Ricochet is the best place on the internet to discuss the issues of the day, either through commenting on posts or writing your own for our active and dynamic community in a fully moderated environment. In addition, the Ricochet Audio Network offers over 50 original podcasts with new episodes released every day.
Polling Perplexity Disappears
Last week, I drew attention to what I called the poll perplex — the fact that some polls had Trump ever so slightly ahead while others had Hillary up by as much as 12 points. The poll that had Hillary with a commanding lead was done on behalf of ABC and Pravda-on-the-Potomac.
Today, that poll has Hillary ahead by a statistically insignificant one per cent — which suggests that there has in the interim been a dramatic change in the race. But Tyler Durden at ZeroHedge reports that the only thing that has changed is the weighting of the sample. Last week, they were over weighting Democrats by a ridiculous proportion. On Thursday, when the tracking poll showed Mrs. Clinton two points ahead, they were oversampling Democrats by only 8 percent.
What we may have had on Thursday and Friday — before the FBI announced that they had reopened the investigation into Hillary Clinton’s emails — was a dead heat. The three most recent polls — all taken before that news was released — show the race tied up or Hillary ahead by one or two points. The reopening of the investigation will surely do her some damage.
At this stage, we may have a wide-open race. This has been a year of incredible surprises. We may encounter another one on the 7th of November. You had better put on your seatbelts. We are in for a wild ride.
Update: As of noon on 10/31 EST, the latest LA Times tracking poll has Trump up by four points.
Published in General
Yet more proof that pollsters can “create” an outcome. No real surprise to me. If you listen to EconTalk or study economics or statistics it becomes obvious that a person can easily choose their data to create their desired outcome.
Especially since this case is about likely voting, the statistical significance cannot be good when there are so many things that would completely change the outcome of the election that pollsters couldn’t possibly account for accurately even if they wanted to.
Thanks, Paul.
Not getting my hopes up. There is a reason that’s Trump’s questioning of the process has this kind of resonance. At this point four years ago, some people were convinced that Romney would win.
Imagine that.
The turnout here in the Houston area – from voters on both sides – has been impressive. The nomination of the two worst candidates in American history doesn’t seem to have dampened any spirits.
Then next time someone who does non random sampling claims it to be ‘scientific’ and then has polls change like this as the event approaches EVERY TIME can we just laugh out loud, or must we pretend to believe them out of politeness and media blather?
Paul,
So this is proof positive of a media / polling cabal that directly tried to suppress the Trump vote by giving a completely false impression. Someone ought to make a list of those media organizations that were part of the fraud.
I’m not vindictive. I just enjoy watching weaseling parasites squirm.
Regards,
Jim
It would indeed be surprising to find Election Day held on a Monday, but I suppose this year anything is possible.
“Drive for show, putt for dough.”
We have passed the long fairway of the campaign where the media was concerned with the impression they were sending. Now they have to justify their existence by bearing down and getting as close to the hole as possible.
I knew it!
Probably the last day for an opinion poll to come out.
Tuesday is the other kind.
I strenuously object to your characterization of the two candidates as the worst. I think Trump is the best we’ve had since Reagan, and HRC is Satan. No wonder our spirits haven’t been dampened. Get ready for the landslide, in spite of extreme vote tampering. And, exactly how do you know the the turnout is equally impressive from both sides? Wishful thinking, and harmful thinking, in my humble opinion. Those who love the USA are about to trounce all who vacillate on patriotism. And by the way, that word is OK.
Best we’ve had since Reagan? He’s certainly different from every candidate since Reagan, and nothing at all like Reagan himself… what criteria are you using to define “best?” I ask in complete sincerity.
I don’t know if it’s been mentioned or not, but the polling strategy is believed to be:
Simple and consistent.
So the pollsters at ABC/Washington Times conduct their polls as they always have and note that the explanation for the shift in numbers is that from the sample of voters polled the ratio of Trump to Hillary voters deemed likely to vote has increased is read by you all as some sort of conspiracy? Give me a break. If they were rigging the whole thing why quit now? If they were rigging the whole thing why even give the transparency in their methodology that they have? Why is it impossible to say that polls may be representative of current opinion but not predictive of future opinion, why does it have to be a conspiracy?
Yes, like all stupid explanations.
