Althouse: Tracking polls — today versus yesterday.

October 31, 2016

Tracking polls — today versus yesterday.

Via RCP. Click to enlarge.

The change looks extreme, don't you think? One day?

89 comments:

robother said...: Shhh. A Reagan Revolution works best under the radar. Also, watching the Networks call the election for for the idiot is much better viewing if the shock is spontaneous.; 10/31/16, 11:36 AM
Bay Area Guy said...: The LA Times/USC poll has had Trump slightly ahead for the past several months. It's kind of an odd poll, but nonetheless, it gives me hope!; 10/31/16, 11:36 AM
Oso Negro said...: I suppose that the news of 650,000 emails overcame the fact that Hillary has a vagina. And the farce of Weiner and the menage a trois, made it too much for even the press loyalists to suppress. And there is still a week to go.; 10/31/16, 11:37 AM
Jaq said...: Traditionally the press polls stop cooking the books, as Wikileaks demonstrates they do, right before the election so as to not completely trash their credibility. In this case, it might blow up in their face because it looks like Trump momentum and will boost Trump turnout.; 10/31/16, 11:38 AM
Unknown said...: Isn't that ABC poll one where they have like 80% democrats and Hillary still can barely get over 50%?

--Vance; 10/31/16, 11:44 AM
Big Mike said...: The late deciders are breaking for Trump and folks who were Clinton leaners are rethinking which way to lean.

Oh Lordy!!! Mick may be right.; 10/31/16, 11:44 AM
Tommy Duncan said...: If Hillary loses the history books will chronicle the elements of her self-destruction. But those elements will predate her campaign.; 10/31/16, 11:45 AM
Rocketeer said...: The change looks extreme, don't you think? One day?

Even moreso when you consider those are (if I am reading the tracks correctly) three to six day rolling averages. But - the trend across all tracks is consistent.

If these tracking polls are correct, Hillary's doomed.; 10/31/16, 11:46 AM
Mike (MJB Wolf) said...: There's a poll next Tuesday that may decide this for good. Or start the court battle. Hard to say...; 10/31/16, 11:46 AM
David Begley said...: It's all over for the Dems except for the crying and the finger pointing. Hillary's finished. Weiner and Huma will be the scapegoats.; 10/31/16, 11:47 AM
David Begley said...: The big issue after Hillary's defeat is will Obama pardon her. I say no. Obama never liked her.; 10/31/16, 11:50 AM
dreams said...: Trump was gaining in the polls before the Comey news.; 10/31/16, 11:51 AM
CStanley said...: @David begley-he will have to pardon her though, to prevent his own involvement from coming to light.; 10/31/16, 11:55 AM
Etienne said...: This comment has been removed by the author.; 10/31/16, 11:55 AM
Achilles said...: The press obviously slants coverage.

There are emails available describing how they manipulate polls.

The results of the polls don't fit a reasonable random sample. The results do fit if you ascribe them to manipulative pollsters trying to push an agenda.

The people who take the NBC/WSJ actually get paid by the Clintons.

There is only one logical conclusion based on the evidence.; 10/31/16, 11:56 AM
dreams said...: "It's all over for the Dems except for the crying and the finger pointing. Hillary's finished. Weiner and Huma will be the scapegoats."

Pity the poor pervert Weiner, after all it was all a cry for help, per Roger L Simon.

https://pjmedia.com/diaryofamadvoter/2016/10/30/whos-huma/; 10/31/16, 11:57 AM
Jimmy said...: Does anyone pay attention to these polls? Especially this year, with the media being so biased. As@Mike said above, Tuesdays poll is the one that counts.
And early voting is a really bad idea. Althouse mentioned that 22 million people have already voted. Before the latest addition to the scandal(s).
oh, and BHO will pardon Hillary. The Clintons and Obama's may not like each other, but they are so deep into each others scandals that they are tied tight now.; 10/31/16, 11:59 AM
David said...: You have to know the margin of error. And the composition of persons polled.

