|
ErIog posted:I like Harry Enten a lot, but this is some Monday-morning-quarterback bullshit. He needs to stay in his lane. Nah, he's absolutely right. Clinton lost Ohio by 10 points, why wasn't she defending her blue wall? Her campaign had the data, they should have been ready to move on it.
|
#
?
Nov 13, 2016 01:44
|
|
|
|
| # ? May 3, 2024 17:12 |
|
|
The systemic flaws in public polling were not necessarily present in the campaign's data. Some insiders have already been talking about how the results weren't far from what their data indicated.
|
|
#
?
Nov 13, 2016 01:45
|
|
|
Vox Nihili posted:The systemic flaws in public polling were not necessarily present in the campaign's data. Some insiders have already been talking about how the results weren't far from what their data indicated. I wasn't talking about systemic flaws in polling. I was talking about systemic flaws in the election itself. That's why I mentioned the FBI gently caress up and the VRA. I know it's easy to blame the top of the ticket because they're the top of the ticket, but I think there's probably more blame that should be directed at the DNC for not doing more earlier after the VRA got gutted. That's what I mean when I say Enten should stay in his lane. He knows polling and analytics. He knows gently caress all about campaigns and what could have been done differently to move those polls/analytics. He's also biasing his thinking towards the people who showed up to the polls without engaging with the fact that who showed up to the polls was precisely an issue and was a point of fuckery. So his opinion on this is accepting the framing that who showed up to the polls was legitimate. There were ~850 fewer polling places in the US for an electorate that continues to grow in size every election. It was harder to vote in 2016 than it was any previous modern election. That's significant. He doesn't care about that because his job requires him to accept that as a prior. It's a blind spot for him, and that's fine. There's other people covering that beat. He needs to recognize that's a blind spot and not spout off with some bullshit laying this entirely at the feet of Hillary's campaign like the other factors don't exist. ErIog has issued a correction as of 02:07 on Nov 13, 2016 |
|
#
?
Nov 13, 2016 02:00
|
|
|
Vox Nihili posted:The systemic flaws in public polling were not necessarily present in the campaign's data. Some insiders have already been talking about how the results weren't far from what their data indicated. trump campaigned here in minnesota just days before the election. this sounded insane when you compared it to all these big professional forecasters and pretty much all of the news media, for example: http://nymag.com/daily/intelligencer/2016/11/trump-and-pence-head-to-minnesota-in-sign-of-desperation.html yet when all was said and done trump lost by only 1.5 percentage points here, closer than any republican candidate since reagan almost beat native son mondale in his massive 1984 blowout. when i heard about the trump rally here i was confused too, but seeing his numbers in MN/WI/MI on election night made me think they probably had very good reason to stop here Lutha Mahtin has issued a correction as of 02:12 on Nov 13, 2016 |
|
#
?
Nov 13, 2016 02:06
|
|
|
ErIog posted:I wasn't talking about systemic flaws in polling. I was talking about systemic flaws in the election itself. That's why I mentioned the FBI gently caress up and the VRA. I know it's easy to blame the top of the ticket because they're the top of the ticket, but I think there's probably more blame that should be directed at the DNC for not doing more earlier after the VRA got gutted. So what? The campaign should be accounting for polling restrictions and focusing energy and resources where they need to be overcome. All you're saying is that there were other factors at play--but the campaign knew about most of those factors. It's not on Clinton personally (though she sucks badly), it's on the strategy people.
|
|
#
?
Nov 13, 2016 02:08
|
|
|
Vox Nihili posted:So what? The campaign should be accounting for polling restrictions and focusing energy and resources where they need to be overcome. Oh okay, her campaign should have been able to immediately deduce what the impacts of the first presidential race minus VRA protections in 50 years is going to be and perform perfectly to combat them. Her campaign tried. The voter registration data was something people were pointing to in the weeks before the election as a reason for her inevitable victory. The idea that her GOTV didn't try like hell is ridiculous. Please educate yourself on voting restrictions and their impact on this election. Comey letter plus VRA being gutted handed it to Trump. Hillary being a good candidate rather than a Barack Obama level great candidate was a factor. I think that's a dumb way to look at it, though. ErIog has issued a correction as of 02:22 on Nov 13, 2016 |
|
#
?
Nov 13, 2016 02:19
|
|
|
ErIog posted:Oh okay, her campaign should have been able to immediately deduce what the impacts of the first presidential race minus VRA protections is going to be and perform perfectly to combat them. Please educate myself? gently caress off, dude, I am extremely well-aware of the voter suppression going on in many states, including Wisconsin and North Carolina. The Clinton Campaign nevertheless bears substantial responsibility for its own failings. Trying to shelter it from criticism is not helpful.
|
|
#
?
