Hypothesis #1: The polls are underestimating Clinton because they don’t factor in her superior ground game....By the way, there's a new Reuters poll (9/22 - 9/26) that has Hillary up by 6. The previous Reuters poll had Trump and Hillary tied (9/15 - 9/19). The new one is still pre-debate.
Hypothesis #2: The polls are underestimating Trump because of shy Trump voters....
Hypothesis #3: The polls are underestimating Clinton because she has a lot more money than Trump and will blitz the airwaves in the last few weeks of the campaign....
Hypothesis #4 (and this one I buy): The polls are underestimating Clinton because the remaining set pieces of the campaign — the things we know will happen — play to Clinton’s strengths, all else being equal. The remaining debates, mostly....
२८ सप्टेंबर, २०१६
"What Could The Polls Be Missing?"
A conversation at FiveThirtyEight:
याची सदस्यत्व घ्या:
टिप्पणी पोस्ट करा (Atom)
पोस्ट्स
९८ टिप्पण्या:
If the polls turn out to be wrong, of course they will come up with some excuse.
However, 538 works off the premise that it doesn't matter if the polls are wrong. Because they don't care about individual polls. They care about the average of all polls. And if the average is telling us it's going to be Clinton, by half a point or 10 points, then it's going to be her.
If it isn't her, then the pollsters will tell us why we can still trust them.
Also, I think that Reuters poll shows Hillary is in trouble and groups like 538 are already trying to hedge their bets.
She gets +6 with a Democrat lean of +8.
Now, maybe if you include NY and CA and MA, you'd get +8 Democrats. But you're not going to get that in purple swing states. And Trump is winning independents by 7 points.
She needs Dem lean +2 just to tie Trump. That's going to be interesting to watch.
Hypothesis # 5. Turn out for Trump will be a tsunami of never voted before voters. But turn out for Clinton will be MIA all of Bernie's voters.
So you think Clinton being required by a judge to respond, in writing, to a series of questions from Judicial Watch, will play to Clinton's strengths?
One final note. I hope this doesn't seem like I'm spamming.
Trump gets 75% Republican support to her 82% Dem support. This is probably accurate. I know a lot of Republicans in California and Washington State who won't vote for trump on principle. But when it comes down to it, the real reason is, because their vote doesn't count in blue states.
Does anyone think in swing states Trump is only going to get 75% Republican support to her 82% Dem support?
Maybe someone should ask 538 about this.
None of them are worth very much. They could never poll me, my wife, any of my adult children nor most of my friends. And we are typical.
Clinton has the media and they will be able to hold her up because they set the background. Turnout will be the key in either direction.
They make a reference to the Bradley Effect. I've wondered about a Coakley Effect: key Democratic voting blocs who won't turn out for a cold fish woman: black males, female 'boyfriend' voters, Bernie Boys.
My biggest concern is voter fraud by the Dems, per usual.
I find it laughable that the Dems are preparing the battlefield to delegitimize a Trump win by suggesting the Russians are going to hack the results. How do we not know the Chinese won't hack the election for Hillary so that they get TPP?
As usual the Nebraska way is the best way. Ballots are "fill in the oval" papers that are read by scanner. Keep the ballots at the courthouse until election contest time lapses.
America needs to be more like Nebraska except for our idiotic ability to split Electoral College votes by Congressional district.
superior ground game
This election will surely test the conventional wisdom that having a superior ground game makes a big difference. Or, more generally, whomever raises and spends the most money will win. The post mortem pieces in the days following the election will be interesting, especially from team Silver.
What the polls have in common is that they are all run by pollsters.
The polls could be wrong if they're using the wrong methodologies--oversampling some groups and mis-estimating the groups they're not getting in phone polls--but for the average of the polls to be off beyond the margin of error the mistake would have to be widespread among different polling firms.
The thing is, this year polls of the parties' primaries have been very accurate (with the exception of the Michigan Democratic primary) so for the general election polls to be off we'd have to account for why they're so different in the general election. Maybe a large contingent of voters who weren't motivated enough to vote in the primaries but will emerge, under the pollsters' radar, in November. Maybe a very large number of people without landlines who will break more for one candidate than the pollsters estimate.
In any event, no one has provided good evidence of why the pollsters are wrong this year--it's guesswork for now. And until Nov. 8, the safest assumption is this race will be close and it'll come down to turnout enthusiasm.
