• Quick note - the problem with Youtube videos not embedding on the forum appears to have been fixed, thanks to ZiprHead. If you do still see problems let me know.

Romney, Obama, Rasmussen

I'm guessing it's been asked, but I don't recall the answer. How do these poll takers adjust the data for the changing demographics of land line users? Are these still mainly telephone polls?
 
Last edited:
Skeptic Ginger said:
I'm guessing it's been asked, but I don't recall the answer. How do these poll takers adjust the data for the changing demographics of land line users? Are these still mainly telephone polls?
Click to expand...
That's the "secret sauce" that goes into every pollsters design of their polls, among other things. Obviously the are aware of that change.

gabeygoat said:
I believe the Rasmussen poll has an over-reliance on land lines?
Click to expand...
Cite?
 
JoeTheJuggler said:
Your point is based on a misreading of what I said.

Meanwhile, I'll turn your reasoning around on you. Despite claims that it's an even race, Romney supporters won't take even money bets. It appears they recognize that Romney has substantially less than a 50% chance of winning the election.
Click to expand...

Why would I take an even money bet here rather than put the money on InTrade or BetFair?

I'll make you a deal, though. If Romney gets to 50% at either site I will make a $100 bet with you. Deal?
 
Skeptic Ginger said:
I'm guessing it's been asked, but I don't recall the answer. How do these poll takers adjust the data for the changing demographics of land line users? Are these still mainly telephone polls?
Click to expand...

I did ask earlier, and the answer seems to be that there's a slight Republican bias, and that the better polling companies don't rely only on land lines. (And that Silver takes these methodological concerns into account somewhat in his analysis.)

I'm also curious about how the question is asked. I saw that Gallup asks the question in their nationwide poll as if there were only two candidates to choose between even though voters in many states will see 3 or 4 candidates on their ballots. But then again, we all know the nationwide poll--the one that gets most media attention--is probably the least significant one.
 
Last edited:
Brainster said:
Why would I take an even money bet here rather than put the money on InTrade or BetFair?

I'll make you a deal, though. If Romney gets to 50% at either site I will make a $100 bet with you. Deal?
Click to expand...

That doesn't make any sense. Intrade doesn't do betting; it's a prediction market where you buy and sell shares in a proposition. I don't know about BetFair.

I'll take a $100 even money bet on Obama to win right now. Are you in?

You keep ignoring the fact that I'm not interested in taking a gamble (betting where the odds accurate reflect the outcome). So I will not commit to betting if the actual odds come down to 50:50.

I don't think they will, but I'm certainly not going to commit to such a hypothetical anyway since it is not consistent with what I said in the post that you still seem to be misreading.

And I note that you ducked my turning the question back on you. Don't you think your reluctance to take an even money bet right now means you don't actually believe Romney has an even chance of winning the election?

This is the approach you're trying to use on me. You're claiming that my unwillingness to take a bet whose payoff odds accurately reflect the probability of the outcome is a sign that I don't believe those odds accurately reflect the probability of the outcome.

But again, I would be a fool not to take an even money bet on Obama to win right now. It would be reasonable for me to take such a bet unless my cost to get money to put up were greater than about 300%.
 
Brainster said:
They are independent in the sense that what happens in one won't influence what happens in each other. But they are not truly independent in that what happens externally to both could influence both.
Click to expand...

OK. So it would be wrong to treat state events as dependent, and it would be appropriate to consider the national polls in assessing a state's probabilities. As I understand it, that's just what Silver does.
 
Since the land line cell phone question isn't so clear, I decided to look further.

This was interesting but led to a number of additional questions:

Does Gallup call cell phones?
Gallup includes cell phones in each national Gallup poll. Gallup has been including cell phone-only households in all national telephone Gallup polls since January 2008. ...

...This is particularly important given recent studies that show that people in cell phone-only households tend to be younger, are more likely to be racial and ethnic minorities, and more likely to be transient....
Click to expand...

It would seem to me that it will be a while before these issues are fine tuned.

Do people who get called on their cell phones refuse to participate more often?
Does the sampling account for rapid changes in the number of cell phone only households?
How do the cell phone results hold up when land-line analysis formulas are applied?
 
JoeTheJuggler said:
....I would be a fool not to take an even money bet on Obama to win right now. It would be reasonable for me to take such a bet unless my cost to get money to put up were greater than about 300%.
Click to expand...

I'm calling the election for Romney at this time.
 
BenBurch said:
I'll go even money too. Any takers?
Click to expand...

Good luck. I've noticed that most of our resident Romney supporters here are talking a good game, but when it comes to betting on their man, they develop a strong propensity for waffling and finding something else to discuss.

Pretty much shows you how much confidence they really have in their man, doesn't it?

ETA: To be fair, after three days of my initial challenge, DDWW did take my $100 bet. But it took three days for someone to take up my challenge, and this was during the height of Romney's "surge" in the polls!
 
Last edited:
Brainster said:
If Romney gets to 50% at either site I will make a $100 bet with you. Deal?
Click to expand...
Why? Why not now? You like everyone else who is asked to put up don't want an even bet on Romney. Why? It's NOT a good bet. Now, I'm happy to take an even bet and so is Joe, ask yourself "why"?

  • Joe and I like Obama's odds and are willing to take an even bet right now.
  • You don't like Romney's odds and want to wait to see if they will improve.
Can you see the point now?

ETA: In the interest of full disclosure I should note that I don't have any money to gamble.
 
Last edited:
Romney at 41% on intrade right now... never seen him this high. Seems the speculators believe Romney to win the debate.... I saw this last debate as well.... my guess is that Romney drops to 37-38%... right before the debate.
 
BenBurch said:
I'll go even money too. Any takers?
Click to expand...

Hell, I'll go even money on Obama and turn around and put the same amount down on Romney at InTrade. Then I'm guaranteed nothing worse than breakeven; if Obama wins my loss on InTrade is canceled out by what I win here. But if Romney wins (as I expect) I lose $100 here and make roughly $150 over there.

This is called hedging one's bet, and it's why I always insist on taking the points when my team is the underdog. If you give someone the opportunity to bet both sides and profit, you're a fool.
 
Brainster said:
Hell, I'll go even money on Obama and turn around and put the same amount down on Romney at InTrade. Then I'm guaranteed nothing worse than breakeven; if Obama wins my loss on InTrade is canceled out by what I win here. But if Romney wins (as I expect) I lose $100 here and make roughly $150 over there.

This is called hedging one's bet, and it's why I always insist on taking the points when my team is the underdog. If you give someone the opportunity to bet both sides and profit, you're a fool.
Click to expand...

Maybe I am misreading this thread, but it seems to me that you believed Romney to have a better than 50% chance of winning, and chose not to accept 50/50 bets before you knew of trading sites like Intrade. If this is true, why?
 
RandFan said:
RCP has Romney ahead in electoral votes.
Click to expand...

That is really odd to me, and not good news (however I think Obama trounced Romney in the debate tonight* so things may change in a few days). I tend to follow Silver because I trust his methods so I don't know how accurate RCP is and Nate still has Obama clearly in the lead on electoral votes....

* said as an unbiased Obama supporter but judging by what I am reading and seeing it seems as if most agree with me on this
 
You must log in or register to reply here.

Back
Top Bottom