Nov 13, 2016 02:22
|
|
|
Vox Nihili posted:Trying to shelter it from criticism is not helpful. How is endlessly bitching about Hillary helpful? The real work to be done in the future for Democrats has nothing to do with campaigning and everything to do with winning state/local races in order to help increase voter participation. Saying Hillary needed to visit Ohio more literally means nothing for 2020. The estimates are she'll win the popular vote by about 2 points. Hillary didn't have a messaging or campaigning problem. Those things could have been better, but they weren't bad. She had a turnout and voter participation problem exacerbated enough by voter suppression that it handed the election to Trump by razor thin margins in the swing states. ErIog has issued a correction as of 02:27 on Nov 13, 2016 |
|
#
?
Nov 13, 2016 02:24
|
|
|
ErIog posted:How is endlessly bitching about Hillary helpful? The real work to be done in the future for Democrats has nothing to do with campaigning and everything to do with winning state/local races in order to help increase voter participation. Saying Hillary needed to visit Ohio more literally means nothing for 2020. How is running away from what happened helpful? Hillary absolutely had a messaging problem, holy poo poo, wake up.
|
|
#
?
Nov 13, 2016 02:27
|
|
|
https://twitter.com/johnpodesta/status/786988264985100288?ref_src=twsrc%5Etfw Excellent message discipline, just excellent.
|
|
#
?
Nov 13, 2016 02:28
|
|
|
Vox Nihili posted:Excellent message discipline, just excellent. We can tell message discipline was an important factor in this election as evidenced by the winner of the EC having excellent message discipline.  Only messaging problem Hillary had was she refused to kowtow to bigots like Trump did.
|
|
#
?
Nov 13, 2016 02:29
|
|
|
ErIog posted:We can tell message discipline was an important factor in this election as evidenced by the winner of the EC having excellent message discipline. No one is going to argue that Trump ran a strong campaign. But the fact that he promised to bring back rust belt jobs mattered. Does he have any real policies that will do so? Of course not. At the end of the day the message was still effective. What message did Clinton present to the rust belt? Many counties flipped 20 points away from Obama's numbers. Suburbs that he won were lost. Rural areas that he lost by 10 were instead lost by 30. That's not just voter suppression; that's a huge, real reaction by voters.
|
|
#
?
Nov 13, 2016 02:34
|
|
|
Vox Nihili posted:https://twitter.com/johnpodesta/status/786988264985100288?ref_src=twsrc%5Etfw lolling @ this smug piece of poo poo
|
|
#
?
Nov 13, 2016 02:39
|
|
|
ErIog posted:
Thunk so, huh? You think "It's her turn" was a message that had any appeal to, like, anyone?
|
|
#
?
Nov 13, 2016 02:41
|
|
|
https://www.washingtonpost.com/news/wonk/wp/2016/11/12/the-advertising-decisions-that-helped-doom-hillary-clinton/
|
|
#
?
Nov 13, 2016 02:44
|
|
|
Hey hey, folks, I'm hiding in this thread to avoid the hindsight based 'who was to blame' stuff. Clinton's camp did mess up (debatable how obvious the errors were at the time though, remember when we criticised the dems for caring too much about the presidential and not the downticket? Whoops) but I agree that this doesn't give Enten any real basis to snark about the legitimate complaint about Comey's actions. That's just punditry, not analysis.FuturePastNow posted:Thunk so, huh? Wasn't that just a comment one of her supporters made early on? It's more a sarcastic anti-Hillary slogan than what she went with. Fangz has issued a correction as of 03:04 on Nov 13, 2016 |
|
#
?
Nov 13, 2016 03:02
|
|
|
Fangz posted:Wasn't that just a comment one of her supporters made early on? It's more a sarcastic anti-Hillary slogan than what she went with. #imwithher was literally her slogan. "It's her turn" is the sarcastic response that points out the futility of her message. "I'm an historic first" doesn't do much for people with real problems. It's masturbatory fodder for the privileged.
|
|
#
?
Nov 13, 2016 03:06
|
|
|
Vox Nihili posted:#imwithher was literally her slogan. "It's her turn" is the sarcastic response that points out the futility of her message. "I'm an historic first" doesn't do much for people with real problems. It's masturbatory fodder for the privileged. I'm not exactly seeing the connection here, but okay. Anyway, this is off topic, no?
|
|
#
?
Nov 13, 2016 03:19
|
|
|
Vox Nihili posted:The systemic flaws in public polling were not necessarily present in the campaign's data. Some insiders have already been talking about how the results weren't far from what their data indicated. Well I don't think 'not necessarily present' does not imply they weren't present. Indeed the campaign's data, if we were talking about differential response rates, could have been worse because of people not wanting to talk to the Clinton campaign. I don't see any process by which the campaign would have managed to avoid the systematic errors we saw, since the smart thing would have been to use industry standard methods, producing as Nate says, large errors on the state level. 'Not far' could just indicate they had a similar assessment to Nate - correlated errors existed, and it was a close election that could look very different on a small screen. You're reaching pretty far to claim that HRC knew they were 10 behind in Ohio via a super advanced polling method no one in the industry had figured out, and chose not to do anything about it because ???.
|
|
#
?