Trump voters don't speak to pollsters. They don't trust them. That said, Trump's lead is tremendously underestimated. But this is the kind of election that could turn on a dime. Trump will either give Reagan a run for the record or Hillary will win with a low turnout. Trump has to be Trump to win. He has to show that he will kick the entire globalist agenda into the trash bin.
" having a superior ground game makes a big difference."
A "superior ground game" includes digging up daed voters.
It worked in Minnesota. Ask Senator Coleman. I just hope Trump has some good lawyers on retainer.
A straight-up, non-rhetorical question:
Was there a primary this year, where Trump significantly outperformed the pre-election polling?
There is this ongoing narrative (epitomized by Darrell) that Trump voters don't respond to polls but they vote.
I am not so sure; what does the data show?
Turnout, turnout, turnout.
What "do" the data show, Chuck. You're welcome.
He didn't have to outperform, he only had to hit his targets. That why he knocked Baby Cruz out on his ass in Indiana. "No mas! No mas!" Baby Cruz cried. . .
Turnout is everything. McCain got more votes in his loss to Obama than Obama got in his win over Romney.
"Trump voters don't speak to pollsters. They don't trust them. That said, Trump's lead is tremendously underestimated."
If that's the case, then why were they more eager to speak to pollsters during the primaries, when he didn't outperform (and in some cases underperformed) his poll numbers? Is there something different about his supporters now?
The polls could be off, but there'd have to be something explaining why they are off for the general election (across most polling companies to skew the polling average) compared with the primaries.
This is looking like 2012 all over again--Republicans insisting the polls are not showing their true support. I'm not sure what that serves, though--if anything, you should want your side to believe it's close to motivate them.
My gut is that the kind of Republicans who won't vote for Trump on principle - the intelligentsia / NRO type - mostly live in blue states - NY, Mass, California, Washington - so their non-votes don't matter. The exception is Virginia, unfortunately, which is pretty much a blue state at this point.
Blogger Chuck said...
A straight-up, non-rhetorical question:
Was there a primary this year, where Trump significantly outperformed the pre-election polling?
I just spend 5 minutes on RCP and went back to April. Doesn't seem to go back further than that, which is a shame. Looked at Emerson (Do you think Emerson is a legit polling firm?) And they had Cruz winning Wisconsin by +5.
But to me that doesn't matter. What does matter is, if you look at the very bottom of the poll you'll see this line, "The Republican Primary was not weighted".
This isn't uncommon in primaries. Pollsters dont weight because it's much more difficult when you split the groups up into Republican, Democrat, Independent, etc, all trying to elect their own candidates. How do you weight for that?
But they do weight for the general election.
And here is my beef. Why don't they give us both numbers and let us decide? Give us the weighted and the unweighted data.
What are they afraid of?
We don't talk to pollsters because we know they are in bed with Democrats. You might have your life turned upside down like those people in Ohio had happen to them last time around. And they only donated money to Romney. Trump did a wonderful job hitting his targets. That's why he set an all-time Republican record for total votes. Sorry, Brando. Your sieve doesn't hold water.
If that's the case, then why were they more eager to speak to pollsters during the primaries, when he didn't outperform (and in some cases underperformed) his poll numbers? Is there something different about his supporters now?
Because he wasn't demonized as an evil racist during the primaries.
Also remember, a lot of primary pollsters don't weight. They just poll and ask, who will you vote for?
Pollsters have to weight during the General election, otherwise, we would all get the impression the Republican will win in a land slide. Pollsters dont want to create that impression, because they want to influence us to vote for the Democrat, not the Republican.
I'd go with #2. There's no stigma for being pro-Hillary. The left has seen to that. I mentioned that I thought Trump's candidacy was revolutionary and the man I said that to, a Bernie supporter, has treated me with absolute contempt ever since. Would I tell a pollster over the phone who I was voting for? No.
Professor, going through the updates section of his forecast, there was an Ipsos (the company Reuters uses for the polling) poll dated Sep 18-22 that showed her up by 5 points. I don't see one done more recently prior to the debate, so I don't know if their number dropped again prior to it. This daily tracker does seem to bounce around quite a bit, from what I have noticed.
A few things about 538...
It does not necessarily skew for HRC in every instance. Oftentimes it gives Trump more favorable odds than the bookmakers.
Aggregating polls is better than just going by one or a small handful of polls. Central limit theorem.
Silver often emphasizes that it is more important to look at polls state by state. Keep that in mind. Hard to speculate until we see new state polls come out.
