macman76
12.06.16 | Mostly, I'm just bored at work |
Rolling Girl
12.06.16 | Polls can certainly be manipulated depending on where and who you ask but I would never call them broken. I also feel that getting a truly unbiased sampling in a place like the U.S can be very difficult |
Spacesh1p
12.06.16 | I've read the same, mac. Polls seem to have taken a lot of hits from the collective media shock regardless of how (un)warranted it is. |
macman76
12.06.16 | I wonder if turnout polling becomes a much more spotlighted part of prediction |
Cygnatti
12.06.16 | "as far as I could tell no one really took the raw demographic data, and included that in their models, rather they trusted pollsters likely voter number)"
yeah i was thinking this. but even so, i think that the numbers from the polls incurred a level of complacency in the clinton campaign and among democrats that let them feel as though they didn't need to try as hard in certain places - like pennsylvania. |
macman76
12.06.16 | I think the retrospective takes that bubbles/fake news/how bad of a a candidate Clinton was are also pretty lazy, there aren't enough people saying that very few people flipped and it may have dozens of causes specific to the region and information diet of those voters. I also think the popular vote push is dumb, too. |
zakalwe
12.06.16 | All manipulated by the media agenda of procuring further funding from the government and suppressing the rage of the voiceless. The voiceless majority lashed out. Thank fuck for democracy. |
Cygnatti
12.06.16 | i think a very important question should be "why did so many democrats/liberals vote for trump?" estimated that 9% of registered democrats voted for him.
"popular vote push is dumb" [2]. recount thing was also dumb, HOWEVER, since trump asserted that millions of illegal votes were cast (in hillary's favor), in his own words - there should definitely be a recount!! |
macman76
12.06.16 | @cyg do tv ads or campaign visits actually change voters, tho? I heard stories that clintons ground team was driving trump voters to the polls by accident, but didn't that also probably happen with Obama against Romney and McCain |
macman76
12.06.16 | Romney won 7% of dems according to roper center |
Spacesh1p
12.06.16 | @mac Prevailing belief in political science is that TV advertising and campaign visits do turn voter opinion. |
Cygnatti
12.06.16 | i think so. florida is always contested and trump managed to win it. pennsylvania should have been a shoe-in for hillary/democrats, so she neglected it, only to have them turn around instead. i do think that she could have won them over, had she campaigned there, but i couldn't tell you resolutely.
tv ads are much more important in the primaries, i think, and less so in the general election. if anything, they just remind people to vote, not that they really changed anyone's opinion (imo, that is). |
guitarded_chuck
12.06.16 | american electoral system is hilarious |
Spacesh1p
12.06.16 | Personally, I think the Democrats just had a losing strategy from the off. Clinton never presented a vision for her presidency aside from continuing the Obama legacy, which was a grave error in an electorate so hungry for change. Furthermore, the way they approached social issues like race and sexuality was another error. They ostracized moderates and your average minority voter in a place like Detroit doesn't get motivated to vote against Trump due to the perception that he is racist. He would have to campaign on bringing back Jim Crow laws for average minority voters to be driven to the polls in significant numbers against Trump. |
macman76
12.06.16 | @space, they wouldn't pay for them if they did nothing, but I'm not sure she could have flipped the three Midwest states she needed if she was performing an 8 hour jam rock set list every day there and if she bought a local station for nonstop advertising lol |
Spacesh1p
12.06.16 | Yeah Mac I don't disagree. You said yourself that you aren't enthused with the "poor candidate" argument, but I think Clinton was exactly the wrong candidate for 2016. |
macman76
12.06.16 | @space she very well could have won though, she lost just enough of the undecided/high leverage voters to lose, but the world in which she wins isn't far off from the one where she lost and in that world we don't call her strategy terrible and question the legitimacy of her candidacy |
Spacesh1p
12.06.16 | Also I think the outrage about the "glass ceiling" for Clinton is also so disingenuous. I completely agree that it is harder for a woman to get elected in America due to how we perceive who/what is "presidential," but people who say that this proves that a more qualified woman will lose to a less qualified man are so off-base. It's an election, not a job interview, and the perfect paper resume doesn't count for much in the eyes of voters. |
macman76
12.06.16 | I'm not sure dems need to stop identity politics either, that also seems lazy to me |
Spacesh1p
12.06.16 | Yeah Mac but I think the numbers generally (if I'm mistaken, please correct me) that Clinton's problem was getting people to turn out and vote for her - not as much that the enthusiasm for Trump was so high across the electorate that she just couldn't ever come up with enough voters to win. The whole "enthusiasm gap" idea, basically. |
Spacesh1p
12.06.16 | Eh I would argue that they really need to shift away from identity politics myself. I think it makes their platform far too narrow and pushes away many centrist voters. It would have been very interesting to see Clinton versus a "standard Republican" as well.