But most of these results look to me like they are within the usual margin of error.

Hillary's support could still collapse. I do not think it has yet. Depends on the news for the next 10 days. And how Trump conducts himself.; 10/31/16, 12:00 PM
hombre said...: The change is not as extreme as it should be.

Genitalia voters and journalists are holding firm for Hillary the Grifter.

BTW, in S. Korea they are demonstrating for the removal of President Park for mishandling "secret state documents.; 10/31/16, 12:00 PM
carrie said...: This comment has been removed by the author.; 10/31/16, 12:00 PM
M Jordan said...: The models lag as do polls but the word in the street is clear: Trump will win. The question now is by how much. I say 5-8 but it actually could go higher. Reagan won the first time on a late surge by 10.; 10/31/16, 12:01 PM
Drago said...: The polls are irrelevant.

The only relevant question is how many % multiples of 100 will emerge from demand controlled precincts in swing states with Hillary select rates at 99%+; 10/31/16, 12:04 PM
Achilles said...: The people who are pushing Hillary are cutting their losses. They could have buried the latest wiener boner for 11 days if they thought they had a chance but it was clear looking at the early voting results and the internal(unmanipulated) polls that trump was going to win.

Now they can just blame it on hillarys wiener. They will use this to ignore the absolute repudiation of the political class that trump is delivering.; 10/31/16, 12:05 PM
eric said...: The polls dont matter, at all. They get closer to the truth, maybe, as we near the election. But polls from a month ago? The constant horse race talk? It's all designed to sell clicks and bolster arguments. Push turn out and donations and keep the faithful tuned in.

I'll be updating my Monday swing state polls analysis later today. Let's see how things have changed since last monday. Have they changed at all?; 10/31/16, 12:05 PM
Alexander said...: I suppose that the news of 650,000 emails overcame the fact that Hillary has a vagina.

Polls were moving before then.

No surprise - wikileaks already unveiled that pollsters were actively seeking to rig the polls to gaslight Republicans. +D samples larger than Obama had, making sure to first include independents from Hillary-favoring districts, upping the expected minority turnout while downplaying Trump's 'first time voter' support. That was how these pollish sausages were made. Texas a tossup! Hillary tied among white males! Right...

They had the means, and they had the motive. But with nowhere else to turn we Trump supporters were going to vote come Hill or high water; they have to now 'tighten' the polls to keep their position as a possible influence into the future. Of course, that narrative now will be that the polls were always right, then the FBI changed the calculus. But of course, these things were already happening.

They lie.; 10/31/16, 12:07 PM
dreams said...: The liberal media uses polls in an attempt to depress the turnout of Republicans which is why they rig their polls. They want to create the impression that a crooked Hillary victory is inevitable.; 10/31/16, 12:08 PM
tim maguire said...: Why do people care all of a sudden? We don't even know what's in the emails, all we know is they exist. What new thing have we learned about Hillary? And yet, suddenly, even some members of that lunatic fringe known as Clinton supporters are questioning Hillary's fitness.; 10/31/16, 12:08 PM
David said...: The media are going to pull out all the stops for Hillary the next few days.

Come violated the Hatch Act. Look fo.r that one. It's ridiculous but will get lots of ink; 10/31/16, 12:09 PM
eric said...: Just as a reminder, here was the swing state tracking polls as of last monday.

Currently polling says Clinton wins in a landslide.

Swing state Polls

AZ Clinton +1.3
CO Clinton +7.2
Florida Clinton +3.8
Georgia Trump +4
Iowa Trump +3.7
Maine Clinton +5.2
Michigan Clinton +10
Minnesota Clinton +4.3
Missouri Trump +6.7
Nevada Clinton +4.2
NH Clinton +8
NC Clinton +2.1
Ohio Trump +.6
PA Clinton +6.2
Virginia Clinton +8; 10/31/16, 12:10 PM
Achilles said...: "You have to know the margin of error. And the composition of persons polled.

But most of these results look to me like they are within the usual margin of error."

