General Discussion
Related: Editorials & Other Articles, Issue Forums, Alliance Forums, Region ForumsCan we please stop with the "polls were wrong in 2016" narritive
I will concede that nearly everyone I spoke to thought Hillary would win in 2016 against Trump. But the polls were not wrong. The national polls had the race a slight Hillary popular vote win. She won by over 2 points. Were some of the state polls wrong? Yes, but by small percentage points, well within the margin of error. If there's anything that might make a poll wrong it's enthusiasm on one side or another in the actual voting. That may have happened within a few states, as older, first time voters without voting history appeared at their precincts. This wasn't an overwhelming number, but just enough to change the anticipated results in a handful of states. But remember, Hillary won the national vote by 3 million.
Right now, polls have Trump nearly 20 points under water. No fucking poll is 20 points off. Let's not start discounting very accurate information because of some misplace narrative. I'll give you a poll that was off. Virginia's Gubernatorial election in 2017. I left work that Tuesday, and the polls had that election one point either way. In the end, Nordstrom beat Gillespie by nearly 11 points. The same in Alabama and Pennsylvania, by a lesser extent. If the polls are missing anything this cycle, it's the commitment of Democratic voters and the enthusiasm of newly registered voters by our side. As a matter of fact, I think that's how Rasmussen has tilted their numbers to become the outlier. Rasmussen is using a "likely voters" model that tracks the voting history of 2014 and 2016 and is missing the huge resurgence in the enthusiasm of the Democratic coalition.
= new reply since forum marked as read
Highlight:
NoneDon't highlight anything
5 newestHighlight 5 most recent replies
TCJ70
(4,387 posts)...the wrong thing was the Electoral College once again denying the will of the people.
lark
(23,091 posts)They used the electoral college against us and hacked the vote just enough to win in 3 states they weren't supposed to.
mythology
(9,527 posts)Wisconsin and Michigan did recounts that were reviewed by the Washington Post and guess what, no votes were changed. That's going to be true no matter how many times people ignore evidence.
https://www.washingtonpost.com/news/monkey-cage/wp/2017/06/06/were-2016-vote-counts-in-michigan-and-wisconsin-hacked-we-double-checked/?noredirect=on&utm_term=.c047d6345753
uponit7771
(90,335 posts)There doesn't need to be evidence of vote tempering for it to exist.
That has been proven many times.
A recount only has meaning if it is hand counted.
I have tried to find information about the recount procedures in those states but have not been able to.
I'm my state, and I believe most, a recount just means you have the same machines count the same ballots again.
That is meaningless.
Anyome who has been studying this seriously knows that if the memory cards are tampered with, there is absolutely nothing you can do but hand count.
If you have touchscreen voting machines you are sol.
I don't care what the Washington Post says. They are not experts on this and you have to be to grok it.
CentralMass
(15,265 posts)"
While Trump managed to gain an electoral college victory, not only did he get fewer votes than Hillary Clinton — a fact that, remarkably, seems to merit nothing more than a footnote in almost every discussion of the election — he got fewer votes than Mitt Romney in 2012, fewer votes than John McCain in 2008, and fewer votes than George W. Bush in 2004. In total, fewer than 26 percent of eligible American voters cast their ballots for the man who will occupy the Oval Office come January."
"What’s also important here is how poorly Hillary Clinton did. She got 6 million fewer votes than Barack Obama did in 2012, and nearly 10 million fewer than he did in 2008:"
"Third, this was the first presidential election since conservatives on the Supreme Court gutted the Voting Rights Act, which allowed Republican-controlled states to pass a series of measures meant to suppress the votes of those who were likely to vote Democratic, particularly African Americans, Latinos and college students. In some of those states on which Trump built his victories, Republican-designed voter suppression laws, including ID mandates, limits on early voting and a reduction in polling locations, seem to have had their intended effect. As Ari Berman noted:
27,000 votes currently separate Trump and Clinton in Wisconsin, where 300,000 registered voters, according to a federal court, lacked strict forms of voter ID. Voter turnout in Wisconsin was at its lowest levels in 20 years and decreased 13 percent in Milwaukee, where 70 percent of the state’s African-American population lives.:"
If Iremember from another article the final numbers show voter turnout in WI was down 19% (17% in Milwaukie).
lark
(23,091 posts)There were numeous instances, far above the norm, where there was no presidential vote in heavily urban area, hmmm. They said the chance of this happening naturally were astronomical. This was found in several of the large MI urban areas, but not in the rural ones. Same thing happened with heavily Democratic Broward county in FL and in heavily Democratic county of Hialeah, but nowhere else in the state. I didn't say they switched votes, they deleted Democratic votes in heavily Democratic areas and no, that was not considered.
raccoon
(31,110 posts)
yuiyoshida
(41,831 posts)maybe? 
