I stand before you to take my lashings...

[Boca Vol]

OP, are you still claiming the talent was equal yesterday? If so, how?

 

[daj2576]

Daj, I always enjoy reading your posts and I personally believe in the 'Moneyball' approach as a numbers man myself. Though attrition works out somewhat evenly over a certain amount of time, UT falls outside of the bell curve when you consider their attrition over the last few years - simply an anomaly.

Also, there's that thing called momentum - either negative or positive. Turnovers produce it (e.g., WKU). The other team breaking several long runs and passes into the endzone so you're down almost 5 TD's at half produces it(UO). This is the kind of thing that happens on occasion that upsets all the talent numbers.

Finally, game breakers: Having a Mariota and Thomas on the same team is hard to beat. Had we had those 2 players on our team, I could see the tables turned yesterday... and that's only 2 players out of 85! Oregon matches up well with UT talent wise (with the exeption of some game breakers) but their system of play has been ingrained into them over the last decade while our guys are adjusting to yet another new coaching staff.

Agreed and thanks for the compliments. It is always refreshing to hear someone who wants to talk rationally and not just throw out inane dribble. Go Vols!

 

[Rifleman]

Laugh at me if you want. I don't care. I still believe Oregon could beat Alabama. Any day of the week. Entirely too much speed. I have found out that speed will beat size any day.

Whose laughing? Did anyone see Bama's defense against A&M? Oregon is better right now, but there is a lot of football between now and January.

 

[daj2576]

OP, are you still claiming the talent was equal yesterday? If so, how?

I am claiming that yesterday is one of the 30% of games where talent ratings do not predict outcomes. 30%!

Talent rankings failed completely for that one game yesterday (although for yesterday 80% of the other games were predicted by talent).

Anyway you boil it down, whether that game was an exception to the rule, or whether UT is an exception to the rule, what we witnessed yesterday was the exception and not the rule.

 

[Boca Vol]

I am claiming that yesterday is one of the 30% of games where talent ratings do not predict outcomes. 30%!

Talent rankings failed completely for that one game yesterday (although for yesterday 80% of the other games were predicted by talent).

Anyway you boil it down, whether that game was an exception to the rule, or whether UT is an exception to the rule, what we witnessed yesterday was the exception and not the rule.

But it didn't take a rocket scientist to realize there was a talent gap in that game. Why else was the point spread 28 points?

I don't care what recruiting numbers you put in front of me there is no way you could convince me that game was evenly matched.

 

[livefaith]

Your thoughts are all well intended, but we lost yesterday for one simple reason: I just realized I forgot to wear my lucky boxers. Sorry to let everyone down.

Way to go genius.

 

[tm3]

IMO CroKev hit the outlier nail squarely on the head.

UT has been an anomaly for several reasons, and the most telling is probably the Kiffin recruiting class that was ranked as the worst in the history of the Rivals rankings. The inflated rank of that class skews the stats significantly.

All teams have attrition, but over the last year UT lost its top 4 playmakers. Take away The Quacks' top 4, put UT's back, and things play out differently although the talent rankings would stay the same. For that matter, just swap QBs and the result may have been different.

 

[Boca Vol]

IMO CroKev hit the outlier nail squarely on the head.

UT has been an anomaly for several reasons, and the most telling is probably the Kiffin recruiting class that was ranked as the worst in the history of the Rivals rankings. The inflated rank of that class skews the stats significantly.

All teams have attrition, but over the last year UT lost its top 4 playmakers. Take away The Quacks' top 4, put UT's back, and things play out differently although the talent rankings would stay the same. For that matter, just swap QBs and the result may have been different.

I agree. You can't take the 2010 class and say that recruiting ranking would still hold true today with Bray, Hunter, and Rogers gone.

 

[Mallicoat]

Now players and coaches are exposed after a real football game was played. Some adjustments need to be made with players. Players and coaches need to be held accountable for the poor play and coaching from assessing QTB play to play calling to defensive breakdowns. This is what the head coach gets paid the big bucks. I or we saw some things stand out things that could and can be corrected. Let's see if Butch does so. No one revamped the defense when we all saw not little but huge problems with last years defensive scheme. Everybody just kept drawing those big pay checks while the fan base went in meltdown mode. If fans see credible areas of concern causing failure then why cannot the coaches.