I expected that there'd be some states where Trump outperformed the polling. South Carolina was one where he underperformed, right? I don't know; there are about 30 non-caucus states to gather up. And state polling data is notoriously plagued with a higher margins of error and less currency.
Is there -- are there -- enough data from the primaries to know whether Darrell is making a valid point, to wit that Trump voters will not respond to pollsters but will vote?
Hypothesis #5: the polls are underestimating shy Hillary voters.
Chuck said...
A straight-up, non-rhetorical question:
Was there a primary this year, where Trump significantly outperformed the pre-election polling?
In the Pennsylvania GOP Primary, RCP Average was 48.3% for Trump; voting was 58.1%
In the Wisconsin GOP Primary, RCP Average was 39.2% for Trump; voting was 48.2%.
In the Indiana GOP Primary, RCP Average was 42.8% for Trump; voting was 54.6%.
Polls bounce all over the place even when nothing has changed. That's why sites like 538 average polls. The 6 point difference in the Reuters poll is a data point which may or may not be significant. It makes good fodder for people like Allahpundit to be pessimistic, which is his jam.
"We don't talk to pollsters because we know they are in bed with Democrats. You might have your life turned upside down like those people in Ohio had happen to them last time around. And they only donated money to Romney. Trump did a wonderful job hitting his targets. That's why he set an all-time Republican record for total votes. Sorry, Brando. Your sieve doesn't hold water."
Darrell, I think you're missing my point--clearly Trump fans had no problem speaking to pollsters during the primaries, where his vote totals matched the polls. My question is what changed that makes you less likely to talk to pollsters now?
"Because he wasn't demonized as an evil racist during the primaries."
He certainly was--surely you didn't miss that. This isn't some new theme his enemies waited until after the primaries to spring.
"Also remember, a lot of primary pollsters don't weight. They just poll and ask, who will you vote for?"
They do weight, but weight differently--they'd be different populations (e.g., Ohio voters likely to vote in a GOP primary vs. Ohio voters likely to vote in a general election). Perhaps that difference in methodology could be missing something.
I don't think, though, as you suggest, that the pollsters are intentionally trying to make this look like a closer race when they see a GOP landslide. First, a lot of these polls are conducted by conservative-affiliated groups. Second, getting an election wrong by a lot (and doing this consistently) should have happened more often if that's the case--and despite similar theories, it didn't happen in '08 or '12. Plus, getting it that wrong and developing a reputation as a poor pollster seems a high price to pay for the dubious advantage of convincing your favored candidate's supporters to turn out. Which brings me to my third point--if the real number is a landslide for Trump, and the pollsters intentionally skew it to look like a nailbiter, isn't that just as likely to increase turnout on both sides? (And before you point out that Trump's voters are so much more motivated than Clinton's and would turn out even if he was up by ten points, consider that most of Hillary's support, and Stein and Johnson's support, is more anti-Trump than anything else, and a sign that supporting third parties or staying home would help Trump would be enough to get them to turn out and switch to Hillary).
We can speculate as to what the polls have wrong, and how it's different in the general election. But so far none of these theories hold water.
Hypothesis #6: The polls are underestimating John Kasich's ability to torpedo Trump in Ohio.
The conventional wisdom--for man, many years--is that Democrats have to see thier candidate in the lead--preferably by a wide margin--in order to come out an vote in an election that counts. Republicans just vote regardless of the polls. The primaries are another matter. If you think your candidate will win, you might skip voting if something important comes up. Some people knew Trump would win their State and they stayed home.
But now that we are getting ready for the real general election, it doesn't matter that Trump wasn't your first choice. Or your second. Or your third. He is now your only choice.
Updating what I said at 12:51 PM, I overlooked a series of Ipsos that showed HRC leading +4, +6, +6, and finally +5 before the debate. Forgive me if I am overlooking a possiblity that Ipsos is running two separate polls, that could be possible, and I haven't had a chance to look into that.
Hypothesis #5: Trump voters have "come home" in the last two weeks, but ultimate Clinton voters, still grieving the loss of Sanders, have yet to coalesce. An admittedly unscientific poll, three of my students, were Bernie-bros (actually Bernie-bras) who are moving to Hillary but not quite there. As much as we keep abreast of politics, we forget most don't have the time or inclination. The debate was eye-opening to them.
The thing that sickens me most is it didn't need to be this way. The media pushed Trump and shut everyone else down, and our most well spoken articulate candidates were shut out for Trump.