TCJ70
(4,387 posts)
ProfessorGAC
(64,995 posts)The justification was to prevent crazy populism from usurping an election. (I know there were other elements, but that was the justification.)
Now, electors just lock step vote on a winner take all basis. No conscience, no thought. They just rubber stamp the state's outcome on a winner take all basis.
It would have taken a very limited number of electors of conscience to say "No, i'm not letting that guy be POTUS." But, that didn't happen.
underpants
(182,769 posts)Trump won Wisconsin Michigan and Pennsylvania by slightly larger numbers than attend a Friday night football game in some towns. That was the difference.
triron
(21,999 posts)But I don't think so. The Russians in league with the Republicans manipulated the vote totals.
The probability this did not happen is enormous. Read Dr. Ron Baiman's affadavit.
mythology
(9,527 posts)Some media outlets post preliminary data when the polls close—that’s the supposedly raw data that, according to the conspiracy-minded, reveal the fraud. But those data have already been merged with the results of those telephone interviews, and they have already been adjusted throughout the day (the interviewers send in their survey results in three waves). Unadjusted data are never released. (If you Google “exit polls adjusted New York,” you’ll get back dozens of posts claiming that the “unadjusted exit polls” varied significantly from the final results. All of those posts are dead wrong, as none of their authors have any idea what the unadjusted data looked like.)
For writers like Fitrakis, these adjustments are inherently sinister. But Lenski notes that all surveys, election-related or not, are adjusted to factual data. “Every telephone survey or online survey is weighted according to Census figures,” he explains, “or if it’s a pre-election survey they’ll weight the data to match the demographics on a voter registration list.” (See here for a detailed explanation of how and why pollsters have to adjust, or “weight,” their samples to get accurate results.)
When the polls close, pollsters don’t adjust the data to “match the official results.” They use the official results from the relatively small number of polling places where they conducted interviews to refine their sample. For example, if their model assumed that 30 percent of voters at a polling place would be black, and that number actually turns out to be 20 percent, or 40 percent, then they’ll weight the data accordingly. During this period, they’re also entering any surveys that were sent in late (again, this is all based on incomplete data).
These adjustments, and the inclusion of those late surveys, can account for significant shifts between the preliminary numbers posted when the polls close and the release of the final exit polls an hour or so later. But the important point here is that the final results are more accurate, not less so.
paleotn
(17,911 posts)If anyone should know better, it's a damn mathematician. Then again, there are quacks in every field.
Amemenhab
(4 posts)Last edited Wed Sep 5, 2018, 02:02 PM - Edit history (1)
You’ve covered the difficult step of weighting, yet exit polls are unfortunately an availability sample, not a stratified random one. Thus we have a worsened self-selection bias problem. I suspect voters who declined to answer the polls were more likely to pick Trump than voters who did respond. There’s nothing pollsters can do about this, for they weight by demographic group—and many non-respondents here are voting against the trend in their demographic.
As example consider black Trump supporters even if few these be. Why, they were ashamed to vote for the man! How could they reveal their choice while so red in the face? Or, succumbing to social desirability bias, they lied to the pollster. Such reasons cause me to take advance and exit polls alike with a grain of salt. This is an instance of what’s called the Shy Tory factor.
For advance elections polls, Nate Silver’s Five-Thirty-Eight web site employs a corrective model where the data from certain groups of demographically similar states are assumed to be correlated, which allows for a revised estimate of the probability that a given lead in the polls will yield a win. An excellent discussion of the Shy Tory factor with a link to Silver’s explanation is found in post #32 on this thread.
garybeck
(9,942 posts)And easiest to hide fraud
Pachamama
(16,887 posts)....then I believe that there may even be something to look at very closely with tampering of the votes in those states. It is always easy to say "well, no one expected those voters would come out and vote or those voters came out in droves..."
I still believe that in states such as Pennsylvania, Michigan, South Carolina and Wisconsin and possibly others, that Russian hackers did something to change vote tallies in machines and the polls were actually accurate all along....
at140
(6,110 posts)Trump supporters were ashamed to admit to pollsters they favored him. So basically they lied to the pollsters.
Pachamama
(16,887 posts)...that explanation by anyone.
We heard that same BS in 2000 too when Bush won in Florida.....and same for OH in 2004....
Right....
PS: Also, people who voted for Trump were proud to say so....they loathed Clinton and even if they held their nose voting for Trump, they didn't hide it
at140
(6,110 posts)and what I said previously is what I heard on TV, that many Trump voters refused to talk to exit pollsters. Note that there is no legal requirements for voters to respond to pollsters.
