Why didn’t libs learn anything from last election - polls don’t matter?

Remember last election? Hillary was certainly going to win, it was just a matter of how long it would take to declare, the margin etc... Trump had no chance.

The polls don’t mean a thing. For some reason they had it completely wrong last time. Also, the polls had BREXIT clearly not happening and in Australia the Labor Party was going to win the election last year and the UK election was going to go down to the wire. I think that the polls either deliberately just ask left-leaning areas or conservatives lie about their intent. 

13 Answers

  • 2 weeks ago

    Because democraps are entitled, whiny, self centered, intolerant, unemployed, uneducated, racist, lazy, emotionally unstable and in debt. Democraps think these are some of the better attributes. I guess living in mommie’s basement playing video games gets you far in life. The left can’t tolerate any view that is different from their twisted point of view. Now ladies, 30 somethings don’t get your panties in a knot. The party of racism throughout U.S. history (KKK) etc.,has always been the democraps. You are so miserable it’s funny. Your biggest hard hitting reply is you made a grammatical error. Ooooh. That doesn’t compare with your warped minds. The same minds that want an Alzheimer’s patient to be president. The reason all you democraps cry all the time is that your diapers are full.

    Attachment image
  • Anonymous
    1 month ago

    what else they got ????  there whole platform is based on lies ........

  • Tmess2
    Lv 7
    1 month ago

    Actually, polls do matter.  They are accurate to a certain degree.  In 2016, the national polls were accurate.  It is the state polls that were inaccurate.  Some of that was due to some states being less frequently polled, but it was mostly due to a demographic change that has now been adjusted for by most major pollsters.  

    There was a lesson from 2016.  In 2016, most pollsters did not include education as a demographic when "reweighing" poll responses.  (Reweighing recognizes the potential for random sampling to leave some groups disproportionately represented in the poll skewing the results.  As such, pollsters -- for example -- unskew the results by treating the responses from black males as if they were 6% of the respondents regardless of the actual percentage of responses that came from black males.)  When the exit polls came out, the results of the exit polls showed that education was a significant demographic with those with a college degree being more likely than those without a college degree to vote for Clinton rather than Trump.  If you look at the state results, Trump tended to outperform polls in states with a higher percentage of whites without college degrees and underperformed in polls in states with a lower percentage of whites without college degree.  After 2016, many pollsters adjusted their weights to take education into account which is a reason why 2018 polls were more accurate.

    Additionally, while some folks were overconfident in reading the polls as favoring Clinton, other folks only had Clinton as a slight favorite as the polls were still within the margin of error.  And the polls were complicated by two factors.  First, the polls showed strong results for the Libertarian and Green candidates.  Normally, support for third party candidates drops tremendously on election day.  Simply put, when people actually vote, they decide against "wasting" their vote. And there was also a significant number of undecided in even the later polls.  Because a poll can't predict how undecideds will vote, a large number of undecideds increases the likelihood of error.  Second, there was a late breaking major news story that impacted one of the candidates -- the short-lived and ultimately unproductive reopening of the e-mail investigation.  The polls showed a small but substantial movement from Clinton to Trump during that period, but there was not enough time to get enough polls to judge the impact at the state level.  The number of times that we have had such an event at the end of a  campaign can be counted on one hand.

    On Brexit, if you look at the polls, they was actually a split between those projecting a narrow win for leave and a narrow vote for remain.  The last poll showed a 4 point win for remain.  Leave won by just under 4 points, but that type of swing is within the margin of error.  An intelligent reporter would have reported the last poll as showing the race too close to call.  

    As to the 2019 UK election, I am not sure what you mean by going down to the wire.  Most opinion polls in the last month of the election had the Conservatives up by 8-12% and the Conservatives ended up winning by just under 12%.  What was going down to the wire was how that margin would translate into seats given the potential for strategic voting in a multi-party race.  The Conservatives slightly overperformed what some analysts were projecting such a result would mean in terms of seats, but everybody had the Conservatives winning a majority.

    For the 2019 Australia election, the results of the polls showed a narrow ALP win and a narrow Coalition loss.  There is some concern of herding (pollsters not releasing polls that are apparent outliers) given how little difference appeared in the released polls.  Even without herding, the difference between the last polls and the results were within the margin of error.  (Like with Brexit, a 4% swing.)

  • 1 month ago

    Because polls can be made to say what the poller wants it to.. FACT.  I ASKED 125 people if they liked Trump.. 75 said they did 50 said they did not.. 

    I only count 100 of them so 75 out of 100 like trump which means that 50 out of 100 did not depending on which 100 I decide to count and include.. 

  • How do you think about the answers? You can sign in to vote the answer.
  • Anonymous
    1 month ago

    This interpretation is actually completely false. 

    You're right that almost all prognosticators got the outcome of the 2016 election wrong.  But the polls were not incorrect, at least not at the national level.  The final set of polls showed Clinton with a two point lead over Trump and that's about how much she beat him by in the popular vote.  What the national polls did not do, because they could not, was see that Trump had a geographic spread of support such that he could lose the popular vote and, just barely by the skin of his teeth, win the electoral college. 

    The lesson of 2016 isn't that polls don't matter.  2018 should have proven that.  In the run up to that election the polls all showed Democrats doing exceptionally well.  People like you returned to your mantra of "the polls are all wrong" and many insisted that Republicans would do well, perhaps gaining seats or at least keeping Democratic gains to a minimum.  When election night rolled around it turned into the biggest midterm defeat Republicans suffered since Watergate.  Sure, Republicans actually gained seats in the Senate, but that was because of an almost comically favorable map.  In fact, given how favorable the map was, Republicans did really poorly, losing most of their chances to pick up seats and winning one of their races by such a small margin that it went to a legally mandated recount.  All in all, Democratic Senate candidates got 20 million more votes nationwide than Republicans (House candidates got 10 million more votes).  