Nov 13, 2016 03:27
|
|
|
Fangz posted:Well I don't think 'not necessarily present' does not imply they weren't present. Indeed the campaign's data, if we were talking about differential response rates, could have been worse because of people not wanting to talk to the Clinton campaign. I don't see any process by which the campaign would have managed to avoid the systematic errors we saw, since the smart thing would have been to use industry standard methods, producing as Nate says, large errors on the state level. Because they chose to go for a blowout strategy rather than focusing on actually winning. Did you read this yet? https://www.washingtonpost.com/news/wonk/wp/2016/11/12/the-advertising-decisions-that-helped-doom-hillary-clinton/ "FiveThirtyEight.com calculated each state's likelihood of being the “tipping point” of the election — which is to say, the one that pushed the winner over 270 electoral votes. By its odds, both candidates were wise to invest in North Carolina (11.2 percent chance), Pennsylvania (12.3 percent) and Florida (17.6 percent). Arizona, on the other hand, had less than a 2 percent chance to tip the race. Georgia was about 2 percent. Iowa was a little over 1 percent. And Michigan, which Democrats basically ignored until the very last days of the race? Nearly a 12 percent chance. To be fair, massive ad spending still left Clinton short in Ohio, North Carolina, Florida and even Pennsylvania. But it appears that in divvying their ad budgets, Clinton and her allies forgot a simple rule of presidential elections: You don't need more than 270 electoral votes to win. By playing so aggressively for a blowout, Democrats allowed the Trump team to poach two of their must-win states."
|
|
#
?
Nov 13, 2016 03:31
|
|
|
Vox Nihili posted:Because they chose to go for a blowout strategy rather than focusing on actually winning. Did you read this from 538 back when these decisions were being made? http://fivethirtyeight.com/features/should-clinton-play-for-an-electoral-college-landslide/ Back before Comey when Clinton was 8 ahead and the map a swathe of blue, the consensus of the 538 panel was that Clinton should try to run up the numbers, to improve the downticket and build a mandate. Our Harry even recommended Arizona and Georgia. Remember what the worry was? http://fivethirtyeight.com/features/senate-update-clinton-is-surging-but-down-ballot-democrats-are-losing-ground/ Fangz has issued a correction as of 04:12 on Nov 13, 2016 |
|
#
?
Nov 13, 2016 03:57
|
|
|
Fangz posted:Back before Comey when Clinton was 8 ahead and the map a swathe of blue, the consensus of the 538 panel was that Clinton should try to run up the numbers, to improve the downticket and build a mandate. Our Harry even recommended Arizona and Georgia. 538 posted:micah: All right, so to put a bow on this: It seems like the consensus here is that Clinton should focus on her firewall states and maybe target one to two reach states? I’m the only one who thinks she should try to play everywhere? Sounds more like they thought she should be playing to win and that the race was more volatile than Clinton's campaign seemed to think.
|
|
#
?
Nov 13, 2016 04:14
|
|
|
Xenoveritas posted:Sounds more like they thought she should be playing to win and that the race was more volatile than Clinton's campaign seemed to think. quote:micah: All right, so to put a bow on this: It seems like the consensus here is that Clinton should focus on her firewall states and maybe target one to two reach states? I’m the only one who thinks she should try to play everywhere? I think you shouldn't have cut those paragraphs out. (I'm sure you have an innocent explanation  )But like, the point is that Clinton didn't play everywhere - where everywhere is defined as Alaska and Texas, what she did is closer to the strategy outlined here of a few reach states in Arizona and Georgia. If their internal polling is basically somewhere between 538 and Upshot (for instance by including that + 1-2% from GOTV that Nate suggests) that strategy makes a lot of sense at that point. Fangz has issued a correction as of 04:50 on Nov 13, 2016 |
|
#
?
Nov 13, 2016 04:20
|
|
|
ErIog posted:Oh okay, her campaign should have been able to immediately deduce what the impacts of the first presidential race minus VRA protections in 50 years is going to be and perform perfectly to combat them. if the dnc didn't abandon and mismanage the poo poo out of downticket races in favor of hillary's warchest, there would not have been nearly as much voter suppression
|
|
#
?
Nov 13, 2016 04:25
|
|
|
anime was right posted:if the dnc didn't abandon and mismanage the poo poo out of downticket races in favor of hillary's warchest, there would not have been nearly as much voter suppression No, there'd be just as much. Only, you know, there'd be the machinery in place to counter it a bit.
|
|
#
?