Is that correct? I have no way to prove either way because like I said, I never worked as a pollster. So all I can do is rely on TV news.
NCTraveler
(30,481 posts)
erpowers
(9,350 posts)I am not buying the idea that the polls were wrong because Trump supporters were ashamed to admit they were voting for Trump. There were reports from multiple swing states of tens of thousands of people not voting for either presidential candidate. I am also not really buying that people, especially black people, went into voting booths and refused to vote for either candidate in the 2016 election. My information may be wrong, but I think it was reported that around 70,000 people in Michigan, maybe Detroit alone, did not vote for either candidate. I do not believe 70,000 people in Detroit, knowing the stakes of the 2016 election, did not vote for Hillary Clinton. I am not trying to claim that votes were changed, but I think something happened where 70,000 votes did not get counted. If those votes had been counted it is likely that Hillary Clinton would have won the election. It was also reported that tens of thousands of people in Florida did not vote for either presidential candidate. Maybe that actually happened, but do you really think tens of thousands of people could not see enough of a difference between Hillary Clinton and Donald Trump in order for them to vote for one of the candidates?
Then there are the reports that tens of thousands of people were prevented from voting due to voting law that made it hard for people to register and stay registered to vote. There been multiple reports that the number of people who were likely prevented from voting outnumbered Trump's margin of victory in each of the swing states that he won.
I do not think the polls were wrong. I think the polls could not have taken into account how voting laws prevented people from voting. I am willing to accept that Hillary Clinton not being candidate that could excite people may have hurt her chances of winning, but I am still willing to believe voting laws had more to do with Hillary Clinton's lose than polls being wrong.
at140
(6,110 posts)I have no argument whatsoever with you that many voters were prevented from voting by one trick or the other.
My only point is about exit polling, which in the past has been the most accurate.
And what I said previously is what I heard on TV, that many Trump voters refused to talk to exit pollsters. Note that there is no legal requirements for voters to respond to pollsters.
Is that correct? I have no way to prove either way because like I said, I never worked as a pollster. So all I can do is rely on TV news.
grantcart
(53,061 posts)This year is a problem because it is much more difficult to predict
1) state polls are more infrequent and the key to polla is watching the change of the same polls in the same place.
2) off year elections have wide swings in turnout
3) the intensity of emotion on the Democrats side cannot be measured.
For example we have a lot of data on "generic congressional ballot" but we have non generic outstanding candidates, including a phalanx of gifted highly intelligent women.
We are going to outperform predictions.
tblue37
(65,328 posts)IOW--young voters.
louis c
(8,652 posts)I don't know how they get the numbers but I got a cell call myself from a pollster.
http://www.pewresearch.org/fact-tank/2016/01/05/pew-research-center-will-call-75-cellphones-for-surveys-in-2016/
Codeine
(25,586 posts)Oddly, I was never polled back in the ancient mists of the Landline Epoch.
Bernardo de La Paz
(48,988 posts)
brooklynite
(94,502 posts)Last edited Mon Sep 3, 2018, 04:28 PM - Edit history (1)
First, favorability does not automatically determine electability, and second, Trump isn't a candidate on the ballot.
louis c
(8,652 posts)there is a direct correlation between the incumbent party's president's approval rating and the outcome of a mid-term election.
The proof is kinda like every mid-term since 1946
braddy
(3,585 posts)progree
(10,901 posts)From some Googling I did recently.
And so.... (?)
On Edit: Obviously a lot of "wasted" votes, if just 2% of that 4.27 million votes had occurred in Wisconsin, Pennsylvania and Michigan (which were lost by a combined 78,000 vote), we'd have had a different outcome (78,000 / 4,270,000 = 1.83%)
braddy
(3,585 posts)
Bradshaw3
(7,513 posts)The polls were within the margin of error when it came to the final results.
deurbano
(2,894 posts)so Comey pulling what Comey pulled (with all the attendant and coordinated magnification of that by Russian bots and others) could have an outsized, last-minute effect hard to capture in polling-based predictions (since it was so unprecedented)... at a time when it was too late to be overcome with the actual facts. Hopefully, it will be harder for them to fight a multi-front war in all those congressional races without that one big target to affect the down-ticket races, too. Also, even some leftists thought it could be a good idea to just "blow things up," assuming something better would emerge from the rubble. Hopefully, the more rational of this contingent and the more rational independents will haver had enough of this "experiment" by now.