Garbage.

Trump wins among independents in every poll.

Trump is in double digits with black voters in every poll.(!!!!)

Trump is outperforming Romney with Hispanics in every poll.

Trump has far more enthusiasm from the base, actual voters, than Romney.

These numbers have not changed for months. The weinie pop was the oligarchs cutting their losses. They could have delayed it 11 days.; 10/31/16, 12:11 PM
Bushman of the Kohlrabi said...: It's going to be a long week for Trump. Hillary's only hope at this point is to turn the media focus from her back to Trump. And given the opportunity, the media will be desperately trying to help. We should know soon if Democrats have any more October surprises to dump on Trump. If they've already shot their load, Hillary is looking more and more like toast.; 10/31/16, 12:13 PM
David said...: "Why do people care all of a sudden? We don't even know what's in the emails, all we know is they exist. What new thing have we learned about Hillary? And yet, suddenly, even some members of that lunatic fringe known as Clinton supporters are questioning Hillary's fitness."

A lot of people know, deep in their hearts, how terrible Clinton actually is. This last episode seems to have lifted the lid they have on those feelings a little bit. But will the lid come off?

Hillary's problem is that there are a number of people who have her fate in their hands. Huma. Weiner himself. Biden. Bernie. Some media types (unlikely traitors.) But most have a lot of self interest involved in Hillary winning. An inner circle betrayal is not likely.; 10/31/16, 12:16 PM
Bob R said...: The big think here is how much both of the candidates are disliked. People are voting for the lesser of two evils and the candidates never seem to stop rubbing the evil part in their noses. People aren't really undecided. They have decided that all of their choices are bad. Doesn't take much to move them in one direction or the other.; 10/31/16, 12:20 PM
Static Ping said...: It does look weird.; 10/31/16, 12:21 PM
bagoh20 said...: The power of the vagina is no match for the power of the Wiener.

Sometimes with pride and sometimes in shame, but either way it leaves a mess behind.; 10/31/16, 12:22 PM
buwaya said...: Peter Thiel - National Press Club, today -

https://www.youtube.com/watch?v=AfYLEPRiIyE

Article -

http://www.businessinsider.com/peter-thiel-trump-speech-national-press-club-2016-10

He points even beyond the remaking of one party to a new American politics that overcomes denial, rejects bubble thinking, and reckons with reality," he said, per his prepared remarks. "When the distracting spectacles of this election season are forgotten and the history of our time is written, the only important question will be whether or not that new politics came too late."; 10/31/16, 12:23 PM
Jaq said...: I see Hillary has just edged out Trump for most hated American by two points. Good. Assange should get a Nobel Peace Prize if he keeps that neocon warmonger Caesar wannabee Goldwater Girl out of the White House.; 10/31/16, 12:25 PM
My name goes here. said...: Let's do the math (and apologies if someone else has already done this).

If we have seven days of data and we get a swing of 3 with the addition of one day.

Take the ABC poll. Assume each day is given equal weight, then one way to have a result of +2 for Clinton is if all seven days have a +2 for Clinton.

23 OCT +5 C
24 OCT +5 C
25 OCT +5 C
26 OCT +5 C
27 OCT +5 C
28 OCT +5 C
29 OCT +5 C

Now, let's remove the oldest day and add in one new day. What does that one new day have to be so that the equally weighted days result in a swing of 3 (from +5 Clinton to +2 Clinton). That would require a change net change of 21 to Trump (from +5 Clinton to +16 Trump) for the math to work.

If we weight each day on a scale so the the most current day is worth 7 times what the oldest day is and we take the above made up sample, then you have to lose the oldest day (23 OCT) and add the new day, then the new day must show a change of 12 (from +5 Clinton to +7 Trump) for the math to work.

Please correct my math. I did this in a hurry.; 10/31/16, 12:32 PM
Jaq said...: Former President Bill Clinton rallied a crowd of about 400 in Charlotte on Sunday with a simple message: Get out and vote, the earlier the better.