Those of us who were skeptical of Trump's chances knew he would be a weak debater. We were promised by the koolaid drinkers that Trump would smash her in the debates. Still waiting.
grimson where are you getting your numbers?
Just asking if you see any states where Trump failed to outperform the state polling. (That is RCP averages for each state, right?)
and our most well spoken articulate candidates were shut out for Trump.
Yeah, the ones supporting open borders and amnesty. When they called Trump a racist and a monster and they wouldn't even agree that a moratorium to allow proper vetting of immigrants to insure Americans' safety was necessary, we didn't need to hear another word from them. If they were in the race now, they would already be out of contention. If a candidate is going to do the same thing Hillary is going to do, you might as well vote for Hillary.
Well Darrell, we shall see--anything could surprise us this year. I just wouldn't be counting too much on a rout either way--I think like the last several elections this one will be pretty close.
"Those of us who were skeptical of Trump's chances knew he would be a weak debater. We were promised by the koolaid drinkers that Trump would smash her in the debates. Still waiting."
We were actually promised that Hillary would be wheeled out on a gurney, and leave in tears after Trump got through with her. And that Trump would destroy the moderator if he tried to pull a Candy Crowley, unlike that wimp Romney.
The one upside of that for Trump is he has two more debates, and if he can actually take advice there's an opportunity to turn it around.
April, no doubt Cruz is a champion debater, but he really stumbled on the NY values exchange. Trump also shut him down when he said, "Ted, I was born here," after Cruz finished up his monologue about eligibility. Unfortunately, the Lincoln-Douglas standard is elusive today.
My guess is he prepares a bit more for the next one. One on one is definitely a different format than 16 candidates vying to be heard.
Hypothesis #7 No matter what the truth may be, the media will shill relentlessly for Hillary Clinton and be appalled when she loses.
There are two more debates to go. If Trump took Hillary down in this debate, she would know all the lines for the next two. Hillarybots would be saying that Trump just keeps repeating the same things. The top fifteen polls showed Trump winning the debate right afterward. You have let DNCbots convince you that the sun never rose. The proper response to them is "Pickles is finished." The next time the meds won't work the same way.
"My guess is he prepares a bit more for the next one. One on one is definitely a different format than 16 candidates vying to be heard."
The other problem was his pre- and post-debate spin, which was all over the place. First, he tries saying "Lester Holt is a Democrat" which happens to be not true--Holt has been a registered Republican since 2003, longer than Trump in fact. Then after the debate he says Holt did a great job, followed quickly by his team (including Guiliani who is coming across more unhinged lately) complaining about how biased Holt was (which may be true, but weren't we told Trump would call him out during the debate the way Romney didn't with Crowley? After all, complaining after the debate didn't help Romney). Then he goes around saying he wasn't sniffing, as if no one actually heard him sniffing, and why keep that conversation going? And to top it off, he spends Tuesday morning saying yes, Miss Universe was a fat pig, because there's nothing swing voters want more than to hear him going on with an attack on someone who has nothing to do with this.
On the plus side, he can't do much worse next time.
If Trump wins as predicted by Scott Adams, you will be able to look back at the crowd and overflow size of his rallies, and realize it was all very apparent from the beginning of his campaign that he would win.
Today's crowd size in Waukesha will be interesting.
Others were saying things. Trump was responding to questions. He should have just given a stock response--"looking forward to our next debate." He is learning a lot about the sickening world of politics, but he is learning. If the DNC dicks around with his mic next time, I assure you he will rip it out and continue like a game show host.
The top fifteen polls showed Trump winning the debate right afterward.
Well, no. A number of internet polls showed those results. Traditional polling showed a different result; that Clinton was viewed by majorities as the winner.
http://fortune.com/2016/09/27/presidential-debate-polls-winner/
Many would suggest that internet polling is regarded as less reliable. See, the Fortune article just above.
Well, yes. I am not going to dig up the list, but interestingly, many of those fifteen were from left-of-center media.
Yeah, the ones supporting open borders and amnesty.
Total Bull. and what good is Trump on those issues? please.
eh - I'm back to thinking this is all just a big scam.
Darrell said...
Well, yes. I am not going to dig up the list, but interestingly, many of those fifteen were from left-of-center media.
Okay. Don't dig up any lists. I'll let people read the Fortune article, follow those links, and make up their own minds.
