12 ms·

This is so terrifying. If you start giving Trump states which are tossups, like FL, the others which were tossups begin to lean red. Choose red for those and yo

This is so terrifying. If you start giving Trump states which are tossups, like FL, the others which were tossups begin to lean red. Choose red for those and you're done. At that point you're in 2016 territory. We have an extremely broken election system.

mey 6y ago
538's map re-simulates with the assumption that if x's goes in y direction, it implies a certain correlation to other states behavior and remodels things.
- bhupy 6y ago
  I found the 538 tool to be a fantastic way to get the intuition for Bayesian probabilities. Wish I had this intuition in college.
dragontamer 6y ago
You're not thinking like a pollster. Florida is currently projected to be Biden's win. So what happens if Trump wins Florida? It means the polls were wrong. Which means, you'd have to reconsider the state-polls in all the other states with that new information: that the country is slightly redder than polls today suggest. If Trump wins Florida, then there's a "hidden vote", so to speak, that's pushing Trump's victory. Once you add that hypothetical hidden vote to the rest of the states, it becomes a closer race. --------- The question is whether or not you think there's a hidden vote for Trump. For all we know, there's a hidden vote for Biden this year: but we won't know that until the polls close and the votes are tallied up. In the absence of knowledge, the best we can do is find indications of cross-correlations and model those.
- nradov 6y ago
  There are numerous anecdotal reports of Trump supporters refusing to answer polls or giving intentionally misleading answers. But it's inherently impossible to know how prevalent that phenomenon is. Pollsters can't force voters to respond or give truthful answers.
  - wtallis 6y ago
    > But it's inherently impossible to know how prevalent that phenomenon is. There was a lot of analysis after the 2016 election cycle about the "shy Trump voter" hypothesis, and it really didn't hold up, and continues to not hold up with under the data available from this election cycle. It is completely unreasonable to postulate that the effect would be drastically bigger this year, after four years of Trump actually being in power and normalizing his status.
    - nradov 6y ago
      Do you have a reference for a good analysis of that hypothesis? On the surface it would seem like a plausible explanation for the 2016 election results, but I don't know how we could prove or disprove it.
      - wtallis 6y ago
        FiveThirtyEight published a well-sourced summary a month ago: https://fivethirtyeight.com/features/trump-supporters-arent-shy-but-polls-could-still-be-missing-some-of-them/ https://fivethirtyeight.com/features/trump-supporters-arent-... The main reasons to disbelieve the shy Trump voter hypothesis are that there's not a big difference in responses to live phone polls vs online polls, and that polls missed Senate and Gubernatorial races in 2016 by about the same amount as they missed the Presidential race.
    - AnimalMuppet 6y ago
      First: There are lots of "shy" voters. I'm one. I do not have time in my life for every opinion poll that calls me to take 5 or 10 or 20 minutes of my time. Even more so when some of them are push polls. No. Just go away. You'll find out what I think when I vote. Now, do people like me lean one way? "Get out of my face and leave me alone" people might plausibly lean toward Trump, but I wouldn't be dogmatic about it. Even more, I wouldn't claim that it's statistically significant. Second, you said: > It is completely unreasonable to postulate that the effect would be drastically bigger this year, after four years of Trump actually being in power and normalizing his status. Um, what? We've had four years of the media relentlessly telling us how stupid, evil, and wrong Trump is, and how stupid and racist all his supporters are. We've seen the media twist the facts to support this narrative. (Yes, Trump does so also. That's not the point.) I would be astonished if there aren't a bunch of Trump supporters who aren't going to tell pollsters what they think. (No, I can't prove it. Even if it's true, I can't quantify it. But I bet that they're out there, and that there are a lot of them.)
- dillondoyle 6y ago
  The polls aren't wrong and the state polls in 2016 weren't on the whole wrong either - one problem is there were too few. Right now there are a tons of Florida polls and they are basically all within the margin right now.
  - dragontamer 6y ago
    https://www.realclearpolitics.com/epolls/latest_polls/ https://www.realclearpolitics.com/epolls/latest_polls/ There are plenty of polls. The 2016 issue was that those polls were __correlated__. They were NOT independent polls, but somehow all the polls made the same mistake across the board. https://fivethirtyeight.com/features/election-update-why-our-model-is-more-bullish-than-others-on-trump/ https://fivethirtyeight.com/features/election-update-why-our... > Assumption No. 4: State outcomes are highly correlated with one another, so polling errors in one state are likely to be replicated in other, similar states. > Basically, this means that you shouldn’t count on states to behave independently of one another, especially if they’re demographically similar. If Clinton loses Pennsylvania despite having a big lead in the polls there, for instance, she might also have problems in Michigan, North Carolina and other swing states. What seems like an impregnable firewall in the Electoral College may begin to collapse. -------- Note that this post was written well before the results of 2016 were known. This was Nate Silver challenging the poll's assumptions back then. It was clear that "correlation" needs to be accounted for in people's models. That's the #1 issue regards to modern polls. Lo and behold: Pennsylvania swung right towards Trump, and all the polls that were predicting a Pennsylvania win for Clinton were wrong. But not only that, it proved a bunch of other polls wrong in other states. Such is the nature of Bayesian statistics: P(A, B) = P(A given B) * P(B). P(A given B) should equal P(A) in an ideal world. But we live in a messy world with correlations: understanding B will change our understanding of other probabilities.
    - dillondoyle 6y ago
      sure. but their post analysis says that national polling in 2016 was CLOSER than 2012. and state polls were even closer - but there weren't as many as this cycle https://fivethirtyeight.com/features/the-real-story-of-2016/ https://fivethirtyeight.com/features/the-real-story-of-2016/ https://fivethirtyeight.com/features/the-polls-are-all-right/ https://fivethirtyeight.com/features/the-polls-are-all-right...
- Simulacra 6y ago
  What will be more interesting is when the polls are wrong again on election day,
- greenpizza13 6y ago
  That’s precisely how I viewed it as well. What I was saying is it’s terrifying that a swing like that implies a larger swing.
rubidium 6y ago
The electoral college is designed (from the start) to prevent the larger, more populous states from dominating the country. Seems to be working as designed. And doing a good job as a system. Politics is extremely broken, yes. But I don’t think the electoral college is the fault.
- bobthepanda 6y ago
  What is broken is not the number of electoral votes per state but the winner takes all allocation system of the EC. As it stands today winner takes all makes big states like California and small states like Wyoming equally uncompetitive since there’s a snowball’s chance in hell they will ever flip majority the other direction. Proportionally allocating a state’s electoral votes would level the state by state playing field and align with popular vote, but you’d need either every state to agree or an amendment.
  - bryanlarsen 6y ago
    Another solution with potentially a clearer path: https://en.wikipedia.org/wiki/National_Popular_Vote_Interstate_Compact https://en.wikipedia.org/wiki/National_Popular_Vote_Intersta... What's really annoying is how close we came to abolishing the electoral college, that 3 segregationists were able to halt almost univeral consensus: https://en.wikipedia.org/wiki/Electoral_College_abolition_amendment#Bayh%E2%80%93Celler_amendment https://en.wikipedia.org/wiki/Electoral_College_abolition_am...
    - AnthonyMouse 6y ago
      The biggest problem with the National Popular Vote Interstate Compact is that in order to get a majority of the electoral college delegates into it, you would need states to sign on who benefit from the existing arrangement. You can't get the swing states because the existing system makes them the focus of national politics and you can't get always-red states because the party that controls them benefits from the extra two delegates per state which go predominantly to that party. And together they're the majority of the electoral college delegates. What you could do is do something like national popular vote weighted by electoral college delegates, because then you could get everyone but the swing states, and that would represent a majority. And then you're not disenfranchising half of both California and Texas.
rsynnott 6y ago
538's model has the (probably reasonable) belief that polling errors don't happen in a vacuum; if there's a polling error in one direction in a state, then other similar states probably have similar errors. Trump winning Florida would imply at least a 3.5 point error based on current polls, so not insubstantial.