Yes, this has all the makings of a Democratic "wave" election... I remember when Obama won, there was a groundswell for him, you could feel it. People were talking about him at checkout counters and lunch counters. Now they talk about Trump, or nothing at all.; 10/31/16, 12:34 PM
Bob Boyd said...: I hate to be a wet towel, but click on the link and look at the little box that says 'RCP Electoral Map'; 10/31/16, 12:37 PM
SteveR said...: In the present environment, the most we can expect is that someone will make a good guess. Beyond sampling issues (how many Dems, Reps, etc) is the uncertainty of who is actually being contacted. I understand statistical certainty and so forth but given the craziness of this election cycle, I wouldn't want to admit to anything beyond wishful thinking.; 10/31/16, 12:37 PM
Owen said...: Tim Maguire @12:08: "...Why do people care all of a sudden? We don't even know what's in the emails, all we know is they exist."

IMHO their existence is the key fact. Never mind the content. What the ever-loving Heck were two-thirds of a million emails from Huma's "work," doing on a (very!) unsecured laptop belonging even in part to Anthony Weiner? Why were they even on a laptop belonging entirely to Huma? How much government business was being conducted *outside the legally-mandated IT system of accounts, devices and access protocols?

The point for me was that under Hillary, in her name and through her cut-outs and minions, a great deal of business was being conducted. Was it necessarily illegal? Well, if it was all so innocent, why did this system exist?

The guilty flee where no man pursueth.

I think Comey's letter to Congress did not require examination of the contents of the files. The simple fact that they were there, said enough.; 10/31/16, 12:41 PM
Michael K said...: "will Obama pardon her. I say no. Obama never liked her."

Hillary has $100 million to buy that pardon and Michelle has big spending habits. The money won't start to roll in for Obama for a year or more.

The early voting issue might be explained by the early voters are the people who are unlikely to change their minds no matter what they are told later. Especially Hillary early voters; 10/31/16, 12:43 PM
Gahrie said...: I have long believed that many Trump voters have been lying to the pollsters out of embarrassment or fear. Any thing that improves trump's chances of being elected weakens that fear.; 10/31/16, 12:44 PM
Big Mike said...: Do you favor the talented, experienced woman, Mrs. Bill Clinton, or are planning to vote for that sleazy sexual predator Donald Trump?; 10/31/16, 12:49 PM
Brando said...: I wouldn't put too much stock in the tracking polls just yet. We'll get a better picture in about a week or so when more polls come in. And with so many ballots cast, a small change in the polls has less impact than earlier in the campaign.

The real effect right now is this takes Clinton off message and has voters thinking about her this week. That's good news for Trump, if he can let her bask in it and not save her. Whether it'll be enough, we'll not know until next week.; 10/31/16, 12:51 PM
Nonapod said...: People are gonna believe what they're gonna believe. For myself, I refuse to get my hopes up. Bitter experience has taught me to avoid assuming too much. The only assumption I have is that the Dems have a built in advantage in terms of fraud and general ground game in the critical swing states, so I still give Clinton an edge.; 10/31/16, 12:54 PM
mockturtle said...: The media are predictably tightening up the polls in order to urge more Hillipedes to get out and vote.; 10/31/16, 1:00 PM
Ignorance is Bliss said...: Owen said...

IMHO their existence is the key fact. Never mind the content. What the ever-loving Heck were two-thirds of a million emails from Huma's "work," doing on a (very!) unsecured laptop belonging even in part to Anthony Weiner?

My understanding is that there were ~650,000 emails, but "only" ~50,000 were Huma/Clinton related. Of that, we don't know how many were copies of email already turned over to the FBI, how many were yoga/wedding related, how many were related to legitimate Clinton Foundation charitable work ( they must have done some, even if only by accident. )

If, after you filter all of that out, you are left with ~100 emails that should have been turned over, of which a couple contain classified information, then it is a big nothing-burger.