Those l-o-c people had probably never watched Trump before. They were expecting him to sound like a stormfront rally and they didn't know what was happening.
Without digging back through the primary polls, I remember a few polls where Trump significantly outperformed the polls (as grimson linked) a few where they were fairly accurate, and one or two where he slightly underperformed.
Are you going to comment later as "P,B,&J"? I am interested in the psychopathy of doing things like that.
Brando wrote
They do weight, but weight differently--they'd be different populations (e.g., Ohio voters likely to vote in a GOP primary vs. Ohio voters likely to vote in a general election). Perhaps that difference in methodology could be missing something.
The one poll I looked at, Emerson, specifically wrote at the bottom that they didn't weight the results. What evidence do you have that they are lying?
This is not the list of fifteen snap polls that I referred to, but--
http://www.dailymail.co.uk/news/article-3809204/Most-snap-polls-Trump-winning-debate-landslide.html
I don't think, though, as you suggest, that the pollsters are intentionally trying to make this look like a closer race when they see a GOP landslide. First, a lot of these polls are conducted by conservative-affiliated groups. Second, getting an election wrong by a lot (and doing this consistently) should have happened more often if that's the case--and despite similar theories, it didn't happen in '08 or '12. Plus, getting it that wrong and developing a reputation as a poor pollster seems a high price to pay for the dubious advantage of convincing your favored candidate's supporters to turn out. Which brings me to my third point--if the real number is a landslide for Trump, and the pollsters intentionally skew it to look like a nailbiter, isn't that just as likely to increase turnout on both sides? (And before you point out that Trump's voters are so much more motivated than Clinton's and would turn out even if he was up by ten points, consider that most of Hillary's support, and Stein and Johnson's support, is more anti-Trump than anything else, and a sign that supporting third parties or staying home would help Trump would be enough to get them to turn out and switch to Hillary).
I don't care what their motivation is. Is there, or has there ever been, a price to pay as a pollster for getting it wrong?
My point is, why don't they give us the raw data? What's all this BS that we only get to see the weighted data? Is it because we are too stupid to know the difference? That they are afraid if the raw data says Trump +5 and the weighted data says Clinton +5, we will all think they're telling us Trump will win?
Give me a good reason why they hide their raw data. Why can they give us unweighted results during the primary, but not during the General?
Unskewing the polls has been mocked since 2012. Isn't that what the pollsters are doing? Skewing the polls by weighting them? It's common practice. So give me the raw data and then convince me your weights make more sense than the raw data.
Blogger Darrell said...
Are you going to comment later as "P,B,&J"? I am interested in the psychopathy of doing things like that.
Are you saying pb&j is a sock puppet?
Cruz already wrote his "sensible" "compromise" amnesty plan, because "we've reached the point where we must do something about the problem and reasonable people compromise." He'd fold when people called him a racist for not taking in "refugees." "People, people, we must do the right thing!" Fuck him. Loser.
A while back P, B. and J said he was Chuck and others even repeated the question. "Chuck" did not disagree. That made me "assume" Chuck is P,B,&J.
The best that can be said about survey research---political opinion polling is one part of that--- is that it is a quasi-science. It is more likely a pseudo-science. The reliability of the mathematics have improved tremendously but it still rests on the clay feet of the assumptions about the composition of the electorate. If the assumptions are b.s., then the math is misleading and thus useless.
"Total Bull. and what good is Trump on those issues? please."
Exactly--not only were several candidates anti-amnesty all the way through, Trump himself is now for limited amnesty. Until he changes his mind next week.
If it was all about the "issues" the Trump fans would have gone with someone consistent on them.
"The one poll I looked at, Emerson, specifically wrote at the bottom that they didn't weight the results. What evidence do you have that they are lying?"
I can't speak to a specific poll, but it's normal methodology for polling firms in every type of polling to weigh and use estimates based on the limitations of their sample. I'm not sure what method Emerson uses.
"My point is, why don't they give us the raw data? What's all this BS that we only get to see the weighted data?"
That I can't answer for you--I don't see any harm in sharing raw data and then showing your weighted results (unless doing so would give away proprietary methodologies they don't want other firms to steal? But I don't see how raw data would do that).
As for a "price" to pay for getting it wrong, I suppose if say Reuters consistently had a result overestimating the Dems by say 10 points, and their competitors (or even one competitor) got it right all the time, how long before you think people would discount Reuters results? This hasn't really happened to any one firm over a consistent basis. Rasmussen overestimated Romney by about five points (not a huge margin, but noticeable) in 2012, but if this became a longterm trend the polling aggregators would either not include their results or take that error into account.
A lot of these companies, remember, do more than just political work--their surveys have to estimate consumer opinions for commercial clients. If they get a reputation for being wrong enough, they'd lose a lot more than just credibility in the media.
Blogger AprilApple said...
Yeah, the ones supporting open borders and amnesty.
Total Bull. and what good is Trump on those issues? please.
Interesting psychological dynamic going on. Some people who say they are "Republican/conservative" seem to be pulling for Trump to lose. I suppose this will give them some satisfaction to have been right all along.
As I type this, I can hear Bernie out campaigning for Hillary on TV. I think I smell fear but nothing counts until the election. He is pushing free tuition, as if there were any chance it would happen.
By the way, I don't expect Trump to prevent the coming crash. It will probably be next year and I suspect will be bad.
I thought in 2012 that Romney was the last chance for a soft landing. I still think so.
I pretty much agree with Angelo Codevilla.
Still, you have to try to stop the runaway train.
Darrell said...
A while back P, B. and J said he was Chuck and others even repeated the question. "Chuck" did not disagree. That made me "assume" Chuck is P,B,&J.
If I didn't, I can assure you that the only reason I wouldn't respond is because I think that you, and comments like this, are such pathetic wastes of time.
Michael, if by crash you mean stock market crash then I agree. Although it won't be a crash so much as a correction, like what we had in 2007-2008.
Something I've noticed in my years of trading.
The Clinton correction came just after Bush got into office and was exacerbated by 9/11. The next correction came right before Obama got into office and was much quicker and more violent.
If Clinton wins, I think the correction can be held off for another four years. But if Trump wins, it'll be a correction for sure.
"Some people who say they are "Republican/conservative" seem to be pulling for Trump to lose. I suppose this will give them some satisfaction to have been right all along."
Well, much as we'd like for Trump to lose, we don't want Hillary to win either, and I can assure you there is no outcome of this election that will be satisfying for small government conservatives. We lost, we know it, now we're just hanging around for entertainment. Whether we vote Trump, third party, or write-in, doesn't matter. The decline continues.
I agree about Romney. I think we missed our chance to right the ship in 2012. And now its' all sinking fast.
I'm not sure what it means but I just got spammed with an Email from the Trump campaign claiming that they've raised a record $18M during the 24 hours following the debate. Can't be a bad sign for his campaign.
...said Chuck, the Hillary Clinton supporter and all-around Democrat shill.
If Clinton wins, I think the correction can be held off for another four years. But if Trump wins, it'll be a correction for sure.
I'm not sure I agree. Clinton might keep the balloon up a bit longer. I expect, for example, that California real estate is toppy and will probably go through a huge crash if the market drops too much. If that happens, California is in deep trouble. The pension bear is kept away only by the very top-heavy tax situation here.
I see a cascade like 1929 coming. I don't think Trump will stop it as it is too late. He might clean out the corruot bureaucracy some.
Just to clarify, Chuck, AprilApple and PB&J are the same person right? Do we have an estimate of how much they/he/she get paid by the Hillary campaign to shill?
I'm guessing it's tops $1000, but I'm open to views.
"Real Republican Chuck."
The polls aren't missing anything. Hill is up. More people plan to vote for her. She will win most swing states and therefore the presidency.
Chuck said... [hush][hide comment]
The top fifteen polls showed Trump winning the debate right afterward.
Well, no. A number of internet polls showed those results. Traditional polling showed a different result; that Clinton was viewed by majorities as the winner.
http://fortune.com/2016/09/27/presidential-debate-polls-winner/
Many would suggest that internet polling is regarded as less reliable. See, the Fortune article just above.
Yeah, I discarded internet polls. Assume they are skewed by multiple votes per person or some sort of tricky stuff geeks know how to do. Lately, though, I have been wondering why they were not "cheated" in favor of Clinton because of all the support among college educated who are more likely to be able to set up some sort of vote flooding systems.
In any event, Trump better up his debate game, for sure. Hillary looked way better than anticipated. Even I am beginning to wonder if she is not actually healthy and was suffering from exhaustion recently.
Micheal K - I'm not pulling for Trump to lose. Trump is.
1929 happened 8 years ago. This next downturn would be 1938, if the pattern is the same and it holds. The market does seem to be moving based on how the polls change. If we anticipate/get a trade war and a continuing global trend towards nationalism, it will be hard for the market to keep trending upward. We can argue the merits of free trade all we want, but it is good for markets.
The San Francisco real estate market in particular looks unsustainable. I can't anticipate how or when it breaks down, but when public and private employers can't attract employees because of prohibitive costs of living there, something has to give.
Chuck: "I've come up with these links that show Hillary absolutely crushed the debates despite what some partisan Republican types might tell you. And I looked up some independent methodology data that show the Republicans who think otherwise are nut jobs."
"But for avoidance of doubt I'm a lifelong Republican and I'd never vote for Hillary - I'm just sharing facts about what happens when Her Highness debates the Republican Party candidate who by the way is the reincarnation of Hitler. Er, I mean Hillary, not Her Highness."
"But I'd never vote for Hillary. And Donald Trump totally lost the debate, which you'd understand if you click the links I sent about how badly he did and how well Hillary did. Though of course as a lifelong Republican I'll never vote for her. Err, I mean him. No, her. Wait, what was the question?"
Seriously Chuck, lifelong Republican, do you think anyone here is buying your story? Dude, just slink away, and tell the campaign manager that unfortunately you've become a joke on the site you were assigned to.
I'm not going to vote, I'm not a Republican, but I'm still offended by how bad at this you are.
Real Republican Chuck, I am not a Republican. But you are so bad at pretending to be a Republican that you offend me. Please stop.
AprilApple and PB&J, could you get together with Real Republican Chuck and decide which identity you're going to post under? Posting under all three is annoying. Please stop. Thanks.
Chuck said...
grimson where are you getting your numbers?
Just asking if you see any states where Trump failed to outperform the state polling. (That is RCP averages for each state, right?)
The numbers were from the RCP site, looking up individual contests. I remembered Indiana was significantly better, and was curious about Pennsylvania and Wisconsin.
Can't you just say outright that you don't comment here under any other alias? What an asshole!
"... there is no outcome of this election that will be satisfying for small government conservatives."
And some day even the brain-dead-holy-shit-how-can-you-even-type types will figure out there never was any outcome of this election with any of the primary candidates that would have been, after all the bullshit had been spewed that is, satisfying.
This could be a lesson for you, a moment of clarity.
Take it.
You believe "small government" lies by people who increased the debt in direct literal partnership almost as much as the Democrats have, to the tune of $20 Trillion, and that, even THAT, won't get you to open your damned fool eyes.
Pawn.
What will you say when the Democrats blame Boehner and McConnell for the debt increase(s) on Obama's watch? You don't have a clue do you. You are going to wait until some dumbass like Hugh Hewitt tells you to say some bullshit about Continuing Resolutions and how complex things are in D.C.
Blogger Big Mike said...
I'm not sure what it means but I just got spammed with an Email from the Trump campaign claiming that they've raised a record $18M during the 24 hours following the debate. Can't be a bad sign for his campaign.
9/28/16, 3:07 PM
I just signed up for monthly donations.
In Lebowski lingo, we (Trump Trash) threw out a ringer for a ringer.
It's all circus clowns all the way down, and now you know.
We made it simple for you.
Vote Donald and thank us.
Left Bank of the Charles said...
Hypothesis #6: The polls are underestimating John Kasich's ability to torpedo Trump in Ohio.
9/28/16, 1:00 PM
Pray tell, how would this be so?
John Constantius,
Punish them. Vote for Trump!
Blogger Darrell said...
Can't you just say outright that you don't comment here under any other alias? What an asshole!
9/28/16, 4:04 PM
What, you want him to lie?
I believe the pollsters should be using turnout models closer to the last 2 mid-term elections where Obama wasn't on the ballet, D enthusiasm was low and the country was trying to send a powerful stop-the-nonsense message to Washington. This angst has only intensified since 2014 making it likely that the data are skewed even more anti-Washington. Are today's polls accounting for the 2010/2014 drubbing?
http://nypost.com/2016/09/26/the-best-debate-takes-come-from-inside-the-bar/
DimeStoreDave I think you make a very good point, but historically presidential turnout has been much higher. Using your view if the turnout rises proportionally then Trump is a shoo-in. The accepted wisdom is that Dems turn out better for Presidential years. Though I haven't seen much on the Senate/House races what I have seen lately has said that the R's are going to retain majorities in both. I have occasionally wondered that if there is confidence in a strong R turnout for down ballot offices why won't that help Trump? Your point I think.