    2016 also should have showed that dismissing the polls out of hand is folly.  In the flush of Trump's victory, many Republicans didn't notice, or perhaps chose to ignore, that the final polls were accurate as to the popular vote.  It's also worth noting that while the non partisan and Democratic pundits got the outcome wrong, the partisan Republican prognosticators got it wrong as well.  Many of Trump's fans predicted that he'd bulldoze Clinton in the popular vote, with some naively suggesting a landslide, either in the popular vote or the electoral college.  Of course, those predictions turned out to be disastrously wrong.  Trump lost the popular vote (and got less of the vote than and candidate this century) and barely won the electoral college. 

    So the lesson that you should take away from 2016 is that the polls are, probably, pretty accurate.  Now, you might say "Okay, but the polls only measure the popular vote and Trump might lose that and once again win the electoral college".  Lets leave aside the issue of how disastrous it would be for American democracy to twice have the candidate rejected by the people elevated to the presidency and just focus on the practical problems with this.  The big problem is that Biden's current lead, should it maintain into the fall, is just too big for Trump to have any chance at winning the electoral college.  The kind of popular vote/electoral college split that happened in 2016 basically happens because one candidate wins big in a small handful of states while another candidate wins small in other states.  That's what happened in 2016.  In particular, Clinton got a historic landslide win in California, the largest state.  She actually got more of the vote than native son candidates like Richard Nixon and Ronald Reagan.  Altogether, California gave her four million more votes than it gave Trump.  On the other hand, Trump won an incredibly narrow string of victories in three states that made up his margin of victory in the electoral college.  Combined, he beat Clinton in these three states by just 80,000 votes (one fiftieth of her margin of victory in California alone).  But Biden's lead is so huge right now that Trump isn't going to be able to do the same thing in 2020.  The polls are all a little different but for our purposes lets say that Biden has a ten point lead over Trump.  If we use the 2016 election turnout numbers as a baseline then that's almost 13 million votes.  And the reality is that there's basically no way that Biden will have all 13 million of those votes clumped either in a) states he was going to win anyway, or b) red states where he doesn't end up winning.  I mean, thirteen million votes is about the same number of total votes cast in California (for both Trump and Clinton).  And the reality is that Democratic numbers are already so high in populous "blue" states like New York, California, and Illinois that Biden probably can't do a whole lot better, certainly not millions and millions of votes.  Nor will those votes likely be "wasted" entirely in deep red states where he doesn't break 50%.  Again, 13 million votes is more than the total number of votes cast in Texas, the most populous Republican leaning state.  So a lot of those votes are probably going to come in "swing" states where they could change the outcome of the electoral college.  And given the shear size of Biden's lead, it's likely that he will carry a lot of those states.  Like I already said, Trump beat Clinton in the three key swing states by just 80,000 votes.  If Biden's lead holds then he will have more than 162 times the votes he needs to wipe out Trump's margin of victory.   Biden won't just carry those three key states (Wisconsin, Michigan and Pennsylvania) but will have such a lead in votes that he will almost certainly carry some or all of other closely contested states like North Carolina, Florida, Georgia, Iowa, Arizona, and perhaps even Texas.  Currently polling has Biden ahead or within the margin of error in all of these states as well as in Missouri, Kansas, and I believe Montana. 

  • Clive
    Lv 7
    1 month ago

    So why does Trump get so upset about polls showing he's losing if they don't matter?

  • Daisy
    Lv 6
    1 month ago

    The 'libs' as you say, are paying attention to the polls.  The pollsters are also polling those who were not polled in the last election.  And, no- they are not placing all of their hopes and dreams on these polls.  The Democratic party has learned from its mistakes- something the Republicans should take to heart.   Their candidate is his own worst enemy.   The Dems will not have to produce many negative ads- Trump is doing all the work for them.

  • 1 month ago

    I think it's worse now because people who answer a call from a pollster today and don't gush leftist drivel risk having their phone number posted on a leftist website as belonging to a RACIST.  

    I used to answer political polls but last year I answered one about the Governor's race here in Louisiana. It seemed like a regular poll except when I said I was going to vote for the Republican the pollster asked if I would still vote for him if he was going to eliminate the Homestead Exemption (a tax break for home owners).  I accurately told him the Governor does not have the authority to do that and I did not believe the pollster. The candidate had certainly not said anything about it.  The pollster insisted that if I wanted to save the Homestead Exemption I needed to vote Democrat.  I hung up.

    The next day I got a call from a person who claimed he was a "supervisor" with the polling organization. He also argued that the Republican would cancel the Homestead Exemption. I pointed out the fact that not only did the Governor not have the authority to do it but neither did the legislature.  The H.E. is in the state Constitution so it can only be eliminated by a Constitutional Convention and I did not believe it would happen.  The "supervisor" hung up on me.

    The Democrat won by a small margin and I wonder how many people voted for him based on the manufactured claim about the Republican wanting to eliminate the H.E.

    I also realized that the polling organization was actually a DNC pressure group disguised as a poll.  Both the pollster and the "supervisor" were very annoyed that I knew they were lying and I suspect I am lucky they did not post my phone number (or worse) on a leftist site to make me a target for harassment. 


  • 1 month ago

    Which is why you have to get out and vote.  Polls are notoriously faulty.  Statistics are notoriously faulty; so remember this come November 2020.

  • 1 month ago

    In the last election Republicans got their asses handed to them and lost the house.

Still have questions? Get your answers by asking now.