On the other hand, if it contains discussion of obstruction of the investigation ( deleting relevant emails ) or contains a significant number of classified emails ( showing that an effort was made to scrub what was turned over to the FBI ) or significant Clinton Foundation Pay-to-Play arrangements, then it is significant.

Unfortunately we won't know which until after the election.; 10/31/16, 1:04 PM
buwaya said...: "We'll get a better picture in about a week or so"

There is data on early voting trends. FL so far is running very Republican vs 2012 based on early and absentee voting. Of course, there is no way to tell if all these extra Republican-registered voters (or the very large number of Independents, this year) are voting for Trump.

This fellow on Freerepublic has been following and analyzing the FL early vote numbers so far vs 2012 - he's been at it from day 1 of absentee requests.

http://www.freerepublic.com/focus/f-news/3487153/posts

At this point based on results so far, he thinks the early/absentee vote will end up with a net early Dem vote DOWN @150,000 vs 2012. That is larger than the total Dem margin (election total) of 74,000 in 2012. Since Republicans normally turn out much better than Dems on the day, this points to, possibly, a substantially greater overall Rep turnout margin vs 2012, several times greater than the Dem margin of victory in 2012. Of course, thats partisan turnout, not voting.

Various other early vote analysts are seeing similar things in Iowa and NC.

Still too early, YMMV. But that I think is better data than polls.; 10/31/16, 1:11 PM
TRISTRAM said...: The big issue after Hillary's defeat is will Obama pardon her. I say no. Obama never liked her.

The more worrying scenario is would he pardon her if she wins?; 10/31/16, 1:12 PM
Matt Sablan said...: "If, after you filter all of that out, you are left with ~100 emails that should have been turned over, of which a couple contain classified information, then it is a big nothing-burger."

-- I'd say if there's even one more classified email, that's bad. Because then you have to ask: Why was this on Abedin's unsecured computer -- which is illegal, and why did we not find this when you handed everything else over?

Even one should be enough to land at least Abedin in jail.; 10/31/16, 1:14 PM
Bay Area Guy said...: The only issue, at this point, is whether you (the voter) are willing to do your small part in stopping the Hillary/Huma/Bill/Weiner machine or not.

Vote Trump-Pence (because the alternative is much, much worse).; 10/31/16, 1:15 PM
Ignorance is Bliss said...: Matthew Sablan said...

Even one should be enough to land at least Abedin in jail.

I agree that what we already know should have landed multiple people in jail. My point was, in terms of the election, if all they find is slightly more of what we already have, it would not change anyone's vote.; 10/31/16, 1:20 PM
Michael K said...: Speaking of machines, here is industrial strength vote fraud by machine.; 10/31/16, 1:40 PM
Yancey Ward said...: I think given the open campaigning that almost all of the media sponsoring these polls are doing on behalf of Clinton, you really should not trust any of their polls to this point. If you are completely willing to bias your news reporting to help Clinton, why wouldn't you do the same thing with the polling? Even the FoxNews poll can be questioned as to whether or not it is actually biased, at the very least, against Trump if not necessarily wanting Clinton to do well, but it was the one poll that could convince me Trump was far behind a couple of weeks ago- the others I simply discounted.

However, I have thought the election was going to be a landslide for Clinton at any point- at no point did I ever expect her to win by more than 3%- the lack of enthusiasm for her tells me her base isn't going to come close to the turnout Obama got in even 2012. So, seeing the polls tighten towards the election is something I fully expected- there is some reputational cost for getting the election badly wrong for a pollster, even if you are paid to try to shape that final outcome in the months prior to the actual election.

