Tale of the Tape: Vols Stats vs. Vandy and Kentucky

[CommodoreCapital]

I would be interested in the stats versus SEC competition. That will be a better comparison since the level of competition would be similar. Auburn / Texas AM would be about a wash, but UT would have had a tougher game Alabama / Ole Miss.

Vanderbilt passing stats are slightly skewed because ACS was on track to throw for over 3000 yards and Robinette is no where near the passer ACS is.

The main stat that stands out to me is points scored in SEC play 29 Vanderbilt to 17 UT. Both defenses have not been very good, though Vanderbilts defense has started to create turnovers and Tennessees turnovers mostly came in the first 3 games. First team to 25 wins and I don't think UT has the offense to score that many.

Bye week could not have came at a better time for the Vols for both Vanderbilt (2-0) and Tennessee (1-0)are undefeated coming off a bye this year.

 

[daj2576]

We have 2 wks for our QB and receivers to get on the same page. VU will sell out to stop the run. We must be able to throw and stop Tiny and Bullard from jumping offsides at critical times.

Great point. Look at it like this: When UT takes the field against Vandy, Dobbs and his receivers will have been getting first team reps for twice as long as when we saw them last.

Between his first full start and his second, didn't his completion rate go from low 50% to low 60%? Don't quote me on that, can someone verify?

EDIT: I checked. Dobbs was 62% against Mizzou (26 of 42, 2 INT), and 64% (16 of 25, 1 INT) against Auburn. Has completed right at 59% of all passes this year. I don't like the INT rate though.

Of 67 passes he has thrown as a starter, 42 were caught by our team and 3 by the other team. So every 100 times the ball leaves his hand it should be caught by our guys 63 times, it will be dropped or miss the target 32 times and completed to the other team 5 times. I guess when I look at it like that, it is a little more confidence inspiring.

 

[Rifleman]

Great point. Look at it like this: When UT takes the field against Vandy, Dobbs and his receivers will have been getting first team reps for twice as long as when we saw them last.

Between his first full start and his second, didn't his completion rate go from low 50% to low 60%? Don't quote me on that, can someone verify?

EDIT: I checked. Dobbs was 62% against Mizzou (26 of 42, 2 INT), and 64% (16 of 25, 1 INT) against Auburn. Has completed right at 59% of all passes this year. I don't like the INT rate though.

Dobbs can likely improve on that next week.

Vandy is 105th against FBS opponents in allowed completion percentage: 65.89%

(KY is 107th at 66.05%)

Tennessee is 19th at 53.28%.

This is partly because everyone spent half the season going after Cam Sutton because he's a freshman and that kid rules his island like no freshman I can remember.

 

[Rifleman]

I would be interested in the stats versus SEC competition. That will be a better comparison since the level of competition would be similar. Auburn / Texas AM would be about a wash, but UT would have had a tougher game Alabama / Ole Miss.

Vanderbilt passing stats are slightly skewed because ACS was on track to throw for over 3000 yards and Robinette is no where near the passer ACS is.

The main stat that stands out to me is points scored in SEC play 29 Vanderbilt to 17 UT. Both defenses have not been very good, though Vanderbilts defense has started to create turnovers and Tennessees turnovers mostly came in the first 3 games. First team to 25 wins and I don't think UT has the offense to score that many.

Bye week could not have came at a better time for the Vols for both Vanderbilt (2-0) and Tennessee (1-0)are undefeated coming off a bye this year.

Any chance Carta-Samuels makes it back for the Tennessee game?

 

[CommodoreCapital]

I have no idea. The coaching staff has been so hush-hush about him. He dressed for the Florida game and was actually throwing balls before the game. I think he may have that injury of the leg (mcl, acl, whatever it is) that apparently you can do no more damage to it and since this will be his last year of organized football he may suit and play. Franklin does not give out much on the injury front but would not rule out his return.

 

[Rifleman]

I have no idea. The coaching staff has been so hush-hush about him. He dressed for the Florida game and was actually throwing balls before the game. I think he may have that injury of the leg (mcl, acl, whatever it is) that apparently you can do no more damage to it and since this will be his last year of organized football he may suit and play. Franklin does not give out much on the injury front but would not rule out his return.

Thanks for sharing what you know, CC. Sounds as though we should prepare to see either QB, as I would not rule out ACS. He's a gamer, for sure.