 

[livefaith]

I can't try to justify what happened yesterday but put somethings in perspective.

Last year we gave up 721 total yards to Troy. TROY!
Say that again to yourself, TROY!

Yesterday we gave up 687 yards to Oregon. Oregon's third or fourth stringers are better than anything Troy could have put on the field, and is probably the most explosive offense ever to take a college football field, yet you act like the defense isn't showing some improvement?

It isn't like in the third or fourth quarter Oregon put fans on the field and just had them sit in a drum circle, they put back ups in and continued to run their system.

We are a disciplined football team, even in a very hostile environment we only had 4 penalties for 40 yards. Usually when teams are very out matched and out gunned (and getting the score run up on them) a lack of discipline will show. That didn't happen.

At this point last year we had 21 penalties for 151 yards. This year we have 6 for 57 yards. That in itself is telling.

I didn't see the guys quit on the coach, and I saw that last year a few times with our players (think about Missouri, Florida and Vandy games, or Kentucky the year before as an example).

We got whipped and they kept playing. I see improvement...

2576,
As strange as I feel saying it, I think you're right about the defense. It's pretty hard to say improved when you give up 600+. UO could have easily put 800+ on us had they not taken their foot off the gas for the entire 4th Q and ran scissors plays every down.
That being said, this was no Troy. We could very well be looking at the best offense in a looong time. They are not splouch on D either. Oregon is for real.
Against realistic offenses, I think we'll be significantly improved this year. With our O, we had better be.

 

[Lifesabutch]

Whether we were in flexbone, wishbone, read option, run n shoot, Wing T, spread, etc, we were going to get housed yesterday. They're just better than us....by a lot.

:lolabove: This.

I don't understand why some of you are already throwing in the towel on Coach Jones and/or his system. Oregon was ranked #2 for a reason...they are really good. Take your hat off to them and move on...or jump ship and pick a different team if you can't handle your emotions any better than to cry about play calling or personnel. This won't be the last time we lose this year, so hang on or be gone.

 

[Tnrdgerunr]

For those of you who know me or have followed my talent evaluation threads, you also know that I was one of the few who was trying to illustrate that the UT v. Oregon game, based on those evaluations, would be a close game.

Tie me up and give me my lashings. I was wrong.

Last night was an "aha!" moment for me.

I went back over some of these predictions and came to another interesting conclusion: talent can be used to predict outcomes of games (I am running about 70% this year) but it is an awful indication of score differential.


If you want to save yourself a lot of time and trouble here's a simple way to compare teams, how many on your team would start for their team and how many from their team would start for your team. Now compare your team to Fla's team, how many UT players would start for Fla? How many Fla players would start for UT? If the answer to how many of yours would start for Fla is 20% or less you're going to get beat and probably badly. If the % is 40 to 60 it'll probably be close. If it's 60% or better would start you're probably going to win. This is who should win, S--- happens.

 

[volswillrise12]

You can throw all the statistics and "Moneyball" research at this Tennessee program that you care to but the fact is the talent gap between us and Oregon is as wide as the Grand Canyon. Butch said it best yesterday in the postgame presser when asked about substituting in guys for the O-line "We don't have anyone". On the defensive line "Can you play D-line?" There are walk ons and first year players in the secondary. These problems can be traced in recruiting all the way back to Fulmer. Let's not just flush Dooleys 3 years of fail. The guy went through a recruiting cycle and recruited zero offensive linemen at one point. ZERO. We will feel that next year guys. Again, totally not surprised on how the Oregon game went---coulda run the Wing T--Worley could have made all the throws--Butch and his staff could have schemed it up--didn't matter. The game needs to be flushed to a certain extent and used as a teaching point and we all need to move on.

 

[IN Jed]

Orange pants!

when was the last meaningful game we won in orange pants on the road?

 

[daj2576]

But it didn't take a rocket scientist to realize there was a talent gap in that game. Why else was the point spread 28 points?

I don't care what recruiting numbers you put in front of me there is no way you could convince me that game was evenly matched.

You have argued for a long time about how recruiting rankings are an illegitimate basis for predicting outcomes of games. You are no less wrong then than you are now, but alas you have found the exception to the rule to try to use as proof.

To answer your question about spread: The spread is set in an effort to assure gamblers will bet roughly evenly on either side of that number. So, the number is set, and adjusted, to play to the better's perception of the actual game. This minimizes risk to the house.

Spreads are, in fact, proof that games can be easily predicted (otherwise a spread wouldn't be required) and are also proof of how poorly the average better can predict the outcome of games.

 

[butchna]

Daj, I always enjoy reading your posts and I personally believe in the 'Moneyball' approach as a numbers man myself. Though attrition works out somewhat evenly over a certain amount of time, UT falls outside of the bell curve when you consider their attrition over the last few years - simply an anomaly.

Also, there's that thing called momentum - either negative or positive. Turnovers produce it (e.g., WKU). The other team breaking several long runs and passes into the endzone so you're down almost 5 TD's at half produces it(UO). This is the kind of thing that happens on occasion that upsets all the talent numbers.

Finally, game breakers: Having a Mariota and Thomas on the same team is hard to beat. Had we had those 2 players on our team, I could see the tables turned yesterday... and that's only 2 players out of 85! Oregon matches up well with UT talent wise (with the exeption of some game breakers) but their system of play has been ingrained into them over the last decade while our guys are adjusting to yet another new coaching staff.

Great analysis!...need more of you on here!
Posted via VolNation Mobile

 

[daj2576]

I agree. You can't take the 2010 class and say that recruiting ranking would still hold true today with Bray, Hunter, and Rogers gone.

I have tried to figure out how to adjust UT's talent to sort for attrition based off of many fan's perception that UT has had an amount to be considered an outlier. What I found , contrary to what many suspect, is that we aren't all that unusual. Did you see the chart where I broke down the remaining talent on the teams in the SEC east after the spring game? That was basically the same ranking that 4 year talent averages put us in. That chart absolutely sorted for attrition, and it made little or no difference on our ranking in the SEC east.

But let's play a little hypothetical game. If Bama and Oregon play in the national championship game, and Oregon wins, does that mean that Bama does not have the most talented roster in college football, or does it mean that Oregon has found a way to utilize lessor talent to be extremely effective?

 

[daj2576]

For those of you who know me or have followed my talent evaluation threads, you also know that I was one of the few who was trying to illustrate that the UT v. Oregon game, based on those evaluations, would be a close game.

Tie me up and give me my lashings. I was wrong.

Last night was an "aha!" moment for me.

I went back over some of these predictions and came to another interesting conclusion: talent can be used to predict outcomes of games (I am running about 70% this year) but it is an awful indication of score differential.


If you want to save yourself a lot of time and trouble here's a simple way to compare teams, how many on your team would start for their team and how many from their team would start for your team. Now compare your team to Fla's team, how many UT players would start for Fla? How many Fla players would start for UT? If the answer to how many of yours would start for Fla is 20% or less you're going to get beat and probably badly. If the % is 40 to 60 it'll probably be close. If it's 60% or better would start you're probably going to win. This is who should win, S--- happens.

Except that you are inserting nothing but subjectivity into an evaluation based on who you think or feel would start on another team. The whole point of this is to remove, or change, views of what it really means to be successful in college football. The answer is talent. The easiest way is to do exactly what I did, and then you get to a prediction rate of about 2/3. (it doesn't take long at all to create a spreadsheet that averages every teams last 4 years of recruiting and then rank them)

 

[Boca Vol]

You have argued for a long time about how recruiting rankings are an illegitimate basis for predicting outcomes of games. You are no less wrong then than you are now, but alas you have found the exception to the rule to try to use as proof.

I'm saying you can't simply base predictions on recruiting rankings and yes, this proves that point. There were people on here pointing to recruiting rankings and calling this an even match-up.

All things being equal, the team with the better talent should win. The recruiting rankings of a player don't mean squat once they step on campus. A lot happens after those rankings are made.

To answer your question about spread: The spread is set in an effort to assure gamblers will bet roughly evenly on either side of that number. So, the number is set, and adjusted, to play to the better's perception of the actual game. This minimizes risk to the house.

Spreads are, in fact, proof that games can be easily predicted (otherwise a spread wouldn't be required) and are also proof of how poorly the average better can predict the outcome of games.

I know how point spreads work, so fine, look at the money line. You had to wager $7000 on Oregon to win $100. Conversly, if you would have bet $100 on UT you could have won over $3000.