So, which polls should get your attention? I personally only pay attention to the LA Times poll, and for two reasons, and I only use for one purpose. The paper is clearly pro-Clinton, and the people who run the poll are academics at USC, so it is likely that they, too, are pro-Clinton all else being equal. My theory about why that poll has consistently shown Trump ahead or just barely behind since late June is simply this- the 3000+ in the sample were all selected over a half year ago, and there is no really plausible way to change that sample to give a Clinton bias- whatever bias it had came the day the final sample was set. If you are going to adjust the results, it has to come from dropping Trump supporters from the sample, and that runs the risk of being uncovered. I think the people running the poll are probably not that dishonest, and so the result is what it is regardless of how they or the paper feel about; and they have gotten pushback from pro-Clinton essayists who were encouraging them to drop certain members of the sample because they were both enthusiastically pro-Trump, and African-American.

Now, does Trump leading that poll by 4% today mean he would win the election if it were held today? I can't conclude that, nor should anyone else. We won't really know how representative the sample is until the actual election- it is possible the original sample was naturally pro-Trump to begin with, though I note that Clinton was well ahead in the late Spring and early Summer. I only use the poll to gauge the trends. In that regard, the sex-talk tape cost Trump about 3-5% support over that polling cycle in the LATimes poll, and he stopped the bleeding in that poll with the final debate. We won't be able to gauge the full effect of the e-mail investigation's resurrection until later this week, but on the first round, I would guess it is going to cost Clinton about 5% from where she stood in the poll 2 weeks ago. A one day surge like that is a bit unusual since only a fraction of the sample is repolled.

As of right now, I think Trump has a chance to win the popular vote- the early voting isn't nearly enough of the final total to matter significantly since it was always unlikely that Clinton was far ahead in the early voting to begin with. I still think, though, that Clinton has an electoral edge- she can win the electoral college without the popular vote, Trump cannot.; 10/31/16, 1:50 PM
damikesc said...: If Hillary loses the history books will chronicle the elements of her self-destruction. But those elements will predate her campaign.

Makes little sense to bury your opponent as inept. It makes a close election sad (you can barely beat the most incompetent buffoon in recorded history?) and makes a loss really embarrassing.; 10/31/16, 1:54 PM
JohnGalt said...: Regarding all the talk of our illustrious "leader" pardoning the dowager empress ... I may be mistaken but don't you have to be convicted of something before you can be pardoned ?; 10/31/16, 2:00 PM
tcrosse said...: None of this forecloses the very real possibility that Trump could run his big mouth and totally fuck things up for himself.; 10/31/16, 2:00 PM
Balfegor said...: Honestly, at this point I wish the pollsters would just release their raw data + turnout models. Because apparently some of the shift (maybe not today but at other times) is just driven by tweaks to the turnout models -- more peasant class whites here, a few more haute bourgeois white mothers there, and some more African Americans over there.; 10/31/16, 2:08 PM
Yancey Ward said...: John Galt, Ford pardoned Nixon who had been convicted of no crime.; 10/31/16, 2:09 PM
Qwinn said...: And Carter pardoned, en masse, everyone who dodged the Vietnam war draft by fleeing to Canada, if I recall correctly. No convictions necessary.; 10/31/16, 2:16 PM
Doug said...: The big issue after Hillary's defeat is will Obama pardon her. I say no. Obama never liked her.

He will ... to buy her silence.; 10/31/16, 2:24 PM
mockturtle said...: Michael K posted: Speaking of machines, here is industrial strength vote fraud by machine.

I didn't understand it but I passed it on to my grandson who is a computer geek.; 10/31/16, 2:25 PM
buwaya said...: "since it was always unlikely that Clinton was far ahead in the early voting to begin with."

But Obama was.
There seems to be an important difference in early voting turnout, 2012-2014, so far anyway. TBD.; 10/31/16, 3:00 PM
Yancey Ward said...: buwaya,

Yes, but Obama had a much more enthusiastic voter base than either McCain or Romney- that right there is the difference; 10/31/16, 3:02 PM
eric said...: Ok. Monday update on battleground states.

First I'll list Clinton up states and then trump.

