Florida survey proves landslide inevitable
Post Reply
 
Thread Rating:
  • 0 Votes - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
01-11-2016, 09:56 PM
RE: Florida survey proves landslide inevitable
You are so clueless, LDH. First of all, 538 has them in its poll. So, since you said you couldn't find them there that speaks volumes about your ability to search the Internet. Second, 311 people are the early voters in the survey, but there were more than 700 respondents. It's not a small poll, it's the first survey that has used ACTUAL voter results. And lastly, the poll at face value seems to be just another poll, but the news here is in the details. The fact that actual voters make up just about half of the survey is huge, but not as huge as the fact that of those actual voters, 28-fucking-percent of the registered GOPers voter for Hillary Rodham Clinton. Oh man that has to sting Purple Helmet!

Check out my now-defunct atheism blog. It's just a blog, no ads, no revenue, no gods.
----
Atheism promotes critical thinking; theism promotes hypocritical thinking. -- Me
Visit this user's website Find all posts by this user
Like Post Quote this message in a reply
[+] 2 users Like WillHopp's post
01-11-2016, 09:59 PM
RE: Florida survey proves landslide inevitable
I can't let myself believe Hillary will win until the day of the election itself. Undecided

Find all posts by this user
Like Post Quote this message in a reply
[+] 2 users Like Escape Artist's post
01-11-2016, 10:00 PM
RE: Florida survey proves landslide inevitable
(01-11-2016 09:59 PM)Escape Artist Wrote:  I can't let myself believe Hillary will win until the day of the election itself. Undecided

Yeah, I'm not taking anything for granted, I'm just busting LDH's balls because he only resurfaces when he can find a headline he likes. Just giving him a taste of his own medicine.

Check out my now-defunct atheism blog. It's just a blog, no ads, no revenue, no gods.
----
Atheism promotes critical thinking; theism promotes hypocritical thinking. -- Me
Visit this user's website Find all posts by this user
Like Post Quote this message in a reply
[+] 1 user Likes WillHopp's post
01-11-2016, 10:05 PM
RE: Florida survey proves landslide inevitable
I didn't say 538 didn't use them. I said I couldn't find a grade.

They already added this poll to their numbers. Something I noticed is that the Targetsmart survey is one of the few that they adjust because they don't trust the results completely.
http://projects.fivethirtyeight.com/2016...t/florida/

"Evil will always triumph over good, because good is dumb." - Lord Dark Helmet
[Image: 25397spaceballs.gif]
Find all posts by this user
Like Post Quote this message in a reply
01-11-2016, 10:08 PM
RE: Florida survey proves landslide inevitable
That's a very good sign, and I expect that you will hear a lot about it tomorrow, but it's a bit overblown.

- There *is* an uncertainty on that. Doesn't matter that they actually voted. You're looking at a subsample of the population and making assumptions about how representative that subsample is. You're minimum uncertainty, assuming you did everything else right, is dictated by counting statistics. Your minimum possible uncertainty is 1/sqrt(n) where n is the sample size. That's why bigger samples sizes are better, more representative, but offer diminishing returns. In this case, 1/sqrt(311) or ~5.7%.

- Early voters aren't a great indicator of the population as a whole. 538 has a decent article on it but basically you're looking at people who self-selected. These are the people who were motivated to get out and vote early. Their mindset can differ substantially from the regular slobs who'll show upon the 8th. That biases the results in ways that are very difficult to predict so it's hard to say what, if anything they mean.

- What this does mean is that Hillary is already building up a decent lead before the 8th. Assuming the numbers are accurate (could be lower, could be higher too!) and a nominal 25% early vote share, Hillary will start Nov. 8th with a 2% advantage over Trump.

- The story itself is probably more important than anything. Spelling >Certain Doom<™ for Trump, it could depress his turn-out in Florida. His rabid supporters will turn out come hell or high water but Republicans who are on the fence may see it as a waste of their time when they're only luke warm about Trump to begin with.