 

[daj2576]

I know how point spreads work, so fine, look at the money line. You had to wager $7000 on Oregon to win $100. Conversly, if you would have bet $100 on UT you could have won over $3000.

You asked me to explain the spread to you, it isn't like I volunteered that information unsolicited.

But back on point, what does the spread or the money line have to do with actually predicting outcomes of games based on talent averages? If anything it continues to prove that games are easily predictable, but that score differentials are not.

Ultimately you are using the exceptions to prove your point. That is like finding a dog that has fleas and saying that because this one dog has fleas, all dogs have fleas. I readily admit that the game yesterday either exposed a weakness in UT's evaluation, or Oregon's evaluation, or another very specific systematic issue. It did not effect the general system as a whole. It could also be that the talent ratings are still on the money, and that UT and Oregon can be evenly talented but that Oregon's talent exposes the exact weaknesses that Tennessee's talent presents. It isn't like the two teams were playing the same style of football yesterday. UT has recruited talent to play a more traditional style of ball, there is no way to argue that. Oregon recruits talent to play a very specific style of ball, that was originally created to do exactly what it does and that is overcome the strengths of more traditionally talented teams.

I have read recently that 30% of 5* players, 15% of 4* players, and 5* of three star players go on to the NFL. So while you are right, there is a quantifiable wash out rate, there is also a quantifiable success rate and that the events after a kid step on campus are far less random in reference to their long term success than you seem to believe.

We are arguing in circles.

You are telling me that talent evaluations don't or can't predict the outcome of games because there are just too many flaws in that system. Okay instead of having me explain to you repeatedly why it works, explain away why if you use nothing more than talent averages from rivals you can predict about 70% of all games played?

Again you must understand that when you point to an exception, you still haven't explained away the rule.

 

[NavyVOLSfan!]

Your thoughts are all well intended, but we lost yesterday for one simple reason: I just realized I forgot to wear my lucky boxers. Sorry to let everyone down.

Sir :hi: hats off for the honesty, But you need to bust out the staple gun and superglue so this dont happen again.

 

[photovol]

No one has taken into consideration the "Curse Factor"

No Mathmatical formulas can explain this other worldly supernatural power, but Vol fans can attest it does indeed exist and flourish.

 

[Volosaurus rex]

I can't try to justify what happened yesterday but put somethings in perspective.

Last year we gave up 721 total yards to Troy. TROY!
Say that again to yourself, TROY!

Yesterday we gave up 687 yards to Oregon. Oregon's third or fourth stringers are better than anything Troy could have put on the field, and is probably the most explosive offense ever to take a college football field, yet you act like the defense isn't showing some improvement?

It isn't like in the third or fourth quarter Oregon put fans on the field and just had them sit in a drum circle, they put back ups in and continued to run their system.

We are a disciplined football team, even in a very hostile environment we only had 4 penalties for 40 yards. Usually when teams are very out matched and out gunned (and getting the score run up on them) a lack of discipline will show. That didn't happen.

At this point last year we had 21 penalties for 151 yards. This year we have 6 for 57 yards. That in itself is telling.

I didn't see the guys quit on the coach, and I saw that last year a few times with our players (think about Missouri, Florida and Vandy games, or Kentucky the year before as an example).

We got whipped and they kept playing. I see improvement...

Please allow me to offer my two cents' worth of perspective pertaining to Oregon’s offense, which is an absolute work of art, so long as you are not on the receiving end of it. From a comparative standpoint, I would say that, whether the offensive scheme is power-run based, pass-happy or balanced, Oregon, right now, executes their offense as well or better than anyone I have watched since the Oklahoma wishbone teams of old. For what it is worth, that opinion covers 46 years of closely observing the sport.

 

[GoDucks349]

daj,

I think the problem is a fundamental one. When using technical analysis, it's good for explaining the past performance, but when it comes to "future" performance it's not a good "predictor". It's good for giving possible indications, but not predictions. Basing the analysis on a narrow group of data categories as predictors leads to more significant outliers.

 

[Fightmaker]

daj,

I think the problem is a fundamental one. When using technical analysis, it's good for explaining the past performance, but when it comes to "future" performance it's not a good "predictor". It's good for giving possible indications, but not predictions. Basing the analysis on a narrow group of data categories as predictors leads to more significant outliers.

You are the absolute worst football fan I have ever seen.



Congrats on the win.