AZ .6 was 1.3
CO 4.0 was 7.2
Maine 6.7 was 5.2
Michigan 6.2 was 10.0
Minnesota 6.0 was 4.3
NV 1.5 was 4.2
NH 5.2 was 8
NC 3.0 was 2.1
PA 5.6 was 6.2
VA 7.7 was 8
WI 5.7 (Don't have last week's for some reason)

Trump

Georgia 3.6 was 4
Iowa 1.4 was 3.7
Missouri 8 was 6.7
Ohio 1.3 was .6
FL 0.5 was Clinton +3.8

As you can see, most of the polls tightened. Florida flipped to Trump. Maine, Minessota, North Carolina, Virginia all grew for Hillary. Missouri, Ohio and Florida all grew for trump.; 10/31/16, 3:27 PM
tim maguire said...: Owen, it seems to me that only matters to those of us who have already decided against Hillary. Those people supporting her must have already decided that the emails aren't a big enough deal to swing their vote. Now if there were an undeniable smoking gun in these emails, that might matter. But nobody knows what's in them yet. So why have these people who have decided they don't care suddenly decided they care?; 10/31/16, 4:03 PM
Michael The Magnificent said...: Please correct my math. I did this in a hurry.

Assuming Hillary is +5 for the first five days:

Five-day moving equal-weight average, she'd have to go to -10 on the sixth day to change to +2. Assuming she stays at -10 for another day puts her average at -1.

Five-day graduated-weight average, she'd have to go to -4 on the sixth day (recent days have more weight) to change to +2. Assuming she stays at -4 for another day puts her average at -0.4.

I arrived at these numbers using a spreadsheet.; 10/31/16, 4:04 PM
Michael The Magnificent said...: While they are said to have found 650,000 emails on that laptop, they are also said to have found only tens of thousands of emails with metadata (to or from work-related email addresses) indicating they were work related (which is bad enough!). So, there's a whole bunch of non work related emails in that 650,000.; 10/31/16, 4:08 PM
JackWayne said...: My sense of the polling is this: Clinton tried the October Surprise and for a week there were people who were in shock that Trump had committed rape. Then they calmed down and remembered what the real issues are. And we have had a huge reminder of what the real issues are: Clinton is a liar and a crook.; 10/31/16, 4:31 PM
Steve said...: It is more interesting to look at the polling of Swing States individually. From these polls it looks like a one day bump of +5 to Trump in swing states. This bump moves Ohio and Florida firmly into the Trump camp, while it leaves North Carolina and Pennsylvania within the margin of error.

North Carolina

Remington Research (R) 10/30 Trump +2
CBS News/YouGov 10/26 - 10/28 Clinton +3
Emerson 10/26 - 10/27 Clinton +3

Florida

Remington Research (R)* 10/30 - 10/30 Trump +4
NY Times/Siena 10/25 - 10/27 Trump +4
Emerson 10/26 - 10/27 Clinton +1
NBC/WSJ/Marist 10/25 - 10/26 Clinton +1

Ohio

Remington Research (R)* 10/30 - 10/30 Trump +5
Emerson 10/26 - 10/27 Tie
Suffolk 10/17 - 10/19 Tie

Pennsylvania
Remington Research (R)* 10/30 - 10/30 Clinton +2
Gravis 10/25 - 10/30 Clinton +3
CBS News/YouGov 10/26 - 10/28 Clinton +8; 10/31/16, 4:53 PM
Michael K said...: John Dean has now weighed in on Watergate in the NY Times.

Pardon me while I retch.; 10/31/16, 4:57 PM
MaxedOutMama said...: They may be changing the LV filter since it is so close to the election. Or the demographic balance.; 10/31/16, 6:30 PM
Ken B said...: Mockturtle,
I think Adams's term is better, Hillbullies.
I dont recall annything in recent elections as vile as that taunting of the homeless woman on Trump's star. That was the deplorables speech acted out. They were PROUD of what they were doing.; 10/31/16, 6:37 PM
Matt Sablan said...: "I dont recall annything in recent elections as vile as that taunting of the homeless woman on Trump's star."

-- I was convinced that was exaggerated by bloggers until I saw that it was exactly as they said.