I just checked and RCP has the Senate even. So I am wrong there.
I believe the pollsters should be using turnout models closer to the last 2 mid-term elections where Obama wasn't on the ballet, D enthusiasm was low and the country was trying to send a powerful stop-the-nonsense message to Washington. This angst has only intensified since 2014 making it likely that the data are skewed even more anti-Washington. Are today's polls accounting for the 2010/2014 drubbing?
This is a very credible notion, stated well.
This general election should have been such a walk, for any Republican without Trump's historically bad negatives. We'd all be talking about Hillary's server and Obama's assumed-name emails to her, not to mention Hillary's inner circle all invoking the Fifth.
In primaries the shame of Trump voters was being for NYC values and for a divorced man who talks like a vulgar NYC person. But after Mr Christianity was beaten. Like a rented mule, the shame of Trump for being a Yankee Atheist Sinner left with Cruz's lies.
Since then Trump has pivoted to being an anti-Globalist and Nationalist. And NYC Yankees can do that easily.
The only hint of racist in Trump was his birther stand and his Mexican border wall. Actually those are Conspiracy theory positions and not racism except as flamed to be seen by LaRazans and by Marxists as WASP tricks. They only knee jerk say that nonsense to Alinski Trump and destroy the USA.
The new Reuters poll used the percentage of Dems lessened Republicans and independent from their prior poll.
Darrell said...
Can't you just say outright that you don't comment here under any other alias? What an asshole!
Sure I can! Will it do any good? Let's see.
I don't comment here under any other "alias" and not under any other name whatsoever. I've never had a Blogger/BlogSpot identity other than this one that I use here. I've never commented on Althouse, ever, under any other name. Professor Althouse knows my name, and my regular email address, and she's had it for years because I happily shared it with her long ago and whenever I've shared ideas with her.
Now, do you want my paper birth certificate? You're a bit like a birther, demanding this silly and frankly irrelevant information. You've proven nothing; you've disproven nothing that I have written by making this demand but I'm actually happy to respond, just to call attention to the silliness of the original charge.
I don't understand why this happens so often on social media; Aha! He has another user name and won't admit to it! I don't get it. Does that matter? Outside of 7th grade, that is.
Anyway, if there is anything else that I can do or write to more conclusively defeat this pointless supposition and (let's not forget) use it to demean and embarrass you, please let me know because I'd like this one to leave a mark.
The problem is, Chuck, because you're such a snake, it's hard to get you to give a straight answer to anything, let alone answers that tend to work against you. I suppose it is lawyerly, which is probably why people want to take lawyers end... Do things to them. You seem to take to playing the role of a manly man and would therefore be expected to know how to talk straight, but I'm assuming facts not in evidence.
"You believe "small government" lies by people who increased the debt in direct literal partnership almost as much as the Democrats have, to the tune of $20 Trillion, and that, even THAT, won't get you to open your damned fool eyes."
Er, no, but nice assumptions there. However, some of us would prefer a candidate who is even somewhat more small government to two candidates who are competing for who can bust the budget fastest with their sweeping government fantasies.
And what's your reasoning anyway--that because "establishment" Republicans have been so lousy at advancing small government values, we may as well go for a candidate who will expand government far more than ever before? Is this the "burn it all down and roast marshmallows over the embers" logic?
Bad Lieutenant said...
The problem is, Chuck, because you're such a snake, it's hard to get you to give a straight answer to anything, let alone answers that tend to work against you. I suppose it is lawyerly, which is probably why people want to take lawyers end... Do things to them. You seem to take to playing the role of a manly man and would therefore be expected to know how to talk straight, but I'm assuming facts not in evidence.
So what straight question(s) do you have for me?
I am not going to play nice, or even pretend to be civil with, people who robotically respond to my posts with claims that I am a Hillary Clinton supporter. Simply to discredit my criticism of Trump.
Nor will I just take it, from people who want to demean and attack the Republican Party.
You want a fight I will give you a fight.
Ignoring the Chuck ghafla, one thing the pollsters ignore is the ability of Democrats to create enough votes after polls close to swing the election to the Dem candidate.
How well has Hillary organized her GOTV campaign that begins the moment polls close, which occurs while the actual votes are counted, and adds votes for her to the actual vote count?
Proof my supposition is not groundless: Senator Franken.
टिप्पणी पोस्ट करा