It's a complicated question, I'd be interested to hear why you think that the current Dem strategy is a good one. |
macman76
12.06.16 | That's possible space, since there was lower turnout for minorities and little or no gain in winning them versus Obama, and uneducated whites moved much greater for trump. But I want to stres that I think a large share of clintons loss is because of randomness and that she couldn't have done very much even in hindsight to flip the election her way. She rolled snake eyes.
It's still her fault that she wasn't more popular, but the takes would be different if she won |
Spacesh1p
12.06.16 | I don't think you're wrong about randomness, although I have no way of proving it. |
macman76
12.06.16 | Even if dem candidates don't play identity politics, dem activists do. Should black women stop talking about the issues that affect them? Or intersex people? Or young black men? |
Spacesh1p
12.06.16 | "the takes would be different if she won "
You're absolutely right about that, although I think that the predominant narrative would be confirmation about people's thoughts regarding how unelectable Trump appeared, rather than talking about how successful Clinton's strategy was. |
macman76
12.06.16 | You know that CNN/MSNBC/Huff Po and whoever would have dem strategists talking up a storm about Clinton tho had she won
I also request a national course on probability and statistical modeling |
Spacesh1p
12.06.16 | "Should black women stop talking about the issues that affect them? Or intersex people? Or young black men?"
Definitely not. I have no issue with people addressing issues that impact them, but I think that they shouldn't be the only thing that is discussed... they could do more to reach out to people on a basis other than skin color or sexuality, I mean. Being born into poverty and thus having a massive opportunity deficit isn't determined by race. Taking up a more class based platform like Sanders advocated makes much more sense to me, although I did not support Sanders either for full disclosure. |
Spacesh1p
12.06.16 | It's true that the left media would be talking up Clinton, but I still think the narrative would be about Trump's defeat more than Clinton's victory, if that makes sense. |
Cygnatti
12.06.16 | "Even if dem candidates don't play identity politics, dem activists do. Should black women stop talking about the issues that affect them? Or intersex people? Or young black men?"
no ofc not, but i think u can do it in a way that doesn't assert that people who don't agree with you are "deplorable". and while hillary was the one who said it, many of her supports actually believe it, especially post-election. |
macman76
12.06.16 | No one calls it identity politics but the fight for 15 is something dems fight for, they proposed free or cheap community college. I'm not sure tho if that's why people voted for or against Clinton necessarily tho |
zakalwe
12.06.16 | Testify space dude. Love ya pal. |
Spacesh1p
12.06.16 | Good point, mac, but the fight for $15 and cheap community college is not really directed at your average middle class voter. That's where Dems lose the most.