 

[daj2576]

I would be interested in the stats versus SEC competition. That will be a better comparison since the level of competition would be similar. Auburn / Texas AM would be about a wash, but UT would have had a tougher game Alabama / Ole Miss.

Vanderbilt passing stats are slightly skewed because ACS was on track to throw for over 3000 yards and Robinette is no where near the passer ACS is.

The main stat that stands out to me is points scored in SEC play 29 Vanderbilt to 17 UT. Both defenses have not been very good, though Vanderbilts defense has started to create turnovers and Tennessees turnovers mostly came in the first 3 games. First team to 25 wins and I don't think UT has the offense to score that many.

Bye week could not have came at a better time for the Vols for both Vanderbilt (2-0) and Tennessee (1-0)are undefeated coming off a bye this year.

If you look at recruiting averages, a great predictor of outcome, Auburn and aTm are nowhere near the same caliber (by the way, both are performing about where talent predicts). Auburn is closest to Bama, aTm is closer to SCAR. And you are right Ole Miss and Bama do not compare either.

If you look at common opponents who were constituted the same when either UT or Vandy played them (APSU, Mizzou, SCAR), UT won 2 of the 3 to Vandy's 1 of 3. Vandy scored 91 points to UT's 89, Vandy had 89 points scored on them to UT's 59. That is from memory, so I am close but probably a little off.

 

[CommodoreCapital]

Auburn and Texas AM are very comparable. Both can score at will and neither can stop anyone on defense.

Interesting you left the Georgia game out and want to add a meaningless Austin Peay game in which UT flexed their muscles.

 

[Rifleman]

Auburn and Texas AM are very comparable. Both can score at will and neither can stop anyone on defense.

Interesting you left the Georgia game out and want to add a meaningless Austin Peay game in which UT flexed their muscles.

I don't think it changes so much if you drop Peay and add Georgia that it is definitive either way.

daj2576 was just generously trying to answer the question you posited, anyway.

Honestly, even with all the stats in the world at my disposal, I really have no idea how this game is going to go. I think it really depends upon the mental composure and passion the Vols bring to the game.

The night atmosphere at Neyland may help us win by more than we would normally, or we may just be mentally fatigued and not ready to answer the bell like last year. I know the Vols have the potential to play a great game, but I've learned that we still aren't ready to embrace that potential consistently.

In short, our best game beats Vandy's best, but our lowest low will lose easily even if Vandy plays as bad as they possibly can. That is where we are.

 

[daj2576]

Auburn and Texas AM are very comparable. Both can score at will and neither can stop anyone on defense.

Interesting you left the Georgia game out and want to add a meaningless Austin Peay game in which UT flexed their muscles.

You mean the UGA game where you played a team with about 8 players absent from when we played them, or roughly 1/3 of the total starting line up? Basically you played the same defense. I said "similarly constituted" teams.

Auburn and Texas A&M aren't comparable just because you want them to be and see that they both can score. That doesn't make Auburn less talented or A&M more talented.

How many points did UT score after half time on APSU? UT called the dogs off. I don't think that game was about flexing muscles as much as you want to believe. Was Franklin just being nice to APSU?

 

[CommodoreCapital]

Yes that game. We can talk about injuries all year long on either team and the fact doesn't change that Vanderbilt ran a freshman QB out there for the 2nd half for his first significant snaps and he WON the game.

Auburn beat Texas A/M by 3 this year in a 44-41 game. So why do you say they are not comparable since you just want to tell me they aren't because I said they are.

The recruiting line is just a small piece of the puzzle and I would take coaching over recruiting any day. Example, Auburn last year (by the way Vandy beat Auburn) does that make then comparable.

 

[daj2576]

Yes that game. We can talk about injuries all year long on either team and the fact doesn't change that Vanderbilt ran a freshman QB out there for the 2nd half for his first significant snaps and he WON the game.

Auburn beat Texas A/M by 3 this year in a 44-41 game. So why do you say they are not comparable since you just want to tell me they aren't because I said they are.

The recruiting line is just a small piece of the puzzle and I would take coaching over recruiting any day. Example, Auburn last year (by the way Vandy beat Auburn) does that make then comparable.

Mathematically recruiting accounts for about 70% of the wins at large, 80% of the wins in the SEC, and about 90% of the wins in the BCS title game.

When you start talking about the times that recruiting doesn't tell the whole story, the minority, you do find some coaches who are isolated.

