90% - 10% doesn't mean a 1/10 chance.
2016 US Presidential Elections | Trump Wins
- Thread starter Raoul
- Start date
- Status
90% - 10% doesn't mean a 1/10 chance.
Pretty much my thoughts.
A one in ten chance of unimaginable horror isn't a comforting situation.
No, but not wanting to die is that reason so it's not really comparable. It also doesn't change that statistically I'm overwhelmingly likely to survive. If you replace your example with one more related where the choice doesn't include death then yes I'd probably play, because they're amazing odds and you'd be a fool not to.
So you privilege fear and mysticism above science and statistics ?
I think more a scepticism of science and statistics, after a long run of mistakes from such in British politics.
Clinton+12 in the post debate abc news poll
It's not really fear and mysticism when we're just after a major vote which went against the polls to an extent, with it being a vote won by people who are arguably quite similar to types likely to vote for Trump.
US election polling may be a lot more consistent and reliable, but this is probably the most unpredictable/bizarre election we've seen in a long, long time, with lots of twists, turns and swings, and if there's any election where it's been more difficult for polling companies to remain accurate then it's probably this one, with possible influxes of new voters, changing allegiances in certain states, third parties etc.
It didn't. People were misinformed, it's not the polls' fault.
Well, bizarre as it is, polls have correctly called about 95% of the contests this cycle.
It's not the polls fault, no, but it was still surprising to many people who didn't see it coming. I'm not disputing the polls' consistency and reliability but this remains a very different contest to some previous ones and, again, it's easy to see why people remain doubtful. Even though I think Trump's chances are, hopefully, just about gone by now.
For me, the Brexit comparisons are just lazy ignorance. They smack of "I know feck all about American politics except that Trump is a populist. We just had similar sentiment before Brexit, therefore the polls in the US must be wrong and Trump will get elected President".
1. The Brexit polls seemed to be more or less a 50/50 proposition whereas the US elects based on the electoral college comprised of 50 small elections. Most statistical models (538, Sam Wang, Larry Sabato, NY Times etc) have this between 87 and 99% for Hillary.
2. The UK polling leading up to Brexit seemed oddly shallow and unsophisticated in terms of polling averages and lack of statistical simulations. They just seemed like regular every day polls with little consideration given to statistics and probabilities.
3. As mentioned before, the US doesn't operate by popular vote (which Hillary would win anyway), it operates via the electoral college where the Democrats have a clear demographic edge before a single vote is counted.
4. In the US, people begin early voting an entire month or more before the election. Millions of votes have already been completed as of today and up to 40% of the entire population will have already voted before election day on 8 Nov.
That's not to say Hillary has a 100% chance of winning, Trump has about a 15% chance of winning, which is how the race should be assessed - as opposed to 'well Brexit happens therefore Trump may win'.
Last edited:
Including those whose jobs depend on them knowing the result, who had commissioned their own private exit polls, at considerable expense.
I'm failing to see how this is not reason for us to think Trump 'may win'.
15% is pretty damn low even if you think he "may" win.
Yeah, 15% isn't some fringe, outsider candidate...it's an underdog, true, but it's still someone with a realistic chance of winning.
He may win because it's a 2 horse race and statistically there's always a chance before Clinton reaches 270 on Election night.
It's pretty much 'Dumb and Dumber' atm.
Quite.
Indeed.
Lets also not forget that those pesky old people don't come online and have a habit of throwing a spanner in the works. Damn oldies.
Not really...15% makes him an outsider but it's still someone with a feasible, realistic chance of winning. I struggle to see it happening but he's around 4/1 or 5/1 in a lot of betting markets and, again, that makes him an underdog but it's still someone with potential to win.
BTW...the 15% is just one model (FiveThirtyEight's)..which is being kind to Trump. The 538 Nowcast now has Trump at about 12.5%.
Sam Wang has this at 97-99% - http://election.princeton.edu/
NY Times has it at 93%.
Larry Sabato's Crystal Ball has one of the biggest leads for Hillary - http://www.centerforpolitics.org/crystalball/
Daily Kos - Hillary 95% - http://elections.dailykos.com/app/elections/2016
Sam Wang has this at 97-99% - http://election.princeton.edu/
NY Times has it at 93%.
Larry Sabato's Crystal Ball has one of the biggest leads for Hillary - http://www.centerforpolitics.org/crystalball/
Daily Kos - Hillary 95% - http://elections.dailykos.com/app/elections/2016
15% is not a realistic chance of winning in the slightest. Even 25% isn't. Statistically, that's so monumentally behind. It's a bigger number than I'd like, I'd like to think Americans are smarter than that, but whether it's 1% or 20% doesn't change the enormous advantage that Hillary has. The only figure that would worry me now would be 60/40.
