Game Thread Southern Cal 18, at tOSU 15 (Sept 12th, 8 pm, ESPN)

[jwinslow]

Pollsters vote on the current best team.

They should do what you suggest, but a large number of folks do exactly as ocre suggested....

 

[ochre]

I seem to remember them continuing to vote in the polls after every team has a loss.

Pollsters vote on the current best team.

Random individual pollsters might head out to the polls with that in mind, but collectively their behavior doesn't seem nearly so cut and dry.

The point of the polls, ultimately, is to identify the top couple of teams (for BCS consideration). I think my hypothesis is evinced by what happened with the polls this week. A team that beat a highly ranked team (but not how the pollsters felt they should) on the road lost ground to a team from the Big12 (where a couple of the "favorites" might not be as good as anticipated, making their path marginally easier) that beat an unranked team and to a team from the SEC that beat an unranked team.

Current evidence included, assuming that, collectively, pollsters are voting on the current "best" team, what would cause one to conclude Texas and Florida are better, relatively, than USC because they rolled teams they were supposed to roll? For that to work, one, more or less, has to conclude the pollsters didn't/don't think OSU should have been rated nearly as highly as they were. Instead, what seems to have happened is that pollsters, collectively, now assume, since USC didn't beat OSU convincingly, they're going to struggle to win out.

Winning out "potential" shouldn't be a condition for judging who the best team is right now. It does become a self fulfilling prophecy though. "See, look! I was right team x is the best! They're still undefeated."

 

[MondayAMGenius]

Or one mature QB away from winning everything.

 

[jwinslow]

I've heard voters state that they vote on potential and where they will finish... It's stupid, but it happens. It wasn't an isolated thing either, I've seen it in online articles and heard it stated/argued on talk radio, both local and national.

 

[ochre]

I've heard voters state that they vote on potential and where they will finish... It's stupid, but it happens. It wasn't an isolated thing either, I've seen it in online articles and heard it stated/argued on talk radio, both local and national.

I first started really noticing it (being overt) when people were justifying how high they had WVU in preseason and early season polls after the big east "big boys" left for the ACC.

 

[zincfinger]

They should do what you suggest, but a large number of folks do exactly as ocre suggested....

Should they? What's the "current" best team? The team that played the best game the previous weekend. The previous two weekends? Personally, I think the pollsters should vote according to who has played the best football, on average, for the entire season to date. And while they're far from perfect in doing this, I think they come much closer to doing this than to voting for the "current" best team.

 

[ochre]

Should they? What's the "current" best team? The team that played the best game the previous weekend. The previous two weekends? Personally, I think the pollsters should vote according to who has played the best football, on average, for the entire season to date. And while they're far from perfect in doing this, I think they come much closer to doing this than to voting for the "current" best team.

well. It would help to get away from all the conjecture that is involved with preseason/early season polling. Polls sell though.

Probably not realistic, but polls should probably start up around the 6th or 7th game of the season.

 

[jwinslow]

Should they? What's the "current" best team? The team that played the best game the previous weekend. The previous two weekends? Personally, I think the pollsters should vote according to who has played the best football, on average, for the entire season to date.

We have the same opinion about how polling should be done... I meant current as in current accomplishments, not guesses about upcoming games and expected results.

Here's what ochre is talking about, and there are plenty who vote like this:

ED GRANEY: College football polls all about (mis)perceptions - Sports - ReviewJournal.com

But here was a national voter on a national network, trying to justify why on his ballot, BYU was No. 13 and the Sooners No. 10.


His reasoning: BYU had not appeared on his preseason ballot, while the Sooners were No. 3. He found it difficult to raise the Cougars to a Top-10 level while dropping an Oklahoma team that lost quarterback Sam Bradford to injury any lower.


The real kicker: "There are so many ways to look at it," the voter said. "How much does one team deserve to rise and how much does one team deserve to fall? There is no question that if this was the final poll and my poll was part of the BCS system, I would have much more seriously considered putting BYU ahead of Oklahoma because of the head-to-head victory.


"But because it's only Week 2, I'm not apt to do that."

Now obviously this isn't one of the BCS pollsters, but there are examples of that out there.

 

[zincfinger]

well. It would help to get away from all the conjecture that is involved with preseason/early season polling. Polls sell though.

Probably not realistic, but polls should probably start up around the 6th or 7th game of the season.

I agree that it would be better to withhold the polls until several weeks of the season have been played out. But that is a completely separate question from whether the polls, regardless of when they come out, should reflect who has played the best recently or whether they should reflect who has played the best overall.

 

[zincfinger]

We have the same opinion about how polling should be done... I meant current as in current accomplishments, not guesses about upcoming games and expected results.

Yep, I misunderstood, and are views are fairly similar, apparently. Although, I think I'm much less an advocate for the little guy than you are. A team being penalized for being in a weaker conference doesn't bother me all that much. A team being rewarded for being in a weaker conference, for example WVU for much of the past 5-6 years, bothers me much more.

 

[jwinslow]

another example:

My 2008 Associated Press preseason top-25 football poll (updated with AP rankings for comparative purposes) | College Sports Hotline

* I’ve written this many times before and undoubtedly will again (many times): My preseason poll is not a projection of how I think the teams should be ranked right now — I couldn’t care less who the best team is three weeks before the season starts.
Instead, my preseason poll is a projection of where I think the teams will finish in the final, post-bowl poll in January. (Yes, that means the No. 1 team is my pick to be the national champion.)
...
* You might notice that this top-25 has been altered since my Ridiculously early top-25 for 2008, which I published on the Hotline in January.
And the reason for that is simple: Back then, I hadn’t evalulated everyone’s schedule — that’s one reason it was a “ridiculously early” poll.

 

[cincibuck]

My own thoughts were that two teams took a hit for playing in the biggest OOC game so far this season. USC took a hit for not blowing the Buckeyes out while Florida and Texas rose despite the fact that they've played an extremely weak OOC schedule, The Buckeyes took a hit for not blowing out Navy and for losing in a tightly contested game to a team that sources have ranked in the top three since the end of the last season.

 

[zincfinger]

another example...

And I agree with you. I dislike the practice of voting teams based on what you think their final record will be, rather than voting them based on what they've done to this point. But I think this problem results primarily from the refusal to significantly drop teams who have won, regardless of who they've won against. So for example, pollsters don't want to put BYU ahead of OU right now because they foresee little chance, according to their devised system, of putting OU ahead of BYU in the future because BYU doesn't fact many tough opponents the rest of the way. If there was less resistance to dropping teams who are winning, but winning against weak opposition, there would be less resistance to putting BYU ahead of OU right now. And there would have been less compulsion to make WVU a preseason top-3 or top-5 for 5 years running. My point isn't to pick on WVU, they undoubtedly had good teams. But their preseason rankings also undoubtedly had a good deal to do with their perceived relatively easy schedules.

 

[ochre]

I agree that it would be better to withhold the polls until several weeks of the season have been played out. But that is a completely separate question from whether the polls, regardless of when they come out, should reflect who has played the best recently or whether they should reflect who has played the best overall.

That's the rub, though. With out knowing the relative merits of the various teams, how can we truly say who is playing "the best"? East Carolina last year going into week 3? They beat the no. 17 and no. 8 ranked teams the first two weeks.

 

[lightyears2012]

we need GAME DAY!!!!!

Snowball...