Nov 13, 2016 04:58
|
|
|
I haven't seen any real numbers from the Clinton campaign's internal polling. If it was anything like the public polling results some of Clinton's late campaign choices and messaging can be forgiven. The public polls indicated a Clinton loss was extremely unlikely and focusing on the swing states was the correct decision from a purely tactical point of view. We can argue about the larger scale strategy of pushing centrist policies and taking MI/WI/PA workers for granted, but the risk of losing rust belt states wasn't reflected in the (public) polling data. Between this and the recent Brexit result there's an argument to be made that pollster's have moved from being non-predictive to actively misleading by systematically underestimating their margin of error. Until pollsters get their act together and demonstrate they can accurately forecast voter turnout it would be fair to add a systematic error of ~5% in quadrature to the quoted margin of error for any US national or state poll. If this means the MOE is much greater than the margin between candidates in any given poll it just highlights the uselessness of modern polling.
|
|
#
?
Nov 13, 2016 20:29
|
|
|
that could work...if journalists and average people understood what is "margin of error" and "sampling" and other statistical concepts. even people who have good math skills in several areas often don't have a solid, basic understanding of statistics. i know i sure don't
|
|
#
?
Nov 13, 2016 21:56
|
|
|
Yeah, people just want the point estimate, when actually the uncertainty around that point estimate is far more important. I wouldn't solve the problem entirely, (arbitrary cut-offs ect), but I think the media should report probability ranges from polls, not the headline number. It would give people a much more honest idea of what was going on.
|
|
#
?
Nov 13, 2016 22:40
|
|
|
War Is Peace. Freedom Is Slavery. Trump Won In A Landslide. By Nate Silver
|
|
#
?
Nov 29, 2016 04:14
|
|
|
anime was right posted:if the dnc didn't abandon and mismanage the poo poo out of downticket races in favor of hillary's warchest, there would not have been nearly as much voter suppression what the gently caress are you talking about
|
|
#
?
Nov 29, 2016 05:15
|
|
|
Nate Silver is cool.
|
|
#
?
Nov 29, 2016 21:24
|
|
|
Nate drop his big "told you so" at the Harvard IOP: https://soundcloud.com/harvardiop/interview-nate-silver "[The New York Times] covered Clinton as the presumptive 45th President"
|
|
#
?
Dec 2, 2016 06:40
|
|
Vox Nihili posted:Nate drop his big "told you so" at the Harvard IOP: goddamn, nate throwing some serious shade at kellyann there. little dude has some fire in him.
|
|
|
#
?
Dec 2, 2016 07:33
|
|
|
Trumps Baby Hands posted:goddamn, nate throwing some serious shade at kellyann there. little dude has some fire in him. apparently the whole event was about 3 steps away from a riot. just open animosity throughout, especially between team clinton and team trump
|
|
#
?
Dec 2, 2016 07:37
|
|
|
good, gently caress the clinton campaign they SHOULD be salty
|
|
#
?
Dec 2, 2016 07:38
|
|
|
I will never look at another poll again.
|
|
#
?
Dec 2, 2016 07:42
|
|
|
Funny that national polls were less bad than state polls this time. I can't believe pollsters weren't accounting for education levels in their sampling... why would you not do that? That's pretty loving important! Hopefully they make the change and polls become good again in 2020 (I'm sure the midterm polls will keep sucking).
|
|
#
?
Dec 2, 2016 22:05
|
|
|
From this article on 538 it sounds like in the past education level hasn't mattered as much as other demographics, so most pollsters didn't attempt to control for it. Clearly it did this time, so in the future, it's likely going to be something they're going to try to track.
|
|
#
?
Dec 3, 2016 02:31
|
|
|
Xenoveritas posted:From this article on 538 it sounds like in the past education level hasn't mattered as much as other demographics, so most pollsters didn't attempt to control for it. Clearly it did this time, so in the future, it's likely going to be something they're going to try to track. Don't pollsters track population stability? The time to correct is as your models are loving up, not after they have just hosed up big time. A simple distribution would have told them that it had become a significant differentiator
|
|
#
?
Dec 5, 2016 10:28
|
|
|
|
| # ? May 3, 2024 17:12 |
|
|
Cheatum the Evil Midget posted:Don't pollsters track population stability? The time to correct is as your models are loving up, not after they have just hosed up big time. A simple distribution would have told them that it had become a significant differentiator Population stability isn't really the key issue. The key issue is that nobody ever got around to figuring out who shows up on election day and why. Pollsters treat it like a black box. They modeled turnout based on 2012 and 2008. They were wrong, and being wrong with the turnout just by that little bit made them fail to see Trump narrowly swinging the EC. You can talk about mistakes with crosstabs, but the biggest problem with all of the polling was the confidence that the same people who showed up in '08 and '12 were going to show up again in '16.
|
|
#
?
Dec 7, 2016 12:30
|
|