- More entertainingly, it could propel Trump into the sort of thin-skinned knee-jerk attack that he's been avoiding lately on account of Kelly Conway putting a password lock on his phone. In a self-fulfilling prophecy it could make him blow his lid and lose himself the election.

- The potentially high Republican vote for Hillary is an excellent sign, though difficult to interpret. I'm predicting a lot of "shy Democrats" this election. Republicans who are going to vote Hillary but are reluctant to admit it in a poll. I'd like to see data from Texas on this. Do they have early voting? Hillary's excedingly unlikely to take the state but Texan Republicans are the type that loathe Trump and may well hold their noses and vote Democrat. Almost certainly not enough for her to win the state, but enough that "shy Democrats" should show up well in what ought to be a very red state.

---
Flesh and blood of a dead star, slain in the apocalypse of supernova, resurrected by four billion years of continuous autocatalytic reaction and crowned with the emergent property of sentience in the dream that the universe might one day understand itself.
Find all posts by this user
Like Post Quote this message in a reply
01-11-2016, 10:14 PM
RE: Florida survey proves landslide inevitable
It's OK, I know you're hurting, LDH. I won't pile on you anymore.

It's just funny to me how the Trump campaign keeps talking about these hidden voters that are going to emerge on Election Day because they didn't want to say they were voting for Trump. Two things about this: What does it say about your campaign when you admit to the world that you're so fucking horrible that you are hoping that there are voters who are so embarrassed to admit they want to vote for you that they can't even tell a pollster anonymously that they are voting for you? And second, this survey proves the opposite, they when the GOP gets behind the curtain they are actually voting for her.

Check out my now-defunct atheism blog. It's just a blog, no ads, no revenue, no gods.
----
Atheism promotes critical thinking; theism promotes hypocritical thinking. -- Me
Visit this user's website Find all posts by this user
Like Post Quote this message in a reply
[+] 1 user Likes WillHopp's post
01-11-2016, 10:23 PM
RE: Florida survey proves landslide inevitable
(01-11-2016 10:08 PM)Paleophyte Wrote:  That's a very good sign, and I expect that you will hear a lot about it tomorrow, but it's a bit overblown.

- There *is* an uncertainty on that. Doesn't matter that they actually voted. You're looking at a subsample of the population and making assumptions about how representative that subsample is. You're minimum uncertainty, assuming you did everything else right, is dictated by counting statistics. Your minimum possible uncertainty is 1/sqrt(n) where n is the sample size. That's why bigger samples sizes are better, more representative, but offer diminishing returns. In this case, 1/sqrt(311) or ~5.7%.

- Early voters aren't a great indicator of the population as a whole. 538 has a decent article on it but basically you're looking at people who self-selected. These are the people who were motivated to get out and vote early. Their mindset can differ substantially from the regular slobs who'll show upon the 8th. That biases the results in ways that are very difficult to predict so it's hard to say what, if anything they mean.

- What this does mean is that Hillary is already building up a decent lead before the 8th. Assuming the numbers are accurate (could be lower, could be higher too!) and a nominal 25% early vote share, Hillary will start Nov. 8th with a 2% advantage over Trump.

- The story itself is probably more important than anything. Spelling >Certain Doom<™ for Trump, it could depress his turn-out in Florida. His rabid supporters will turn out come hell or high water but Republicans who are on the fence may see it as a waste of their time when they're only luke warm about Trump to begin with.

- More entertainingly, it could propel Trump into the sort of thin-skinned knee-jerk attack that he's been avoiding lately on account of Kelly Conway putting a password lock on his phone. In a self-fulfilling prophecy it could make him blow his lid and lose himself the election.

- The potentially high Republican vote for Hillary is an excellent sign, though difficult to interpret. I'm predicting a lot of "shy Democrats" this election. Republicans who are going to vote Hillary but are reluctant to admit it in a poll. I'd like to see data from Texas on this. Do they have early voting? Hillary's excedingly unlikely to take the state but Texan Republicans are the type that loathe Trump and may well hold their noses and vote Democrat. Almost certainly not enough for her to win the state, but enough that "shy Democrats" should show up well in what ought to be a very red state.

