a brief history of the UT v. Bama rivalry: If I Bleed Orange, I Am Bled Out: Hug a 'Bama fan today.

 

[Crazy Ivan]

The best news is that we have definitive proof that Jones can drive the Ferrari he has been given.

Huh? Our talent level is the equivalent of a Ferrari? I thought Butch inherited a Dodge Dart?

 

[BennisDrewer]

How is Arky's talent considered that bad? They're only a couple years removed from Petrino and being ranked in the top 5....did they suffer that much attrition from his departure, graduation, etc?

P.S.
My reaction when reading any of your posts

 

[daj2576]

The best news is that we have definitive proof that Jones can drive the Ferrari he has been given.

Huh? Our talent level is the equivalent of a Ferrari? I thought Butch inherited a Dodge Dart?

Butch inherited a team with solid talent, contrary to what many people want to believe after seeing Dooley handle the team for three years. At Cincy, Butch was driving something closer to a Dart.

The talent on this team is capable of predicting a 7-5 finish and predicted a win over SCAR. If that is a Dart, then so be it. I am not overly attached to that metaphor.

 

[daj2576]

How is Arky's talent considered that bad? They're only a couple years removed from Petrino and being ranked in the top 5....did they suffer that much attrition from his departure, graduation, etc?

P.S.
My reaction when reading any of your posts

Arky's talent hasn't really changed that much since Petrino left. Petrino was a coach who over-performed on a level of 4 games a year. Last year, Arky performed exactly as talent predicted and they ran their coach out of town on a rail. Until now, we at UT haven't had a coach that could simply perform as talent predicted since arguably 2001.

 

[BanditVol]

OP....I did some digging and noticed that we actually out-talent Oregon in recruiting rankings. Their average Rivals ranking was a 17, I believe.

But I'm not worried about that loss, believe me. They have been well coached and have honed their system to an art.

But it does suggest that they are a bit overated, yes?

 

[MockingVol]

How is Arky's talent considered that bad? They're only a couple years removed from Petrino and being ranked in the top 5....did they suffer that much attrition from his departure, graduation, etc?

P.S.
My reaction when reading any of your posts

Arkansas, even during Petrino's tenure, was never a very highly "talented" team. They have never recruited well. Petrino recruited guys that fit into his system and developed them well.

 

[KBVol]

The best news is that we have definitive proof that Jones can drive the Ferrari he has been given.

Huh? Our talent level is the equivalent of a Ferrari? I thought Butch inherited a Dodge Dart?

'78 Gremlin IMO. Omg, don't let stj18 see this.... gonna be insufferable

 

[SmokeyX]

Awesome post and analysis OP

 

[HollowaytoSeivers]

The best news is that we have definitive proof that Jones can drive the Ferrari he has been given.

Huh? Our talent level is the equivalent of a Ferrari? I thought Butch inherited a Dodge Dart?

Perhaps, but so far CBJ and his staff have shown themselves to be pretty fair mechanics.

 

[VolFan4Life87]

Butch inherited a team with solid talent, contrary to what many people want to believe after seeing Dooley handle the team for three years. At Cincy, Butch was driving something closer to a Dart.

The talent on this team is capable of predicting a 7-5 finish and predicted a win over SCAR. If that is a Dart, then so be it. I am not overly attached to that metaphor.

1) does you analysis look at recruiting rankings only or does it factor in players no longer with the team for various reasons and generate deductive scores accordingly

2) math cannot factor in that while Jones inherited whatever talent level he did, that talent level was never coached to its potential during the previous coaching regime.

Always like your posts, I'm an engineer so I love statistical analysis of things.

 

[Special Ed]

Easy on the Dodge Dart. I bought my daughter one because it's one of the safest cars on the road.

 

[BanditVol]

The best news is that we have definitive proof that Jones can drive the Ferrari he has been given.

Huh? Our talent level is the equivalent of a Ferrari? I thought Butch inherited a Dodge Dart?

We have pretty decent talent. I have to admit, I was surprised that our average national ranking was 15 per Rivals, which is what the OP is using.

However, it's worth pointing out that this is behind bammer (2), Florida (5), Auburn (7-8), LSU (9) and Uga (11), or only 6th in the SEC.

So you could say that we "shudda cudda" been a middle of the pack team with Dooley, but that now we are with Jones, but we have the potential to be in the top group if the current recruiting success continues.

 

[daj2576]

OP....I did some digging and noticed that we actually out-talent Oregon in recruiting rankings. Their average Rivals ranking was a 17, I believe.

But I'm not worried about that loss, believe me. They have been well coached and have honed their system to an art.

But it does suggest that they are a bit overated, yes?

My numbers show that we had the same recruiting average over 4 years. My numbers also show that when teams are closely matched, the home team wins. Seems obvious enough, right? I thought that game would be closer but have since learned that these numbers don't always predict the spread, just the likelihood of a W or an L.

My numbers also show that Oregon performs precisely as their talent average predicts, in other words, they recruit as well as UT but can typically out recruit every other PAC team besides USC. USC, under Kiff, was a non-entity on the field, but an extremely talented team.

 

[daj2576]

We have pretty decent talent. I have to admit, I was surprised that our average national ranking was 15 per Rivals, which is what the OP is using.

However, it's worth pointing out that this is behind bammer (2), Florida (5), Auburn (7-8), LSU (9) and Uga (11), or only 6th in the SEC.

So you could say that we "shudda cudda" been a middle of the pack team with Dooley, but that now we are with Jones, but we have the potential to be in the top group if the current recruiting success continues.

Talent predicted a 9-3 season last year. Thanks Dooley!

Here are some interesting charts showing our recruiting v. the SEC and our SEC wins over time. Notice a correlation?

 

[daj2576]

1) does you analysis look at recruiting rankings only or does it factor in players no longer with the team for various reasons and generate deductive scores accordingly

2) math cannot factor in that while Jones inherited whatever talent level he did, that talent level was never coached to its potential during the previous coaching regime.

Always like your posts, I'm an engineer so I love statistical analysis of things.

Attrition is always brought up in this discussion, but it is basically a non-entity.

Here are the SEC east rosters after the spring game (in other words after all of the attrition of the draft, etc). Basically matches the recruiting average rankings.

 

[VolFan4Life87]

Talent predicted a 9-3 season last year. Thanks Dooley!

Here are some interesting charts showing our recruiting v. the SEC and our SEC wins over time. Notice a correlation?

9-3 was my preseason prediction last year. It absolutely should've happened, thanks Doofus indeed

 

[BanditVol]

Talent predicted a 9-3 season last year. Thanks Dooley!

Here are some interesting charts showing our recruiting v. the SEC and our SEC wins over time. Notice a correlation?

Good stuff man! Thanks for posting.

 

[Volunteer_Kirby]

My numbers show that we had the same recruiting average over 4 years. My numbers also show that when teams are closely matched, the home team wins. Seems obvious enough, right? I thought that game would be closer but have since learned that these numbers don't always predict the spread, just the likelihood of a W or an L.

My numbers also show that Oregon performs precisely as their talent average predicts, in other words, they recruit as well as UT but can typically out recruit every other PAC team besides USC. USC, under Kiff, was a non-entity on the field, but an extremely talented team.

I know you like to use "recruiting average" as a good indicator to "predict" games, but it really carries no relevance when half of our 4 stars, which originally beefed up our "recruiting average", failed to sign in the first place or are no longer with the team. See 2010. Attrition over the last few years really has been at an all time low, so do you factor that into these numbers at all? Would be interesting to see what the RATED starting two deep (of say Missouri or Auburn vs Tennessee) is vs the inflated recruiting average. Seems like it might yield even better / more accurate results.

 

[Crazy Ivan]

Perhaps, but so far CBJ and his staff have shown themselves to be pretty fair mechanics

I will definitely agree with that. I will also agree with the OP that the cupboard was not bare for CBJ, at least at certain positions. But we are hard-pressed to find much SEC talent behind the starters at most positions. The Dodge Dart assessment may be a bit unfair on my part, but at the same time, I just don't see the Ferrari analogy. If we are applying it to Bama's talent, then yes. I do believe Jones will end up building us a Ferrari, but we are 3-4 years away from that IMO.

 

[daj2576]

I know you like to use "recruiting average" as a good indicator to "predict" games, but it really carries no relevance when half of our 4 stars, which originally beefed up our "recruiting average", failed to sign in the first place or are no longer with the team. See 2010. Attrition over the last few years really has been at an all time low, so do you factor that into these numbers at all? Would be interesting to see what the RATED starting two deep (of say Missouri or Auburn vs Tennessee) is vs the inflated recruiting average. Seems like it might yield even better / more accurate results.

See below (or above as I have already posted this once). Also understand that this does not count the 2013 signing class, as this was the roster AFTER THE SPRING GAME. Attrition is always viewed as some mitigating factor, but it appears that as I have long suggested, it really changes nothing and all schools have attrition that is relatively similar. We just notice ours a lot more.

 

[daj2576]

Perhaps, but so far CBJ and his staff have shown themselves to be pretty fair mechanics

I will definitely agree with that. I will also agree with the OP that the cupboard was not bare for CBJ, at least at certain positions. But we are hard-pressed to find much SEC talent behind the starters at most positions. The Dodge Dart assessment may be a bit unfair on my part, but at the same time, I just don't see the Ferrari analogy. If we are applying it to Bama's talent, then yes. I do believe Jones will end up building us a Ferrari, but we are 3-4 years away from that IMO.

If we are driving a ferrari, then Bama is driving an Indy Car. There are always useful analogies, if you don't get too caught up in trying to prove/disprove what is essentially a visual aid.

 

[Rifleman]

See below (or above as I have already posted this once). Also understand that this does not count the 2013 signing class, as this was the roster AFTER THE SPRING GAME. Attrition is always viewed as some mitigating factor, but it appears that as I have long suggested, it really changes nothing and all schools have attrition that is relatively similar. We just notice ours a lot more.

This makes a lot of sense to me. Recruiting ranks is a great base variable to build a prediction model on. Obviously our coaches would be knocking around the hallways of every high school and Juco from coast to coast otherwise.

I also agree we suffer more from poorly developed talent than a bare cupboard.

To that end:

What, if any, difference do you think seniority of roster makes? For example, if team A and team B both have roughly 3-stars average rosters, but A's starters are mostly freshman and B's starters are mostly seniors?

Does team B usually win, or does it remain a coin toss?

 

[Sheik Yerbouti]

...

As the season progresses, I will keep it updated here--> https://docs.google.com/spreadsheet/ccc?key=0AkwyQgwl-hyfdEpCdTBVRldVaS1HT3Vnczh4SFdEZFE&usp=sharing

Also, it is Bama hate week and here is a brief history of the UT v. Bama rivalry: If I Bleed Orange, I Am Bled Out: Hug a 'Bama fan today.

I always enjoy your analysis; thank you for posting.

Two slight quibbles.

First, it's hard for me to count this evidence as "definitive proof." I'd say, "the evidence so far supports the conclusion that our head coach has the team on track," or something like that.

Second, from a readability standpoint, I'd like to see these charts reorganized in descending order of talent from left to right, rather than in alphabetical order. Seeing the yellow talent line step down from left to right may make the chart a bit more intuitive to understand. I recognize, of course, that this would require you to do some potentially annoying cut-and-paste work. I'd just like to see it the other way.

Cheers! Nice work. eace2:

 

[Rifleman]

If we are driving a ferrari, then Bama is driving an Indy Car. There are always useful analogies, if you don't get too caught up in trying to prove/disprove what is essentially a visual aid.

I like it! I think Oregon was in the SR-71 Blackbird the last time we crossed paths.