UT Vols #12 in Sagarin

[DiderotsGhost]

Now that we have four weeks of data, the Sagarin ratings are becoming more meaningful.

Sagarin Ratings

Here's the top 5, our UT Vols, the SEC, and all of our opponents this season.

1. Ohio State
2. Alabama
3. Michigan
4. Stanford
5. Louisville

6. Texas A&M
10. Ole Miss
12. Tennessee
14. LSU
20. Auburn

21. Florida
23. Virginia Tech
26. Arkansas
48. Missouri
52. Mississippi State

53. Georgia
72. Vanderbilt
73. App State
74. South Carolina
81. Kentucky

93. Ohio
198. Tennessee Tech

Virginia Tech has been played phenomenally the past two weeks, which makes that win look more impressive in hindsight. This is the lowest I can ever remember seeing Georgia ranked in Sagarin. Sagarin was very skeptical of Georgia even before the blowout loss to Ole Miss. Also note that Sagarin believes that Missouri may be a tougher opponent for us than Georgia.

 

[WesternKyVol]

Sagarin tends to overvalue blowouts. They can be undervalued at times, but after you win by 3 scores, does any more really matter. Beating Delaware State 79-0 should mean less than beating Vandy by 1. They are that bad. And while losing to Ole Miss by 30 is bad, it it's just one data point for a team that is historically very up and down.

 

[ImportedVoL]

While it is decent to look at, and gives a little insight..

Still do not believe there is enough data yet..95% of the teams out there have their toughest games ahead of them.

I mostly like these types of polls for end of season rankings..JMHO

 

[WiseOlVol]

How is Michigan in the same area code as Alabama?
The Tide boat raced Michigan last time.
Michigan has played 4 practice games.

 

[WesternKyVol]

Where do you want Michigan ranked?

 

[DiderotsGhost]

Sagarin tends to overvalue blowouts. They can be undervalued at times, but after you win by 3 scores, does any more really matter. Beating Delaware State 79-0 should mean less than beating Vandy by 1. They are that bad. And while losing to Ole Miss by 30 is bad, it it's just one data point for a team that is historically very up and down.

I don't think Sagarin discloses the actual formula, but if you've monitored the site for many years, you'll note that it's much more influenced by strength of schedule and quality wins than margin of victory over weak competition. A 23 - 10 team victory over the #6 team would definitely help more than a 77-0 over the #245 team. But certainly, score is important in the Sagarin model. Losing to the #1 team by a score of 28 - 27 is much more impressive than losing to the #1 team 35 - 10.

IMO, that's one of the big strengths of Sagarin. The human pollsters will often reward a 7 - 0 team that hasn't faced a single top 50 opponent all season with a top 15 ranking, but they'll send a 3-4 team with very close losses to the #1, #2, #3, and #5 teams off to unranked oblivion, even if that team also has wins over the #10 and #12 teams. This is why Sagarin is generally much better than the human pollsters, at least when it comes to ranking teams outside the "top 5" (I tend to think the human pollsters get the top teams more right, but are incapable of distinguishing between #12 and #30, so they simply take win - loss records).

 

[Boston Vol]

How is Michigan in the same area code as Alabama?
The Tide boat raced Michigan last time.
Michigan has played 4 practice games.

The last time they played Michigan was in 2012. What the hell does that have to do with this season? You have an unhealthy obsession with Michigan. And btw, their victory over Colorado looks pretty good right now.