Thanks for the well-thought-out response. A couple of things: History has shown that Florida Dems traditionally lag behind in early voting (absentee, mail-in) and always play catchup. If they actually lead going into Election Day, it's game over.

I understand why larger samples are more reliable, but in this case, it's not predominantly "likely or registered voters" it's actual voters. They weren't self-selected. This firm got the info from the Supervisor of Elections (or whatever) and called the people and asked if they would take the survey. So, when the final report on this comes out, it will have where they live, possibly race, voting tendencies, party, etc.

We'll see. Texas does have early voting and the Dems are leading that, too, but there's not actual voters polled yet. Also, the surge is from Latinos, which has surpassed 100% of their numbers there at this same point in 2012.

Fingers crossed.

Check out my now-defunct atheism blog. It's just a blog, no ads, no revenue, no gods.
----
Atheism promotes critical thinking; theism promotes hypocritical thinking. -- Me
Visit this user's website Find all posts by this user
Like Post Quote this message in a reply
01-11-2016, 10:32 PM
RE: Florida survey proves landslide inevitable
In similar news, 538 is reporting that Hillary is overperforming her polls in early voting in Nevada. That is not a poll, that's voter registration data for 500,000 votes already cast. Significantly, early voters make up >50% of Nevada's vote.

The discrepancy between polls and voting appears to be ~4%, a roughly Brexit-sized error in Hillary's favour. Big Grin

---
Flesh and blood of a dead star, slain in the apocalypse of supernova, resurrected by four billion years of continuous autocatalytic reaction and crowned with the emergent property of sentience in the dream that the universe might one day understand itself.
Find all posts by this user
Like Post Quote this message in a reply
[+] 2 users Like Paleophyte's post
01-11-2016, 10:49 PM
RE: Florida survey proves landslide inevitable
(01-11-2016 10:23 PM)WillHopp Wrote:  A couple of things: History has shown that Florida Dems traditionally lag behind in early voting (absentee, mail-in) and always play catchup. If they actually lead going into Election Day, it's game over.

Encouraging, yes. Hardly game over. I won't count Trump dead until they stake the fucker through the heart and bury him face down at a cross-road.

Quote:I understand why larger samples are more reliable, but in this case, it's not predominantly "likely or registered voters" it's actual voters. They weren't self-selected.

Sorry, I was unclear. They self-selected by virtue of the fact that they voted early. Not by any selection mechanism of the survey. These are the people with more zeal, for one reason or another, so they don't necessarily reflect the general population. Still, encouraging given the historical trends.

Quote:We'll see. Texas does have early voting and the Dems are leading that, too, but there's not actual voters polled yet. Also, the surge is from Latinos, which has surpassed 100% of their numbers there at this same point in 2012.

Really? That is the story that the Dems should be grinding into Trump's face. Losing Texas would be embarassing. Laughat Evil_monster

---
Flesh and blood of a dead star, slain in the apocalypse of supernova, resurrected by four billion years of continuous autocatalytic reaction and crowned with the emergent property of sentience in the dream that the universe might one day understand itself.
Find all posts by this user
Like Post Quote this message in a reply
[+] 3 users Like Paleophyte's post
02-11-2016, 03:40 AM
RE: Florida survey proves landslide inevitable
(01-11-2016 09:33 PM)WillHopp Wrote:  But, you see, this doesn't have a plus-minus margin of error disclaimer. Of the 311 people, they ALL voted. Of the 311 people, 53% chose Hillary. Of the 311 people, 28% of its Republicans voted for HER. There is no other way to interpret this. Nearly a third of the GOP here voted for her. It's a horrible sign for Drumpf.
The polls for brexit showed leads everywhere for remain Will, and look what happened there. Don't get me wrong I love your optimism but it ain't over til its over, hopefully then the world can breathe a collective sigh of relief.
Find all posts by this user
Like Post Quote this message in a reply
Post Reply
Forum Jump: