Survivor Round 3 -- Who'd you vote off the Presidential Election island?

[AP]

A week or so ago, somebody started a survivor-style poll -- Who would you vote off the island? But they only went two rounds. Lieberman lost in the first and Gephardt in the second round (if memory serves).

So, we're down to these seven:

Remember, the question is 'who would you vote OUT of the election?/who wouldn't you ever vote for? Got it?

(Likely Democratic Primary voters only, please).

[JVS]

1. My choice isn't present

[AP]

2. If your choice isn't present, you've already won. You're voting

for the person you don't want to get the nomination the most.

[JVS]

3. Well that is a nice way to look at it.

[kixot]

4. WTF???

:wtf: What's with the high numbers on Graham? He's a good candidate.

[Glade]

5. Did * win the immunity competition?

I want to vote for *. Just this once!

[AP]

7. We'll have a championship round

with the final Democratic survivor facing off against Bush. OK?

[AP]

6. Kick.

I'm not going to call this one until there's a, uhm, statistically significant number of votes.

[mndemocrat_29]

8. How many votes do you consider enough?

n/t

[AP]

10. If I say how many, it might encourage cheating.

Just kidding.

When I read those polls in the paper everyone gives so much weight, you see sample sizes. I'm trying to get close to a sample size I have in mind from a poll I read earlier today. Invisible goalposts, I know, but that's all I'm saying.

[redeye]

13. In statistics, 30 is considered big enough

Howevr, in that case the margin of error is outrageous (I can calculate the margin of error for you if you want me to).

[AP]

15. What's the margin of error for 60, 80, and 100?

I'm just curious. Wondering if this one is outside the margin of error already.

Thanks

[redeye]

23. Hmmm...

IIRC, the margin of error is equal to the standard deviation divided by the square root of the sample size.

Now, in a binomial distribution (i.e. every person votes either for or against someone independently of the the other people's votes), S.D = npq, whereas n is the sample size, p is the probability that the test will be successful (i.e. the percentage of the vote the candidate got), and q is 1-p. Note that this applies to each candidate separately.

So, let's look at Sharpton, how has 22 votes against him out of 72. In his case, SD = 72*(22/72)*(50/72) = 22*50/72 = 15.2777777... and SQRT(N) = 8.49, so S.E = 1.8 votes.

IOW, Sharpton clearly loses.

[AP]

9. I don't want to influence the voting, but...

...the Keryy folks have to be happy with these results, so far.

[Ein]

11. He seems to be playing it right n/t

[redeye]

12. Sharpton

Not only as strategic voting to ensure that Dean stays, but also because I loathe the guy's racism and race-baiting.

[AP]

14. This guy has talked nothing but sense since he entered the race.

But I guess you represent a segment of "thought" among Democrats.

[ComerPerro]

16. You should have called this "The Weakest Link"

The Kucinich and Kerry supporters are voting against Dean because they don't want to face him in the final rounds. Meanwhile the Dean people are voting against Sharpton. They were successful in getting rid of Lieberman, but I think everyone realized how damaging he was.

[AP]

19. Now, normally I'd agree that strategic voting skews things.

However, I just have the sense that people are being pretty honest about this. It sort of matches expectations.

Lieberman was the obvious first round loser.

Gephardt makes sense to me. He is clearly the guy with the least charisma. He totally rides on his tight connection with labor, and labor is going to go all out for whomever the nominee is, so nobody really feels the need to feed that beast (and labor certainly has friends in the field who remain).

I would think Graham would go next, because his campaign seems like it's just about one issue -- skewering Bush on homeland security, and he doesn't even seem like he expects to do anything other than harp on that issue. However, I understand Sharpton going next, even though I don't think it's wise to foreclose his wit and irreverence at an early stage. Hostilitiy towards Dean also makes a certain amount of sense too.

[mndemocrat_29]

17. Interesting Results so far

I knew that Lieberman would be the first one gone here, but I was kind of surprised by Gephardt. I'm really shocked by Kerry's low numbers (though I'm glad because I'm a Kerry supporter) and Graham's higher numbers.

By the way, I think that this is a great idea for a poll. Is it something that happens every week, every couple of days, every night. I'd like to know so that I could watch for it next time.

[AP]

20. I think spontaneity results in more reliable numbers.

[AP]

18. Fascinating insight into Dean's support.

Or lack of it, among a certain segment of DUers. Dean regulary draws about 50-60% of the support on DU. Well, it looks like half to 5/8ths of the people who don't vote for him really don't like him. Interesting.

[TSIAS]

21. Dean is probably the most polarizing candidate

He has the strongest support, yet the most detractors. There doesn't seem to be much middle ground in most DU'ers feelings about Dean.

[AP]

22. and I wonder if you can extrapolate that to Republicans too.

Rennie408 suggested that one of the keys to Democratic success in 2004 is that you don't run a candidate who freeks out some Republicans so much that they get energized and vote against him or her.

I think Clinton in '92 was a good example of that kind of candidate, so the Republicans really turned up the bullshit to make him a polarizing figure, but even with the millions spent, the ground they made up still wasn't enough to unelect Clinton in 96, with impeachment, or, say, if he had run in 2000. He just wasn't THAT polarizing for a big enough group of people.