bucolic_frolic
(43,128 posts)which included Rudy's cabal which lived there forever and which mingled with the populace in those areas which included little Russia in Brooklyn (forgive my understanding of NY geo). So the mind control was not just occurring on Facebook.
deurbano
(2,894 posts)bitterross
(4,066 posts)There was enough crap going on in swing states to nullify that win.
bucolic_frolic
(43,128 posts)You've nailed it again, louisc. In my local voting place, I've seen two close elections, one local, and the presidential one in 2016. In the local, a prominent businessman was trying to take over the town to gut zoning, basically. Every last uneducated, high school dropoout, and their parents and cousins, were at the polls in jeans, Sears polyester shirts, and workboots. These people gave not a hoot about politics. How did he motivate them. I don't know for sure, but I do think there was motivation of some sort.
In 2016, these same people voted. Everyone in a pickup truck who thinks you vote a "the poles" was there in full regalia, the same as the local businessman turned out. What was their motivation? Again, these people never thought about politics. It's a big deal when politicians start promising voters something, and come to town with jobs, loans, BBQ, and promises of paradise. How were they motivated? See the last sentence of the previous paragraph.
So there's your assignment louisc. "older, first time voters without voting history". Find out why and how they decided politics was their sudden thrill, these people who care about coffee, donuts, Dodge trucks, and football.
louis c
(8,652 posts)with Trump, it was racism. No party has offered an open racist for President since Woodrow Wilson. The closest I can come was Wallace in 1968, or Thurmond in 1948, but neither were major party candidates.
Racists will vote against their own best interests, because they a racists and bigots first and their personal best interests come second. We all know a racist or bigot who has turned his back on his children for who they dated or married. This was not in his best interest, but he served his racism or bigotry first.
Now, for my opinion on the voting dynamic, these people who voted for the first time for Trump will not come out for a mid-term. I believe they are being qualified by Rasmussen as likely voters, but I think they are not, especially in 2018. I think your enthusiastic voters who are registering for the first time, or sat out 2016 because they couldn't cotton to Hillary (I think they learned a hard lesson), who are, maybe, 18 to 26 years old are not recorded in the polls as likely voters. But, it is my opinion, they will not miss this election to vote Democratic under any conditions.
Thus, the enthusiasm gap. I work on GOTV and, let me tell you, it's a lot easier to get a person to the polls who wants to vote than one who spends all day trying to avoid you. I think we have a lot of the former this year.
AlexSFCA
(6,137 posts)We have witness a complete failure of the US intelligence and justice.
NRaleighLiberal
(60,014 posts)
k8conant
(3,030 posts)Pachamama
(16,887 posts)I completely agree
Mike Nelson
(9,951 posts)… the polls were correct and Hillary won. They were not polling the Electoral College. The danger is that more pollsters will change their methods to favor Republicans. In fact, I predict this will happen.
Awsi Dooger
(14,565 posts)That's been an increasing tendency, somehow the conventional wisdom even though it is brutally misplaced.
Then everyone fixates on certain states as if those polling margins are absolute and if we have enough state polls then they have to be correct. It is garbage and has always been garbage. Many states are simply modeled poorly and it doesn't matter how many times they are polled if the sampling is ridiculous. I've emphasized since I joined this site in 2002 that Georgia and Alaska polling is awful, invariably overstating the Democrat. Somehow I get push back even though it has proven out time and again.
Right now we have a strange new dynamic of paranoid white right wingers with one stupid thought after another, and part of it is distrust of media or any type of organization they associate with media. So that's something that can't be ignored in relation to polling. I remember discussing it here with Febble and OnTheOtherHand as far back as 2005, the transfer of the Shy Tory tendency to this country. And the landscape has exploded exponentially in that direction since 2005. In the heavy white midwestern states I don't see how we can fully trust the polling if it is sharply in favor of Democrats.
The states are not independent of each other on election day. Nate Silver was one of the few who spotlighted that aspect repeatedly in the final weeks of 2016. If there is polling error it would attach to one similar state after another:
https://fivethirtyeight.com/features/election-update-why-our-model-is-more-bullish-than-others-on-trump/
"In 2012, Obama beat his polling by 2 or 3 percentage points in almost every swing state. The same was true in 1980 when Ronald Reagan won in a landslide — instead of the modest lead that polls showed a few days before the election — and claimed 489 electoral votes by winning almost every competitive state. You also frequently see this in midterms — Republicans beat their polling in almost every key Senate and gubernatorial race in 2014, for example.