According to the local news early voting has heavily favored Democrats. Last I heard it was on the order of 200,000 votes state wide. Other than that, the polls have been swamped. My wife and I waited 45 minutes to vote (we live in a heavily Republican area). We’ve seen long lines at other polls as well.
Update: According to the Texas Monthly, the first day of early voting shattered earlier records.
Prof. Rahe,
Fool me once shame on you. Fool me twice shame on me.
I am not buying what you are selling.
Ryan, very glad to answer sincerely. By what measure do the Bushes compare to Trump? He’s a fighter and not a member of the uniparty. Don’t get me started on the Maverick and Mittens, both pushovers. Trump is the first to show some spirit, and he gets casual comparison to the Satanic HRC? Besides, a comparison of Trump to HRC is entirely unsupportable, on the bases of character, accomplishment and simple likability. Remember, I don’t believe most of what I see and hear. Cheers!
Understand. But now I have to question how the local news knows the Dems are favored? Are they tallying and publishing results already? Yikes!
Some people made the same mistake in 2012 of looking too closely at national polls. There were polls with Romney in the lead shortly before the election. We heard lots of Republicans saying “he will win!” He lost in an electoral landslide.
The state polls are what matter. And they don’t look good for Trump to obtain 270. Even if he wins Ohio and Florida, and holds all other Romney states, he still loses the election.
I love a grim realist
Ok, so what you’re saying is that he’s the strongest contender since Reagan? I don’t think it can be argued that he’s the best conservative since Reagan, and I suppose we could argue whether that is a good or bad thing, but if you’re saying that he is the most accurate reflection of public sentiment since Reagan, I might just grant you the point.
That was clearly ridiculous and barely worth laughing at, but let us not fool ourselves. Trump is the underdog and for good reason. He will not sail into the White House, he will have to fight for it tooth and nail, it is not over even this close to the vote. But Trump has yet to seal the deal, many are uncertain about him.
At 538 http://www.fivethirtyeight.com the Polls-Only forecast gives Trump a 21.1 percent chance of winning and a popular vote of Clinton at 49.4 – Trump at 44.2, a 5.2 percent lead for Clinton.
Nate Silver is very transparent about how the various state polls and national polls are applied in his model. If I were betting on the outcome of this election, I’d bet on Silver’s model.
NYSlimes
“Mr. Trump leads among white voters without a college degree by an impressive margin of 63 percent to 24 percent. He’s so strong that Mrs. Clinton has just 55 percent of the vote among white registered Democrats without a degree, compared with Mr. Trump’s 32 percent.”
Yes, you read that right, Trump is getting 32% of the noncollege white DEMOCRAT vote in a NYT Florida poll.
Not necessarily true. The state polls are already moving in tandem with the national polls. I think the pollsters have never had a firm grip on this election. Their model may have been 2012. 2016 is definitely not 2012. Clinton is NOT Obama the first black US president. Even before the latest news some swing states were moving towards Trump as the RCP graphs attest. The state polls were full of outliers similar the ABC/WaPoo poll with huge margins for Hillary that moved the RCP averages dis-proportionally.
The NY Slimes own poll shows Trump getting 32% of the white non-college Democrat vote in Florida. There are a number of indications that he is going to get a much higher proportion of the black vote than conventional wisdom indicates.
““Perhaps more frightening for the Clinton camp, the latest NBC/WSJ/Marist poll shows Trump opening up a 9-point lead in Florida and a 2-point lead in North Carolina.”
Trump is campaigning now in Michigan and Wisconsin while Hillary occasionally – but mostly her surrogates – are defending the traditional swing states. What does that tell us?
Oops.
Because no other poll shows a comparable slide among likely voters towards Trump. There was last week a fourteen point discrepancy between the ABC-Wapo tracking poll and the LA Times tracking poll. Someone was way off — and the evidence was in the sampling.
I am not selling anything. I am simply looking at the tightening of the polls and at ABC-Wapo’s radical shift.
For the record, I still think that Hillary is likely to win. She has a get-out-the-vote operation and bucketfuls of money. There are likely to be plenty of illegal aliens voting. John Podesta certainly hoped that this would happen. And Trump is a terrible candidate.
But I have been wrong on nearly everything this year, and no one really knows who will turn out. If the polls are any indication (and they may not be), it could be very, very close.
FWIW, realclearpolitics composite as of this morning shows Trump gaining steadily.