Well hopefully this explains IBD/TIPP polls. If true, expect them to announce Clinton comfortably leading on election night
Check the link posted by Nate
Check the link posted by Nate
Let's put it like this, it's David Moyes's Sunderland vs Mourinho's United at Old Trafford. They are down 3-0 and 10 men at HT. The likelihood of them winning is about equal to the likelihood of Drumpf squeaking it, by most models.
It's essentially one in six, right? I understand that's not how elections work, of course, and 'realistic' is probably the wrong word for me to use, but he's not someone with an almost impossible chance of winning...plenty of things will have happened in sport/other betting markets this week wherein someone won something with a 15% chance of doing so.
Although I'll also of course acknowledge that it's just one model of determining the race, and as @Raoul said a lot of the other models predicting this have it at 90%+, which is more reassuring.
It's 1 in 6, but it's nothing compared to 5 in 6. You can make it sound as scary as you want until you realise that the flipside is so overwhelming that it's essentially nothing to worry about. If Trump was 1 in 6 and Hilary was 2 in 6 then sure, but that's not the case. I would put money on this being an absolute landslide, I've said for a long time that he'll get nowhere near winning.
Only by the ones that are on 99% or so. I'd imagine odds for what you're describing happening would be well, well above 100/1. Whereas odds for Trump winning are somewhere around 4/1 to 5/1 in most betting companies.
Oh yeah, I don't deny that...I've said plenty of times in the past week or two that I think Clinton will win it, and she'll probably win it comfortably. It's just that one in six isn't some rare, freak occurrence, as some seem to be implying. Again, though, that one in six probability is about the best he's being given by any reputable source, so he's even less likely to win.
I'm already resigned to the depressing reality of President Clinton (not as horrifying as President Trump granted, but depressing nonetheless), but I am excited at the propsect of the senate being flipped and Bernie becoming the budget committee chair 
Let's see how fans of the IBB/TIPP, LATimes/USC, and Rasmussen polls explain this A+ rated monster.
Nah, that's me pegging it at about 5% of him winning, about the mean between models. Stranger things have happened in football.
Betting market is essentially useless. They can't offer prohibitively high odds just in case things go tits up, it's a defense mechanism against the unthinkable rather than any foresight into the race. You will see the odds rising if polls hold steady as time runs out .Brexit shock was largely due to the market as well.
Sunderland coming to OT right now would probably struggle to get much more than a 5% chance of winning at the start of the game...they certainly wouldn't be given anywhere near 5% in the situation you describe, although that's going off topic...
And true, the betting markets are playing it carefully but my general argument is that someone with a 5-15% chance of winning is still a threat to many people. Even if only a distant, unlikely threat.
I'm prepared for the worst and won't relax until it is officially announced that Hillary has become the 1st female president of the US.I know there are people in this thread who are quite bullish about what the outcome of the election will be and have every faith in the polls etc, but we're all still going to breathe a massive sigh of relief when he loses, aren't we?
Edit: I trust polls and models to a great extent (at least those with a scientific approach, nt the online polls Drumpf likes to refer to) but don't trust that people tell interviewers their true preference of candidate. I do hope they prove me wrong.
Well my parents just told me they voted for Drumpf here in Arizona yesterday. They asked why I hadn't voted yet, I told them because I'm currently stateless.
What if she becomes the male President? Cue shock to everyone...
No matter how she improves in polls, Clintons 538 percent chances can't climb much more according to 538 themselves so they are likely understating her chances.
Yeap. Their poll-plus model has been significantly more conservative. Nate posted a graphic the other day and Clinton would need a +10 poll average to climb past 95% iirc.
I'm still waiting for the November surprise video being released. Did Trump use the N-word or is it something worse?
Still hoping to see Texas go blue.
Still hoping to see Texas go blue.
Another way to look at chances. Hilary's chances of winning Mississippi (9%) are greater than Trump winning the presidency (7%). Do you have any thought that she will win Mississippi?
Edit: based on NYT numbers
Edit: based on NYT numbers
So polls from A+ pollsters in the last ten days:
Monmouth Univ. Clinton +12
Selzer & Company + 9
ABC News/Washington Post + 12
And if we add A rated ones
Fox News +6
SurveyUSA +10
Median: 10, Mean: 9.8
I'm happy
Monmouth Univ. Clinton +12
Selzer & Company + 9
ABC News/Washington Post + 12
And if we add A rated ones
Fox News +6
SurveyUSA +10
Median: 10, Mean: 9.8
I'm happy
So polls from A+ pollsters in the last ten days:
Monmouth Univ. Clinton +12
Selzer & Company + 9
ABC News/Washington Post + 12
And if we add A rated ones
Fox News +6
SurveyUSA +10
Median: 10, Mean: 9.8
I'm happy
All these polls are basically doing is telling Clinton light voters not to go vote - her vote and position as POTUS is safe. Not what you would want in the swing states where there is 1-2% in things.
For some perspective... At this point in 2012 I believe that 538 had Romney at something like 30% to win the election.
- Status