The quant polls say Tennessee should be in the CFP

[DiderotsGhost]

The quant polls are getting interesting, particularly where they radically diverge from the human polls. I prefer the good quant polls over humans, since the human pollsters tend to underestimate strength of schedule. The Massey Composite Index is the most important quant-oriented poll IMO. It's an average of several polls (including the human ones, but mostly quant polls). Right now, Massey Composite says:

1. Georgia
2. Ohio State
3. Tennessee
4. Michigan
5. TCU
6. Bama
7. LSU
8. Clemson
9. Penn State
10. Utah
11. USC

Not only are we #3, but we're pretty firmly entrenched in that spot now. Indeed, we're #3 in the vast majority of the quant polls.

Sagarin is also interesting. We've been at #5 in Sagarin for a long time, but the Mizzou win pushed us up to #4. We also moved up to #3 in Sagarin's Recent ratings.

Here are Sagarin's rankings (with Recent Rating in parenthesis)

1. Georgia (1)
2. Ohio State (2)
3. Alabama (5)
4. Tennessee (3)
5. Michigan (4)
6. Utah (6)
7. Penn State (7)
8. Oregon (8)
9. TCU (9)
10. LSU (11)
11. Kansas State (10)
13. USC (14)
14. Clemson (15)

IMO Sagarin is the gold standard in football quant polls, particularly this late in the season, it tends to get things right. Sagarin thinks we're either the 3rd or 4th best team in the country right now (depending which measure you use). But more telling is that it puts TCU down at #9 and USC down at #13. And I tend to agree; TCU is a very good team, but I think if they slip in as the #4 seed, they'll get blown out by Georgia or Ohio State in the first round of the playoffs.

There is some hope. If Kansas State wins out, they'll play TCU in the Big 12 Championship Game. And Sagarin thinks Kansas State is underrated. It picks TCU-K State as a toss-up more or less. Sagarin also thinks Baylor is very underrated and has them at #18. TCU plays Baylor next week.

While I think the case for us at #4 is very strong, and the quant polls agree, I do think we have to hope that TCU loses 1, or alternatively, either Ohio State or Michigan gets blown out by the other. Worst case scenario is that TCU wins out, Michigan loses to Ohio State by 3 points (or some small margin), and we get stuck at #5.

But the quant polls are seeing our overall performance against a Murderer's Row of opponents, which now includes 3 top 10 teams (Georgia, LSU, Bama) and we went 2-1 against that (with 2 of those being road games). No one in the country has a resume on par with us and the quant polls realize that more than the human voters.

 

[GoVols9800]

The quant polls are getting interesting, particularly where they radically diverge from the human polls. I prefer the good quant polls over humans, since the human pollsters tend to underestimate strength of schedule. The Massey Composite Index is the most important quant-oriented poll IMO. It's an average of several polls (including the human ones, but mostly quant polls). Right now, Massey Composite says:

1. Georgia
2. Ohio State
3. Tennessee
4. Michigan
5. TCU
6. Bama
7. LSU
8. Clemson
9. Penn State
10. Utah
11. USC

Not only are we #3, but we're pretty firmly entrenched in that spot now. Indeed, we're #3 in the vast majority of the quant polls.

Sagarin is also interesting. We've been at #5 in Sagarin for a long time, but the Mizzou win pushed us up to #4. We also moved up to #3 in Sagarin's Recent ratings.

Here are Sagarin's rankings (with Recent Rating in parenthesis)

1. Georgia (1)
2. Ohio State (2)
3. Alabama (5)
4. Tennessee (3)
5. Michigan (4)
6. Utah (6)
7. Penn State (7)
8. Oregon (8)
9. TCU (9)
10. LSU (11)
11. Kansas State (10)
13. USC (14)
14. Clemson (15)

IMO Sagarin is the gold standard in football quant polls, particularly this late in the season, it tends to get things right. Sagarin thinks we're either the 3rd or 4th best team in the country right now (depending which measure you use). But more telling is that it puts TCU down at #9 and USC down at #13. And I tend to agree; TCU is a very good team, but I think if they slip in as the #4 seed, they'll get blown out by Georgia or Ohio State in the first round of the playoffs.

There is some hope. If Kansas State wins out, they'll play TCU in the Big 12 Championship Game. And Sagarin thinks Kansas State is underrated. It picks TCU-K State as a toss-up more or less. Sagarin also thinks Baylor is very underrated and has them at #18. TCU plays Baylor next week.

While I think the case for us at #4 is very strong, and the quant polls agree, I do think we have to hope that TCU loses 1, or alternatively, either Ohio State or Michigan gets blown out by the other. Worst case scenario is that TCU wins out, Michigan loses to Ohio State by 3 points (or some small margin), and we get stuck at #5.

But the quant polls are seeing our overall performance against a Murderer's Row of opponents, which now includes 3 top 10 teams (Georgia, LSU, Bama) and we went 2-1 against that (with 2 of those being road games). No one in the country has a resume on par with us and the quant polls realize that more than the human voters.

And Alabama at 3 above two teams that they lost to. And two losses. Makes complete sense.

 

[ProudVOLfan]

I cant see Bama above LSU, or even Oregon for that matter.

 

[DiderotsGhost]

And Alabama at 3 above two teams that they lost to. And two losses. Makes complete sense.

It actually does make sense. Alabama played 3 top 10 teams on the road this season and went 1-2 against those teams, and the two losses were by 3 points and in overtime.

Politically speaking, 3 SEC teams in the CFP is not going to fly, and Bama's 2 losses basically disqualify them barring a complete collapse by several teams. But in reality, I think 3 of the top 4 teams are SEC teams. Michigan has a case, as well. But no one in their right mind should think TCU would be undefeated if they played Georgia, Bama, and LSU all in the same season. TCU is maybe the 7th or 9th best team in the US; I don't really buy them at #4. If TCU played our schedule, they would have at least 2 or 3 losses.

 

[GoVols9800]

So Bama with two losses and ahead of both the teams that beat them makes sense? And with one more loss than us. Ok

 

[loservol]

Common sense also says the exact same thing.

 

[Special Ed]

Quant?

 

[Stoerner Fumbles]

Quant?

Interestingly it is an abbreviation for quantitative logic, but much of the logic being deployed here is qualitative. In this instance, he means a computer model using a half-ass algorithm developed by programmers and data scientists who couldn’t get hired elsewhere— over the opinions lesser idiots like coaches and sportswriters.

 

[DiderotsGhost]

Interestingly it is an abbreviation for quantitative logic, but much of the logic being deployed here is qualitative. In this instance, he means a computer model using a half-ass algorithm developed by programmers and data scientists who couldn’t get hired elsewhere— over the opinions lesser idiots like coaches and sportswriters.

Literally none of that is correct.

Reminds me of the Billy Madison quote:

Mr. Madison, what you’ve just said is one of the most insanely idiotic things I have ever heard. At no point in your rambling, incoherent response were you even close to anything that could be considered a rational thought. Everyone in this room is now dumber for having listened to it. I award you no points, and may God have mercy on your soul.

It's funny how there are so many people here that hate that quant polls even though they blow away the human polls on predicative accuracy. There is a reason they are used in sports betting instead of garbage coaches' and media polls.

 

[Stoerner Fumbles]

Literally none of that is correct.

Reminds me of the Billy Madison quote:

Ok. As a data scientist who works with quantitative models and algorithms all day, I await your assistance.

 

[jim pique]

The only poll that matters is the CFP ranking.

 

[FlyFishnVol]

The quant polls are getting interesting, particularly where they radically diverge from the human polls. I prefer the good quant polls over humans, since the human pollsters tend to underestimate strength of schedule. The Massey Composite Index is the most important quant-oriented poll IMO. It's an average of several polls (including the human ones, but mostly quant polls). Right now, Massey Composite says:

1. Georgia
2. Ohio State
3. Tennessee
4. Michigan
5. TCU
6. Bama
7. LSU
8. Clemson
9. Penn State
10. Utah
11. USC

Not only are we #3, but we're pretty firmly entrenched in that spot now. Indeed, we're #3 in the vast majority of the quant polls.

Sagarin is also interesting. We've been at #5 in Sagarin for a long time, but the Mizzou win pushed us up to #4. We also moved up to #3 in Sagarin's Recent ratings.

Here are Sagarin's rankings (with Recent Rating in parenthesis)

1. Georgia (1)
2. Ohio State (2)
3. Alabama (5)
4. Tennessee (3)
5. Michigan (4)
6. Utah (6)
7. Penn State (7)
8. Oregon (8)
9. TCU (9)
10. LSU (11)
11. Kansas State (10)
13. USC (14)
14. Clemson (15)

IMO Sagarin is the gold standard in football quant polls, particularly this late in the season, it tends to get things right. Sagarin thinks we're either the 3rd or 4th best team in the country right now (depending which measure you use). But more telling is that it puts TCU down at #9 and USC down at #13. And I tend to agree; TCU is a very good team, but I think if they slip in as the #4 seed, they'll get blown out by Georgia or Ohio State in the first round of the playoffs.

There is some hope. If Kansas State wins out, they'll play TCU in the Big 12 Championship Game. And Sagarin thinks Kansas State is underrated. It picks TCU-K State as a toss-up more or less. Sagarin also thinks Baylor is very underrated and has them at #18. TCU plays Baylor next week.

While I think the case for us at #4 is very strong, and the quant polls agree, I do think we have to hope that TCU loses 1, or alternatively, either Ohio State or Michigan gets blown out by the other. Worst case scenario is that TCU wins out, Michigan loses to Ohio State by 3 points (or some small margin), and we get stuck at #5.

But the quant polls are seeing our overall performance against a Murderer's Row of opponents, which now includes 3 top 10 teams (Georgia, LSU, Bama) and we went 2-1 against that (with 2 of those being road games). No one in the country has a resume on par with us and the quant polls realize that more than the human voters.

Too much subjectivity in human polls. The Northeast has been anti-SEC.

 

[Gandalf]

And I am sure the CFP committee will consider this - but also will think about how to ensure MAXIMUM viewership. If based on deserve I am pretty sure we get in. If viewership, I dont know. There are a lot of TN fans this season.

 

[Raebo]

I assume some of these polls, maybe most, factor in home field advantage. If they do then I can see why Bama might be rated higher than UT.
Does anyone believe that playing Alabama at home had no affect on the the game or playing Georgia away didn’t have an affect??? How many posts have I seen from our fans saying they want to play Georgia on a neutral field? And I would love to play them again on a neutral field.

 

[Arrested Devolopment]

CFP selection seems to be more about revenue, optics, and confirmation bias than choosing the most competitive teams.

 

[Cartervol]

CFP selection seems to be more about revenue, optics, and confirmation bias than choosing the most competitive teams.

They're about brand and getting as many tv markets involved as possible.. That's why they'll not do Tennessee any favors in this deal. They'll have to be forced to put us in it, they'll exhaust ever possible scenario before they will.

 

[Vol in Buckeye Land]

And I am sure the CFP committee will consider this - but also will think about how to ensure MAXIMUM viewership. If based on deserve I am pretty sure we get in. If viewership, I dont know. There are a lot of TN fans this season.

We’ve been involved in the two most viewed cfb games this season.

 

[CNCVol88]

CFP selection seems to be more about revenue, optics, and confirmation bias than choosing the most competitive teams.

idk about that. The BCS, if it still existed would allegedly be pretty similar