For me to just copy another poll without having my own opinion and reason for how I rank the teams is pointless.

If we want to just be a copy of other polls then maybe every poll ought to be submitted here each week in an official manner.

One of the things I like about our poll is the feedback we get from each other. I see a lot of polls that I don't agree with but at least I can get an idea of how they came up with their ranking. If someone is just going to submit another poll system there can be no explanation.

I guess the question is do we want pollsters or copies of other polls? Independent thinkers or dependent amusements (a=non muse=to think)?

One other opinion of mine on this matter:

Even if copying a poll is a legitimate way to vote without your bias, isn't the (very lofty) goal of this website to eventually be considered in the BCS calculations down the road? and isn't Jeff Sagarin (or many other polls people might consider copying) already included in the BCS formula?

I have no problem with someone copying the poll of someone else if they completely agree with how the poll is calculated. Unfortunately, this can't be the case in the early season with Sagarin rankings, because before the graph is connected, the Bayesian analysis will rely on prior probabilities that are, in effect, JS' opinions (and I can't see this person knowing how JS formulates his opinions). Furthermore, the calculations that JS uses in his polls are not published to my knowledge, so again, unless one has inside information, a pollster can not agree completely with the way that JS calculates his rankings because they don't know what those calculations are.

If the calculations have been published and the pollster agrees with them, then by all means, copy away. Or of course if you are JS, use your rankings.

I agree that someone just copying anything with little thought doesn't provide much benefit for including everyone and their ideas, and I wouldn't (and don't) do it for my poll. But that's me and my opinion. Someone else may think copying Sagarin is the ideal calculation and way to perform and their poll should matter and they want to essentially give THEIR voice away to that rankings. That's their choice. If they want to spend the 5 minutes a week to come here and enter the close of Sagarin, that's their choice.

As for claiming Sagarin and "copying" (my words) the formula, cdavis627 hasn't said they are Sagarin and perhaps they are using an ELO-CHESS system that would also produce similar rankings. I don't run that system, so I don't know if teams are all connected yet for last year to matter still. It's possible that after week 3, all teams are connected so running the system without last season or with until they are all connected would be identical. That wouldn't explain week 2, unless they all got connected after week 2. Hard to believe that would be possible, but it could be.

The ELO-CHESS system Sagarin used is a known system. I'm not sure if the exact calculation is known, factoring in scores or game margin (which the BCS doesn't use), but if you're doing the ELO-CHESS system, you probably could play around and figure it out.

As an analogy, say when we US citizens voted for the president some celibrity comes out and says "I'm voting for X", and fans of the celebrity say they're going to vote for X because the celibrity is going to. Does that invalidate their vote? Aren't they an adult, and while we disagree to choose to vote or "copy" a vote, it's their decision to do so?

It would be "easy" to screen out all polls that are identical to Sagarin or another computer/poll, but what if you just happened to have the same poll that week? Should you not be allowed to vote that because it happens to be the same that week? I could see 1 or 2 weeks that happening, especially late in the season, and it not be a problem. But trying to have computer smarts to compare someone's poll to another computer/(AP) poll would be difficult to do. The current comparison system doesn't even allow the coaches or AP or BCS or Harris to compare to a user (That's an idea with a new comparison system I want to work on).

I don't know why someone would bother participating in this website if they are merely copying another's poll. On the other hand, if someone else has completed a poll and you fully agree with it, why recreate the wheel? Just as long as you can back up what it is about the poll you liked and why you think it's valid. And with all the other polls out there, there are bound to be duplicates.

Besides, if someone chose to copy my poll, I'd be pretty proud of myself....and yes, you are free to use my poll if you'd like.....if you actually existed, fake poll copy person.

The pollsterin question isn't just using the ELO-Chess system (which is what the BCS uses), but Sagarin's combination of ELO-Chess and his predictor (which is not what the BCS uses. So the concern that this would just mimic the BCS instead of providing a separate ranking isn't too bothersome.



As of week 3, neither JS' ELO-Chess, nor his predictor graphs are connected.



As soon as both the Predictor and ELO-Chess are connected, this would be very possible.

Sorry for jumping to conclusions so early. In retrospect, I don't mean to accuse anyone of copying without hard evidence. It is entirely possible that the pollster in question coincidentally has the same poll as JS. Until the pollster confirms or denies this fact, I cannot pass my judgement in this particular case.

That said, the discussion about the possibility of pollsters simply copying other polls is pretty valid. We have had pollsters in the past that happened to have the same poll as the AP and no one really cared all too much. Who's to say that someone couldn't happen to have the same poll as a computer.

What's the point of being a member of this website if you just copy and paste a computer generated poll and claim it as your own?? Even if you come clean that it's not your own, I thought the whole point of this website is to bring together a group of intelligent college football fans to discuss interesting topics (such as this one) and to rank teams according to THEIR OWN opinions. I'm all for using computer polls as a starting point for your own poll (especially early in the season when it's tougher to distinguish contenders from pretenders) but copying a computer poll and then posting it as your own takes all the fun out of building your own poll.

Right. But if THEIR OWN OPINION is that Computer/Poll X is super awesome, and they just want to use it at theirs, is that wrong? You wouldn't do it. I wouldn't do it. But is that wrong?

And if enough people were to do this, wouldn't it just be a popularity contest of whose poll is better, and not ranking teams. 4 people vote like Sagarin, 7 for Billingsley, and 13 for TheBlitz. "Yay, TheBlitz wins! wait, you have who at number one?"

But I agree, if you agree with how another poll is put together, then use it. Just be ready to defend the merits of that poll and why you like it...

I'm really curious as to where cdavis is in the middle of all this, and am very interested to hear his response...

If you read his earlier posts he seems to suggest that he bases his poll on an 'automated presence', as well as opinion, so if he is just copying Sagarin, that's not cool. If you come out and say that you're using someone elses system, I have less of a problem with it, but suggesting you're coming up with it yourself when you're just copying and pasting from someone else is, as Sonic the Hedgehog would say, Nooo good.

It isn't fair to those who form their own polls, using their own methodolgy, not relying upon another's methods, which may, or may not yield a better result. And, I know something about statistical methods, if you want to make your poll more valid, you try to minimize those things that tend to make a poll the same. So, if one person uses Jeff Sagarin's poll, another may follow suit.

I wouldn't say it's WRONG. It just seems to take all of the fun out what we're doing here and I disagree with it. If someone thinks that a particular poll has a sure-fire, perfect algorithm to determine rankings, that's great but what's the point in copying that poll week after week and presenting it on this website?

However, if someone's poll is identical to a computer generated poll on a particular week; that is, they simply agree with that particular poll for that particular week and they can justify why they agree with what teams are ranked where, I have absolutely no problem with that and that's certainly not WRONG.

Just thought I'd point out that cdavis' poll is not the same as JS' this week.

http://www.kiva.net/~jsagarin/sports/cfsend.htm

thank you.

maybe it was coincidence... maybe not... i guess until cdavis chimes in we will never know...