Basically, this means that you shouldn’t count on states to behave independently of one another, especially if they’re demographically similar. If Clinton loses Pennsylvania despite having a big lead in the polls there, for instance, she might also have problems in Michigan, North Carolina and other swing states. What seems like an impregnable firewall in the Electoral College may begin to collapse."
louis c
(8,652 posts)I think young people, educated white women, minorities, especially black women, are being undercounted in the polls.
I think the landslide this time will be blue.
lostnfound
(16,173 posts)There’s always a story they tell to explain it
uponit7771
(90,335 posts)I pray this enters the conversation more widely.
triron
(21,999 posts)
former9thward
(31,981 posts)Clinton had negative favorability polls all through the campaign. What poll has Trump losing by 20 points to a specific Democratic candidate?
louis c
(8,652 posts)Since 1946 every mid-term election has mirrored the approval rating of the incumbent party's President. The better the President's approval rating, the fewer seats the incumbent party lost. The worse the president's approval rating, the more seats lost. Below are the results of the elections and the approval of the incumbent party's president.
I hate to burden you with imperial evidence, but here you go.
Link:
http://www.presidency.ucsb.edu/data/mid-term_elections.php
Initech
(100,063 posts)Please????
louis c
(8,652 posts)beachbum bob
(10,437 posts)so, no the polls were not wrong
mahatmakanejeeves
(57,393 posts)Ralph Shearer Northam (born September 13, 1959) is an American politician and physician serving as the 73rd and current Governor of Virginia since January 13, 2018. A physician by occupation, he was a member of the Medical Corps from 1984 to 1992. Northam served as the 40th Lieutenant Governor of Virginia from 2014 to 2018 prior to winning the governorship against Republican nominee Ed Gillespie in the 2017 election.
Tarc
(10,476 posts)Not everyone wants to admit to a pollster that they support a white supremacist.
The Bradley effect (less commonly the Wilder effect) is a theory concerning observed discrepancies between voter opinion polls and election outcomes in some United States government elections where a white candidate and a non-white candidate run against each other. The theory proposes that some voters who intend to vote for the white candidate would nonetheless tell pollsters that they are undecided or likely to vote for the non-white candidate. It was named after Los Angeles Mayor Tom Bradley, an African-American who lost the 1982 California governor's race despite being ahead in voter polls going into the elections.
Honeycombe8
(37,648 posts)The Russians hacked several key districts that were close. That's all it took.
Trump was not elected. 12 polls weren't wrong. And not all were w/in the margin of error, as I recall.
RhodeIslandOne
(5,042 posts)Wisconsin.....Michigan.....Pennsylvania.....
A few votes more here......a few there.....
progressoid
(49,978 posts)paleotn
(17,911 posts)is predicting the future. It is not an exact science. I don't think it's a science at all. It's an art. We struggle with many of the same issues as polling operations. Taking into account and weighting the potential impacts of drivers we know about, while struggling to discover and understand those we don't. To uninitiated exec's and vp's, I impress on them up front that we are wrong. We will always be off. What is important is how far off we are. And more importantly, why. Assumptions are documented religiously and measured against actual outcomes as a postmortem analysis, allowing us to improve accuracy...sometimes. But we will rarely, if ever be right. If by some miracle we are "right" it's by sheer chance. The realm of probabilities. Given that, I've been asked many times why our function exists. My answer is, without us, they wouldn't have a fucking clue where we're going and couldn't plan shit. That's a bad way to manage.
The same holds true for polling operations. Any attempt to measure human endeavors of any kind is fraught with error, misunderstanding and omission. But, it's the best system we have ever had for predicting future events. They're not prophets so don't treat them as such. They predict as close to reality as they can, given the tools available and our constraints within the dimension of time.
Amemenhab
(4 posts)Nice post. The problem with election polls isn’t that they’re wrong, but that voters enter the booths some days after having responded to a poll and might not vote for the same person they’d named in that poll. Candidates wanting to use poll data for campaign strategy then face your problems of unforeseeability.
Hillary obviously didn’t whistle-stop Wisconsin, Michigan or Pennsylvania aggressively enough in October 2016, omitting the first of these altogether, yet she couldn’t have known this before November 8. As few of her poll leads ran a decisive 20%, she had to make decisions about where to campaign and she felt Ohio, Florida and North Carolina more important. Florida has 29 electoral votes. Regardless if she lost the states in the latter group, waging battle there was crucial and the window of time and opportunity wasn’t large enough to include all six states.
Therefore, Hillary probably made the best campaign decisions using available information in a gamble Trump happened to win by chance against the odds.
gopiscrap
(23,756 posts)populistdriven
(5,644 posts)Trueblue Texan
(2,425 posts)Russians influencing the election by changing the vote tally in at least 3 important states. Yeah...the polls missed that completely.