Separate issue but I think turnout hurts Dems quite a bit. If election were a weekend or a holiday, I think you'd see a lot more Dem voters, but again, I can't back that up with data. |
Spacesh1p
12.06.16 | @zak I still wouldn't have voted for Brexit though dude! |
macman76
12.06.16 | If you have any sort of political views, there will always be someone deplorable to you, some pro lifers see abortion as murder, some see income tax as theft |
Spacesh1p
12.06.16 | "If you have any sort of political views, there will always be someone deplorable to you, some pro lifers see abortion as murder, some see income tax as theft"
I find that to be a cheap argument though. It's no excuse to break the system by making everything a sacred cow. |
TVC15
12.06.16 | Mac has a job? |
macman76
12.06.16 | If it sounds like I'm being contrarian, it's because everybody seemingly has a solution and explanation that they seem to think explains everything, I wish we had more candidates and even voters who said "I don't know" when they faced a complex issue |
wham49
12.06.16 | I think they werent that far off, but trusting polls, when people intentionally mislead the polls, is a bad thing, I do it
I think people were just thinking there was no way, no matter how close the polls got, that he could really win regardless |
Spacesh1p
12.06.16 | That's fair mac and I will be the first to say that I don't know it all. I've just presented my opinion and some speculation here in this thread. What I would like to hear more of is simply balance. Boring rhetoric of compromise and reaching across the aisle doesn't draw the viewers but politics should be pretty unexciting when it's done correctly. |
macman76
12.06.16 | This election probably won't make people more moderate, and I'm not sure they should. It's gonna be hard to tell a BLM activist that they need to cool it for the sake of electoral politics, for instance |
Spacesh1p
12.06.16 | Indeed. It's a tough question. Do you double down and try to generate more momentum by being louder or recalibrate and try to appeal to the larger but less ideologically invigorated moderates? I would like to see the latter but that doesn't seem to be the strategy most are taking in today's world. |
ShadowRemains
12.06.16 | meanwhile the kids aren't alright |
Spacesh1p
12.06.16 | I think there's just a lack of perspective today on many sides. There's probably never been a better time or place in human history to be a minority than in a place like the states in 2016. It's far from perfect, but the fact that the narrative largely consists of microaggressions and unconscious biases when discussing minority issues is indicative of how far we have come.
Things like police violence are more pressing issues and deserve a heightened level of rhetorical intensity, but like most things, there's so many issues that fold into police violence that aren't strictly race related. |
Cygnatti
12.06.16 | "If you have any sort of political views, there will always be someone deplorable to you, some pro lifers see abortion as murder, some see income tax as theft"
completely agreed and ofc, but apparently when she said 'deplorable' she actually meant that you were one of the following: "racist/sexist/homophobic/islamaphobic/etc". she didn't, in fact, just mean "deplorable" :x |
Spacesh1p
12.06.16 | I've basically just taken over your thread to vent, mac, apologies dude. |
macman76
12.06.16 | @space I made the thread to vent, but I'm venting that people are too confident and simple minded specifically about trumps chances |
zakalwe
12.06.16 | The big issues are being avoided.
Why are CD albums wrapped in that cellophane which is impossible to take off?
When putting petrol in a car why can't you just input how much you want so it automatically cuts off so you're not fart arsing around and end up paying £20.02p
Why has sowing got such shit taste?
|
Spacesh1p
12.06.16 | I hear you on that. It's been very popular this election cycle to talk about how Trump has a "mandate," whatever that exactly means, but as you said, a lot of things fell for Trump or against Clinton that could be down to randomness. |
Spacesh1p
12.06.16 | Zak, in the states, you can pay cash and get an exact amount. |
macman76
12.06.16 | Yeah, y'all came after my boy Silver and trashed numbers so... thread |
unclereich
12.06.16 | The polls never seemed off to me. Both candidates admitted they were not looking to win the popular vote hence their strong campaigning in the swing electorate states. If dems knew anything about the electoral college they would not have been surprised. She always held a lead from a popular vote standpoint the polls showed this but neither candidate cared thats why both especially hillary was in Ohio and Pennsylvania for weeks right before the election. She knew she was in trouble. |
zakalwe
12.06.16 | I'm heading over space, settling down the lot.
Problem is I look a bit like a muzzie so I'll get fucked off out of it at the application stage. |
unclereich
12.06.16 | Voter turn out is the more interesting topic of discussion. Did the media paint hillarys chances in such a way that people assumed shed win so they didnt feel the need to vote for her? This will be a question analysts will fumble around with for the next 4-8 years |
macman76
12.06.16 | @fourth they'll never answer that, fourth. Her poll numbers only moved significantly after comey. Early returns suggest that whites were undersampled. |
Spacesh1p
12.06.16 | I don't think that media portrayal of a probable Clinton victory dampened voter turnout to a significant degree in states that were in play. Election day just needs to be on a weekend or a holiday so that people can actually show up without missing work or getting up really early/staying out late to vote. I think that would make a big difference but I don't have any evidence to back that up. Just my opinion. |
Spacesh1p
12.06.16 | @mac, you mean after the Comey announcement close to election day? I figured that that would have a pretty minimal impact since most people appeared pretty set on their opinions regarding the email controversy prior to those comments, but it sounds like I was wrong. |
unclereich
12.06.16 | Clinton never presented a vision for her presidency aside from continuing the Obama legacy, which was a grave error in an electorate so hungry for change
This this this. This is why candidates have to be weary of taking endorsements so seriously. It is well known obama and clinton nevet got along. So why take his endorsement? She had a huge opportunity to state why his 8 years were mostly a failure and how shed reignite the dem party. But she assumed the endorsement would garner more votes when really people didnt care and are sick of barack obama. What could she have done talked down about his polices right in front of him lol? What a dumb move |
macman76
12.06.16 | 538 dipped a lot after the coverage (2.5 to 3 points in the national polls). It seems natural to suggest that it was all comey, but it's also possible she would have dipped even without the news |
macman76
12.06.16 | @fourth, I'm working on the premise that she could have won with a little more luck, would you call her strategy dumb if she won? |
Spacesh1p
12.06.16 | Interesting dude. Comey was in a tough situation, though. Some of it was his own doing, mind, but I didn't envy his position in all of this. Transparency was probably the right call but I honestly don't even know. Talk about a hard question.
@Fourth, I just think Clinton could have done more to come up with a cohesive policy narrative. She had a ton of positions on her website, but her bottom line call to action was continuing the Obama legacy. That just doesn't sell like "make America great again," no matter how empty the statement may be. |
unclereich
12.06.16 | Absolutely, you dont think so? |
Spacesh1p
12.06.16 | Hindsight is 20/20 of course but I questioned her strategy throughout the election so I maintain that I would have done so had she won too. Maybe if she had blown him out I would have changed my mind, but in the end, if Trump lost, I would have been far more convinced of a case of electoral suicide than homicide. |
macman76
12.06.16 | I think of Clinton like if she's Andy Dalton, you can win with him, but don't ask him to take over a game. I don't think Clinton could have done much to get people to change their minds about her. |
unclereich
12.06.16 | Idk myself and a lot of my fellow poli sci bros in the masters program her were dumbfounded long before the election was over with her strategy. The last time we had consecutive democrat presidents after a full term was the 1850's i believe. Thus the thing she needed most to do was stray away from the obama admin NOT ride his coattails |
Spacesh1p
12.06.16 | "I don't think Clinton could have done much to get people to change their minds about her."
This is something I completely agree with. Those who didn't like her would never come around and those that did were far more anti Trump than pro Clinton. |
unclereich
12.06.16 | Agreed with both i love the dalton analogy |
Spacesh1p
12.06.16 | Most enjoyable political thread on Sput, thanks mac. |
macman76
12.06.16 | @space thanks. I'll fight you if you talk shit to numbers tho lol |
Trebor.
12.06.16 | macman is a fivethirtyeight junkie SHOCKER |
macman76
12.06.16 | Fight me Robert |
Spacesh1p
12.06.16 | I would pay to watch you guys fight. |
Trebor.
12.06.16 | dude I love fivethirtyeight dude |
macman76
12.06.16 | blood still needs to be shed lol |
macman76
12.06.16 | I also want to say that the people suggesting electoral voters flip are delusional |
Spacesh1p
12.06.16 | Agreed. That would be catastrophically bad. |
anarchistfish
12.06.16 | The same things happened in the UK 2015 general election and in the referendum this year. Polls obviously have limitations and sometimes they don't capture a certain demographic or people aren't honest with their voting intentions and there's not much you can do about that. Betting odds are garbage and should never be used to predict likelihoods in matters outside sport. |
macman76
12.06.16 | Betting odds are garbage and should never be used to predict likelihoods in matters outside sport.
Why tho? They suggested a 20% chance for brexit |
macman76
12.07.16 | Agreed, Nate silver doesn't make calls, he assigns probabilities based on statistical models. He's not a seer like the press has hoped or played him up to be. |
macman76
12.07.16 | I got to 1500 comments itt |
GhostOfSarcasticBtrd
12.07.16 | Correct me if I'm wrong, but he had like a 1/3 chance of a Trump victory, did he not? That's not a slim probability by any means so this wasn't terribly unusual. |
macman76
12.07.16 | Yeah but people see 30% and think 0%. Also models and betting markets are assumption based. Poll aggregators like huffpollster had it at 2% for trump, there assumptions were clearly wrong and people that didn't like the less certain models would point to those to feel better. |
anarchistfish
12.07.16 | "Why tho? They suggested a 20% chance for brexit"
they only represent how likely people think something is to happen which in many cases can be extremely unreliable and will never pick up on factors that haven't gained wide public attention. Leicester City winning the premier league last season was put at 5000/1 odds. This year the equivalent was much higher (I think like 700/1), because before no one conceived that a team like Leicester could actually win it, but I doubt their odds were actually 5k/1. In reality you can't really represent the likelihood of something like that in numbers cos it's so removed from material reality and precision is out of the window |
macman76
12.07.16 | But the alternative is having no information. Polling also suggested around 20% chance for brexit. As for Leicester, all models are wrong but some are useful. I imagine the 5000 to 1 was pretty consistent with the top heavy nature of the PL. |
anarchistfish
12.07.16 | I don't like it because its a lazy substitution for proper analysis (which often does conclude that "we don't know"; but it's better to know that you don't know and why than to pretend you do) that gives people the idea that they know what to expect and which will always gloss over important details that often provide the crucial deciding factor. Betting odds will always favour conditions of inertia based on received wisdom. Received wisdom is often a bad thing to base your opinions off and if it's your go-to tool in every discussion you won't ever make discerning points when it matters |
macman76
12.07.16 | The housing market crashed because of improper assumptions, but models and odds are a starting point. I've heard someone suggest that models can sometimes be an excuse not to think critically about something. But you don't make money in betting if you accept the odds as completely accurate. |
anarchistfish
12.07.16 | They ultimately don't mean anything more than what people expect to happen, which is a bad way of analysing a situation, and in political and social scenarios can be pretty dangerous |
macman76
12.07.16 | People that bet on Leicester probably were fans or thought the bet had good value. Most other information suggested brexit was unlikely. Improbable things happen all the time though. |
Spacesh1p
12.07.16 | Betting odds aren't useless, in my opinion. Just like all other measures you have to look at them with the proper considerations in mind. As a simple indicator of "this is what the betting public perceives as the odds of x occurring," betting odds can be informative. That doesn't mean you have to carry that conclusion to "this is the likelihood of x occurring." |
anarchistfish
12.07.16 | "That doesn't mean you have to carry that conclusion to "this is the likelihood of x occurring.""
Many people do that though |
macman76
12.07.16 | It's not your fault people are dumb, fish |
Spacesh1p
12.07.16 | "It's not your fault people are dumb"
[2]
Personally it's interesting to see what outcomes people are willing to stake money on. It's not always (read: often) not based on rationality but it is an interesting and quite readily available indicator of sentiment. |
macman76
12.07.16 | Cable news would be more informative/entertaining if pundits and hosts had to put serious money on their analysis and predictions |
macman76
12.08.16 | This is pretty relevant to the discussion itt
http://andrewgelman.com/2016/12/08/19-things-learned-2016-election/ |
macman76
05.30.18 | My dudes, Nate silver dropped some data.
https://fivethirtyeight.com/features/the-polls-are-all-right/ |
guitarded_chuck
05.30.18 | 1) polls are never accurate because the sample size is only of people willing to take the poll
2) the polls were incorrect because they took a snapshot of time just before the election when hilary's popularity was plummeting due to the email thing and indicated the election was going to be closer than had been forecasted previously |
macman76
05.30.18 | 1) you’re asking for too much certainty for polls. Non-response bias is real, but so far not super damaging to poll accuracy.
2) if you read the article the polls were just as close at predicting races as they usually are. It just so happened to miss high leverage states on directions that flipped the election. It also wasn’t destiny that the email thing would come up or that it would stick going into Election Day.
Lesson: be less certain but more informed with data |
guitarded_chuck
05.30.18 | >not super damaging to poll accuracy
source?
>It also wasn’t destiny that the email thing would come up or that it would stick going into Election Day.
not sure what your point is here but anyway nbd |
macman76
05.30.18 | 1) silver says the accuracy in polls hasn’t greatly reduced even as response rates have plummeted.
2) I guess i misread your initial response, if anything the polls recovered a little just before Election Day in favor of Clinton according to Nate’s model. She did, however, end up underperforming in the states she needed. So, they were inaccurate, possibly, because they were not responsive enough to that scandal. Nate seems to think so. https://www.google.com/amp/s/fivethirtyeight.com/features/the-comey-letter-probably-cost-clinton-the-election/amp/ |