The worst part? It is entirely possible someone like Robert Creamer PAID them to do that.; 10/31/16, 6:49 PM
Curious George said...: Part of it is that Obamacare premium increases are now know, and they're ass ticklers. My salesguy went from $600 to $1000 this year, to $1600 next. That's for a $7k 60/40 deductible bronze plan. I feel bad for him, but he voted for Obama. You'd think that he learned the consequences. Nope, he early voted for Anderson. I'm sure that protest vote will feel awesome as he is cutting the check to Anthem. Dumbass.; 10/31/16, 6:56 PM
Fen said...: Nothing for me to say that others haven't already said. But it's refreshing to see so many people are getting it. Yes, the MSM is not using polls to report votes, they are using fake polls to herd votes.; 10/31/16, 7:18 PM
FullMoon said...: Of course Trump has gained ground. It is because of the FBI, KGB, Republican cabal.
James Carville lays it all out, with emotion, on MSNBC

Now is the Time for Your Tears; 10/31/16, 8:12 PM
mockturtle said...: James Carville. I remember him in Deliverance. The dueling banjos scene, right? ;-D; 10/31/16, 8:16 PM
Darrell said...: James Carville. No need to wonder what a sapient penis would look like.; 10/31/16, 8:49 PM
rcocean said...: "James Carville. No need to wonder what a sapient penis would look like."

Yeah, the biggest Dick in politics.; 10/31/16, 8:51 PM
traditionalguy said...: IMO we are seeing voter suppression being closed out as election count day approaches.

The trick was reverse engineering from using RCP Running average as the gold standard with an outlier totally oversampled poll showing Clinton +12 which drags the average to +6.

The smoke in the Media mirrors just is blowing away.; 10/31/16, 9:12 PM
gadfly said...: Nate Silver likes the USC Dornsife/LA Times Daybreak poll but he insists that you have to add six points to Hillary's number because the actual poll numbers are adjusted based upon each participants preference in 2012.

This has been a fairly consistent difference between this poll and most others. Take the LA Times poll, add 6 points to Clinton, and you usually wind up with something close to the FiveThirtyEight or RealClearPolitics national polling average. What’s the source of the LA Times poll’s Trump lean? There are good “explainers” from The New York Times’s Nate Cohn and Huffington Post Pollster’s David Rothschild. Long story short: The poll’s results are weighted based on how people said they voted in 2012. That’s probably a mistake, because people often misstate or misremember their vote from previous elections.

The LA Times, analyzing all polls as of 10/30/16 conclude that the new emails haven't hurt Clinton through the weekend.; 10/31/16, 9:18 PM
narciso said...: uh no, you think hillary is more popular than obamas was in 2012, shirley you can't be serious,; 10/31/16, 9:47 PM
Aarradin said...: Aren't those tracking polls 4 day averages?

So, say a +5 for Hillary was 4 days of +5 for a total of +20/4.

To go from +5H to +2H in one day, you'd have to have a +7 TRUMP (-7H) on your most recent day. +5, +5, +5, -7 = 8/4 = +2H

So, yeah, pretty extreme.; 11/1/16, 3:58 AM
viator said...: 1. Bogus polls are adjusting themselves to realty so they don't look too bogus.
2. Undecideds are moving towards Trump (+10 to +20).
3. Trump is getting votes from "unexpected" quarters - African American, women.
4. There is a quiet Trump vote particularly where it was exposed to ridicule.
5. Many Trump voters are excited, Democrats are the opposite.
6. Events beyond even Clinton's control are working against her.; 11/1/16, 5:37 AM
viator said...: Trump coattails?; 11/1/16, 6:26 AM
jg said...: if every poll moved 3 points in that 1 day then that's insane
but each individual poll has a +-3 at least measurement uncertainty

we should ask: what happened ending on 10/23 (the day that dropped out) that was boosting hillary's numbers, or what happened on 10/30 (the day that got added) that hurt them?

i guess we know the answer. thanks, weiner.; 11/1/16, 11:59 AM