Tennessee ranked under Virginia Tech and Florida (per S&P)

#1

VOL-uh-tile

Accept me into your immortal oath
Joined
Jul 11, 2016
Messages
757
Likes
669
#1
I am a proponent of using statistics to determine outcomes of upcoming games. Phil Steele is too, and he gets his predictions pretty darn close. In the past I have been skeptical of one particular set of advanced stats, S&P. Last year I vividly remember referring to them in the middle of the season, after the Georgia game I believe, and saw that Bowling Green was about 10 spots above us, around #25. Well, as of 9/24/16, Tennessee is ranked 15th, VT is 14th and Florduh is 7th. We actually moved down 2 spots and the gators up 2 spots after we beat them.

All of this is to say I'm not concerned with this specific set of stats. It obviously has some deficiencies in it's calculations. But it does frustrate me that this fairly commonly used method can be so inaccurate. So if anyone claims their team is better than our Vols and cites this metric, laugh in their face and tell them to use FPI. As always, Go Vols and beat Georgia.
 
#4
#4
I don't care what a computer or talking head says at this point in the season all that matters is we beat both of their butts. If we win out none of these rankings will matter at all.
 
  • Like
Reactions: 1 person
#5
#5
All polls are 100% meaningless till the last game of the season has been played. And even then, sometimes they are wrong.
 
#6
#6
Bill Connelly, who made these, has a good podcast called Podcast Ain't Played Nobody where he explains this stuff. He admits his numbers are not great this year and are turning out some weird results. The system might be up for a tweaking soon.
 
  • Like
Reactions: 1 person
#10
#10
Bill Connelly, who made these, has a good podcast called Podcast Ain't Played Nobody where he explains this stuff. He admits his numbers are not great this year and are turning out some weird results. The system might be up for a tweaking soon.

Maybe they only take first half stats into consideration. That would explain a lot! :p
 
#11
#11
Maybe they only take first half stats into consideration. That would explain a lot! :p

Hell if I know.

All I know is that, while I love me some statistical analysis, it sometimes really doesn't tell the whole story. Almost his entire top 25 is a complete farce. I'm not sure how any poll, computer or otherwise, would churn out the rankings the S&P currently is.
 
  • Like
Reactions: 1 person
#13
#13
as long as games are won and lost on a real field, I'll just use that as my Poll Indicator...:)

GO VOLS!
 
#19
#19
We're 4-0 with a decent SOS and yet we're ranked below the preseason ranking(AP Poll). :confused:

Yes, it's been extremely odd. UT dropped more for barely beating App State then they moved up for beating UF and Va. Tech combined. And for all the hand wringing about UT starting games poorly, App State is the only truly close game (1 possession) they've played.

I think it all has to do with the mindset of the voters when the season started. They had UT ranked highly but didn't entirely trust the ranking because the track record in recent seasons wasn't there. When they struggled against App State they overreacted.

UF and Va Tech were pretty thorough beatdowns at the end of the day but voters are getting hung up a bit on the fact UT was down early. If anything, the early struggles make me think this team is nowhere near it's ceiling. A team that can go on a 38-0 run and a 45-3 run against those defenses is good enough to beat anybody. We'll see how the season plays out but the next 3 weeks should be very interesting.
 
  • Like
Reactions: 1 person
#20
#20
The problem with computer-gemerated polls this early in the season is that the sample size is too darn small to use until week 6 or 7, at the earliest. Anyone that's taken a basic stats class can see that. Hell, some computers rank Tennessee #1 or in the top 5...
 

VN Store



Back
Top