• 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.

Do polls underestimate Trumps support?

Caper said:
I'll preface this by saying Trump should not be president. I am right wing, but Hillary is the much better candidate.

[...]

I could see myself voting for Trump (a wrong choice, just because I hate HC).
Click to expand...

That seems like an odd position to take.

If you hate HC and you think Trump should not be president, isn't it better to vote for a n other?

I think the polls are overestimating Trump support in the light of "being wrong about Brexit" and media bias/spin.

It's possible that Trump could win, but so many things have to go right that I think it's a far lower probability than 33%

Also the narrowing of the polls in recent days is likely to spur more Democrats to get out and vote.

Sadly I doubt it'll be a landslide, (any party nominating a candidate who is so ill equipped to be President deserves an absolute pasting at the polls imo) but I would be shocked if it even turned out to be close.
 
God, a Britxer who dislikes Trump. Ironic,since Trump calls himself "Mr.Britexer".
 
Argumemnon said:
Well, I find it odd that you'd admit that, rationally, Clinton's the better choice but you're going with your guts and voting against her instead. That sounds, well, irrational.
Click to expand...

First off. I'm not voting. I'm Canadian. And yes, you are right, it would be irrational.
 
Ambrosia said:
It's possible that Trump could win, but so many things have to go right that I think it's a far lower probability than 33%
Click to expand...


538 is giving Trump a 36% chance of winning right now. To me, that's a very big chance. That would have to be higher then Romney or McCain right?
 
Have Trumps businesses been taken of the sharemarket while this is going on?
 
there seemed to be a lot of wishful thinking amongst the press and pollsters during this campaign cycle. Amazing how wrong they got it.
 
TragicMonkey said:
I think polling accuracy hinges on methodology.
Click to expand...

The most important variable in polling is in the candidates themselves and issues that affect polling, but not methodologies as such. The who thing is a rerun of the 2000 election, when Bush snatched a victory from similar predictions. The difference this year was more pronounced, but not greatly - by ~3.5% instead of 3%.

Pollsters were generally correct in 2004, 2008 and 2012, but made the same mistake in 2016 as in 2000. To me this suggests that methodology is important, but all established methodologies make a fairly strong systemic bias in a specific situation. There are many similarities between 2016 and 2000, but a reasonably popular democratic two term incumbent is probably the critical one. Apparently a sizable segment of voters are more willing to go to the polls and vote Republican than answer a pollster they're going to do that in this situation.
Another important issue may be the candidates themselves, but I suspect Al Gore would be insulted if he were compared to Hillary Clinton, and George Bush would be insulted if he were compared to Donald Trump - which could explain the larger scale of the difference to some extent.

McHrozni
 
Last edited:
Brexit was the canary in the coal mine for tonight. It should have been noticed.
 
McHrozni said:
The most important variable in polling is in the candidates themselves and issues that affect polling, but not methodologies as such. The who thing is a rerun of the 2000 election, when Bush snatched a victory from similar predictions. The difference this year was more pronounced, but not greatly - by ~3.5% instead of 3%.
Click to expand...

Bush was favored in most polls going into the 2000 elections; in polls taken in October of that year, he held the lead in 10 out of 14 polls; Gore led in only two (both from early October) and the other two were tied. In the two polls taken in early November, Bush led by four percentage points.
 
Last edited:
Caper said:
Brexit was the canary in the coal mine for tonight. It should have been noticed.
Click to expand...

Those of us who kept bringing it up were accused of all kinds of things :(

Ironically, this will be bad news for Brexit because Trump will want any future trade deals to be heavily weighted in favour of the U.S. Any hope that the U.S. will fill any void in UK exports will be dashed.
 
Brainster said:
Bush was favored in most polls going into the 2000 elections; in polls taken in October of that year, he held the lead in 10 out of 14 polls; Gore led in only two (both from early October) and the other two were tied. In the two polls taken in early November, Bush led by four percentage points.
Click to expand...

Right, polls made an error in favor of Bush, not Gore, my bad.
Same thing as here but the opposite way, apparently.

I think we need to give up on finding the perfect methodology and focus on individualizing methodologies based on the specifics of every election.

McHrozni
 
cullennz said:
Clinton will win by a landslide

You heard it here first

The media are just hyping it up.

She will have the majority of the swing states sewn up and that is all she needs
Click to expand...

Bwaaahah!

Holy crap! I am really sorry to see Trump win, but the consolation is seeing all the overconfident dick-swinging "skeptics" making ridiculously over-the-top predictions which turned out to be wrong. There should be a lesson in humility here about how certain we should allow ourselves to be, but if Trump winning the Republican nomination didn't work I doubt this will. It seems to me that the cautiousness with which fivethirtyeight forecast a likely victory for Clinton was a result of the Republican nomination process.
 
You must log in or register to reply here.

Back
Top Bottom