Dooley/Chizik/Muschamp/Kiff, all significantly under perform compared to talent. On the flip side of that coin are coaches who over-perform. Yes Franklin is on that list at +2 this year, Petrino at Arkansas over-performed at a rate of about 4 games a season. Jones at Cincy over-performed about about 2-3 games a season on average. There are others but that is the general explanation.

Of course, where you sit, you believe that coaching is the preferable course of action. That is exactly what you see, but that only happens in a minority of times. You can't take that data and extrapolate out from, my dog has fleas, so all dogs have fleas.

I caution you against using the transitive property to draw conclusions between Auburn and aTm by that heads up game alone. If you use the transitive property, you end up with this: Mizzou beat UT who beat SCAR who beat Mizzou. Doesn't work. Although, 2 out of 3 of those games are accounted for by talent averages.

If you have series of events where teams played similarly constituted teams, you can discover trends. Which is the reason that UGA and UF are discounted when discussing common opponents between UT and Vandy. I know, you think I am leaving them out because it helps my argument. Although I could say that you are leaving them in, not because they were similar teams (besides wearing similar uniforms), but because it only helps your argument.

The hardest thing for me to grasp, and I have been studying this for over a year now, is that talent averages account for wins/losses, but not for score differentials. In other words, sometimes teams who have a gargantuan talent average disparity have a tight game, and sometimes teams who are evenly talented are blow outs. I don't see that as being important. All that matters is that talent accounts for roughly 70/80/and 90% of the outcomes of games depending upon the setting.

Here is my chart of teams ranked by talent averages in the SEC and how that is measuring up against actual outcome. https://docs.google.com/a/mybloodis...Qgwl-hyfdEpwUHpyWXUzY3JWRFU1Skc1UTRiZ2c&gid=0

I realize that it probably isn't intuitive so there is an instructions tab on top. Its a few weeks old so the examples given are out of date, but it should help explain.

EDIT: To your point about a freshman QB winning a game, I will add something I read. Dave Barton (cfbmatrix.com) provided me with some numbers that suggest that, contrary to fan fiction, the difference between an experienced an inexperienced QB is about .2 games a year. Yes...point two. Not two. According to Bartoo's figures replacing a kicker accounts for a bigger fall off in outcome. Interesting stuff.

 

[BaldBiker]

Mathematically recruiting accounts for about 70% of the wins at large, 80% of the wins in the SEC, and about 90% of the wins in the BCS title game.

When you start talking about the times that recruiting doesn't tell the whole story, the minority, you do find some coaches who are isolated.

Dooley/Chizik/Muschamp/Kiff, all significantly under perform compared to talent. On the flip side of that coin are coaches who over-perform. Yes Franklin is on that list at +2 this year, Petrino at Arkansas over-performed at a rate of about 4 games a season. Jones at Cincy over-performed about about 2-3 games a season on average. There are others but that is the general explanation.

Of course, where you sit, you believe that coaching is the preferable course of action. That is exactly what you see, but that only happens in a minority of times. You can't take that data and extrapolate out from, my dog has fleas, so all dogs have fleas.

I caution you against using the transitive property to draw conclusions between Auburn and aTm by that heads up game alone. If you use the transitive property, you end up with this: Mizzou beat UT who beat SCAR who beat Mizzou. Doesn't work. Although, 2 out of 3 of those games are accounted for by talent averages.

If you have series of events where teams played similarly constituted teams, you can discover trends. Which is the reason that UGA and UF are discounted when discussing common opponents between UT and Vandy. I know, you think I am leaving them out because it helps my argument. Although I could say that you are leaving them in, not because they were similar teams (besides wearing similar uniforms), but because it only helps your argument.

The hardest thing for me to grasp, and I have been studying this for over a year now, is that talent averages account for wins/losses, but not for score differentials. In other words, sometimes teams who have a gargantuan talent average disparity have a tight game, and sometimes teams who are evenly talented are blow outs. I don't see that as being important. All that matters is that talent accounts for roughly 70/80/and 90% of the outcomes of games depending upon the setting.

Here is my chart of teams ranked by talent averages in the SEC and how that is measuring up against actual outcome. https://docs.google.com/a/mybloodis...Qgwl-hyfdEpwUHpyWXUzY3JWRFU1Skc1UTRiZ2c&gid=0

I realize that it probably isn't intuitive so there is an instructions tab on top. Its a few weeks old so the examples given are out of date, but it should help explain.

EDIT: To your point about a freshman QB winning a game, I will add something I read. Dave Barton (cfbmatrix.com) provided me with some numbers that suggest that, contrary to fan fiction, the difference between an experienced an inexperienced QB is about .2 games a year. Yes...point two. Not two. According to Bartoo's figures replacing a kicker accounts for a bigger fall off in outcome. Interesting stuff.

Wow. I feel really stupid after reading that post. Good stuff.

 

[BigOrangeTrain]

Mathematically recruiting accounts for about 70% of the wins at large, 80% of the wins in the SEC, and about 90% of the wins in the BCS title game.

When you start talking about the times that recruiting doesn't tell the whole story, the minority, you do find some coaches who are isolated.

Dooley/Chizik/Muschamp/Kiff, all significantly under perform compared to talent. On the flip side of that coin are coaches who over-perform. Yes Franklin is on that list at +2 this year, Petrino at Arkansas over-performed at a rate of about 4 games a season. Jones at Cincy over-performed about about 2-3 games a season on average. There are others but that is the general explanation.

Of course, where you sit, you believe that coaching is the preferable course of action. That is exactly what you see, but that only happens in a minority of times. You can't take that data and extrapolate out from, my dog has fleas, so all dogs have fleas.

I caution you against using the transitive property to draw conclusions between Auburn and aTm by that heads up game alone. If you use the transitive property, you end up with this: Mizzou beat UT who beat SCAR who beat Mizzou. Doesn't work. Although, 2 out of 3 of those games are accounted for by talent averages.

If you have series of events where teams played similarly constituted teams, you can discover trends. Which is the reason that UGA and UF are discounted when discussing common opponents between UT and Vandy. I know, you think I am leaving them out because it helps my argument. Although I could say that you are leaving them in, not because they were similar teams (besides wearing similar uniforms), but because it only helps your argument.

The hardest thing for me to grasp, and I have been studying this for over a year now, is that talent averages account for wins/losses, but not for score differentials. In other words, sometimes teams who have a gargantuan talent average disparity have a tight game, and sometimes teams who are evenly talented are blow outs. I don't see that as being important. All that matters is that talent accounts for roughly 70/80/and 90% of the outcomes of games depending upon the setting.

Here is my chart of teams ranked by talent averages in the SEC and how that is measuring up against actual outcome. https://docs.google.com/a/mybloodis...Qgwl-hyfdEpwUHpyWXUzY3JWRFU1Skc1UTRiZ2c&gid=0

I realize that it probably isn't intuitive so there is an instructions tab on top. Its a few weeks old so the examples given are out of date, but it should help explain.

EDIT: To your point about a freshman QB winning a game, I will add something I read. Dave Barton (cfbmatrix.com) provided me with some numbers that suggest that, contrary to fan fiction, the difference between an experienced an inexperienced QB is about .2 games a year. Yes...point two. Not two. According to Bartoo's figures replacing a kicker accounts for a bigger fall off in outcome. Interesting stuff.

Excellent post. It seems you have used logic and stats to destroy commodores argument. Congrats :good!:

 

[daj2576]

Wow. I feel really stupid after reading that post. Good stuff.

Don't feel stupid. I accidently stumbled on this while trying to figure out if we were being too harsh on Dooley.

We weren't.

 

[Volosaurus rex]

Daj2576, with respect to your statement that "talent averages account for wins/losses, but not for score differentials," I would suggest that one factor accounts for the variability inherent in that phenomenon. Unfortunately, it is not empirically measureable although it most definitely is observable. I refer, of course, to momentum. Football is such a game of emotion that, when one of two evenly matched teams clearly seizes the momentum, particularly in a home game, it snowballs, often resulting in a rout that does not reflect relative talent levels accurately. Momentum can, of course, be established in any number of ways: an 80-yard pass to Joey Kent "on play number one," a kickoff return in which Willie Gault "runs all the way to the state capitol," or a blocked field goal attempt that "Floyd Miley can fly" all the way for a touchdown against Notre Dame.

Momentum can, of course, be diverted by an equally effective counter-punch, but, if it snowballs, it can truly be transformed into a runaway freight train. We obviously were not 28 points better than Miami in the 1986 Sugar Bowl but, once we seized the momentum, we continued to stoke the furnace with big play after big play and, as a result, played with the ferocity of eleven grizzly bears.

Momentum can be a capricious taskmaster and it certainly can not be predicted prior to the game, but I would submit that it accounts for, on a game-by-game basis, much of the score differentials you cited.

 

[Rifleman]

Love the work, daj2576! Another thing that occurs to me is that recruiting and coaching are not really exclusive of each other.

Nick Saban has been known to say, for example, that he is an average game coach, but the best recruiter in the country. This makes sense. If you have the best talent, then you do not need to get that talent to overperform. If Bama plays to their basic potential, they win a national championship.

Another coach may be outperforming his recruiting rankings year in and year out by finding the perfect recruits that fit their particular system. So they are not assembling just any old group of 3-stars recruits, but getting the 3-stars recruits best suited to their system physically and mentally.

I think this aspect of recruiting is a big part of Coach Franklin's success at Vanderbilt. I think Coach Jones has also been great at finding talent at previous stops. Hopefully, he is an above-average game coach as well so we can have the best of both worlds.

 

[daj2576]

Love the work, daj2576! Another thing that occurs to me is that recruiting and coaching are not really exclusive of each other.

Nick Saban has been known to say, for example, that he is an average game coach, but the best recruiter in the country. This makes sense. If you have the best talent, then you do not need to get that talent to overperform. If Bama plays to their basic potential, they win a national championship.

Another coach may be outperforming his recruiting rankings year in and year out by finding the perfect recruits that fit their particular system. So they are not assembling just any old group of 3-stars recruits, but getting the 3-stars recruits best suited to their system physically and mentally.

I think this aspect of recruiting is a big part of Coach Franklin's success at Vanderbilt. I think Coach Jones has also been great at finding talent at previous stops. Hopefully, he is an above-average game coach as well so we can have the best of both worlds.

I have never heard him say that, but that has long been my assertion. Saban gets out-coached about once a season with a roster that should never lose a game. The spread/read-option gives him absolute fits. He is, by any measure, an absolute prodigy when it comes to recruiting. When I say that though, I am told that I should not say such things about his highness. Although, I am never sure why anyone feels the need to speak with such reference about a man who has to wear lifts in his shoes.

Sorry for hijacking your thread, my friend. I really enjoy reading your stuff. You are a kindred spirit.

 

[daj2576]

Daj2576, with respect to your statement that "talent averages account for wins/losses, but not for score differentials," I would suggest that one factor accounts for the variability inherent in that phenomenon. Unfortunately, it is not empirically measureable although it most definitely is observable. I refer, of course, to momentum. Football is such a game of emotion that, when one of two evenly matched teams clearly seizes the momentum, particularly in a home game, it snowballs, often resulting in a rout that does not reflect relative talent levels accurately. Momentum can, of course, be established in any number of ways: an 80-yard pass to Joey Kent "on play number one," a kickoff return in which Willie Gault "runs all the way to the state capitol," or a blocked field goal attempt that "Floyd Miley can fly" all the way for a touchdown against Notre Dame.

Momentum can, of course, be diverted by an equally effective counter-punch, but, if it snowballs, it can truly be transformed into a runaway freight train. We obviously were not 28 points better than Miami in the 1986 Sugar Bowl but, once we seized the momentum, we continued to stoke the furnace with big play after big play and, as a result, played with the ferocity of eleven grizzly bears.

Momentum can be a capricious taskmaster and it certainly can not be predicted prior to the game, but I would submit that it accounts for, on a game-by-game basis, much of the score differentials you cited.

You make some great points, and I wish I could totally agree. Alas, I cannot agree totally. I can only tell you this: I have seen and been involved in very detailed evaluations that account for incredibly specific talent traits, and how those traits predict success when pitted against each other. The end result is that, if you know what to look for, and have the time and money to invest in sussing that out of the rosters, that you can predict score differentials with some repeatable certainty in limited situations. I say "limited situations" because the evaluations look for specific enough traits that not every game will have teams with rosters that have those traits.

You just can't do anything like that with these simple talent average evaluations.

Of course, I can't show you any of those as I do not own them, and the owner would not be happy with me publishing data that was very expensive and time consuming to gather and collate. I understand how this must sound. I am basically telling you something that sounds so spectacular that I might as well tell you that I have proof of aliens. Aliens!...but sorry, I can't show you. :neener2:

Yup, aliens...you just have to believe me.

 

[Rifleman]

See original post for updated stats. Last week's featherweight bout between our two final opponents did not change much at all. Vandy got a lot of INTs, so they moved up in categories affected by that stat.