mey 6y ago

538's map re-simulates with the assumption that if x's goes in y direction, it implies a certain correlation to other states behavior and remodels things.

bhupy 6y ago
I found the 538 tool to be a fantastic way to get the intuition for Bayesian probabilities. Wish I had this intuition in college.

dragontamer 6y ago

You're not thinking like a pollster. Florida is currently projected to be Biden's win. So what happens if Trump wins Florida? It means the polls were wrong. Which means, you'd have to reconsider the state-polls in all the other states with that new information: that the country is slightly redder than polls today suggest. If Trump wins Florida, then there's a "hidden vote", so to speak, that's pushing Trump's victory. Once you add that hypothetical hidden vote to the rest of the states, it becomes a closer race. --------- The question is whether or not you think there's a hidden vote for Trump. For all we know, there's a hidden vote for Biden this year: but we won't know that until the polls close and the votes are tallied up. In the absence of knowledge, the best we can do is find indications of cross-correlations and model those.

nradov 6y ago
There are numerous anecdotal reports of Trump supporters refusing to answer polls or giving intentionally misleading answers. But it's inherently impossible to know how prevalent that phenomenon is. Pollsters can't force voters to respond or give truthful answers.
- wtallis 6y ago
  > But it's inherently impossible to know how prevalent that phenomenon is. There was a lot of analysis after the 2016 election cycle about the "shy Trump voter" hypothesis, and it really didn't hold up, and continues to not hold up with under the data available from this election cycle. It is completely unreasonable to postulate that the effect would be drastically bigger this year, after four years of Trump actually being in power and normalizing his status.
  - nradov 6y ago
    Do you have a reference for a good analysis of that hypothesis? On the surface it would seem like a plausible explanation for the 2016 election results, but I don't know how we could prove or disprove it.
    - wtallis 6y ago
      FiveThirtyEight published a well-sourced summary a month ago: https://fivethirtyeight.com/features/trump-supporters-arent-shy-but-polls-could-still-be-missing-some-of-them/ https://fivethirtyeight.com/features/trump-supporters-arent-... The main reasons to disbelieve the shy Trump voter hypothesis are that there's not a big difference in responses to live phone polls vs online polls, and that polls missed Senate and Gubernatorial races in 2016 by about the same amount as they missed the Presidential race.
  - AnimalMuppet 6y ago
    First: There are lots of "shy" voters. I'm one. I do not have time in my life for every opinion poll that calls me to take 5 or 10 or 20 minutes of my time. Even more so when some of them are push polls. No. Just go away. You'll find out what I think when I vote. Now, do people like me lean one way? "Get out of my face and leave me alone" people might plausibly lean toward Trump, but I wouldn't be dogmatic about it. Even more, I wouldn't claim that it's statistically significant. Second, you said: > It is completely unreasonable to postulate that the effect would be drastically bigger this year, after four years of Trump actually being in power and normalizing his status. Um, what? We've had four years of the media relentlessly telling us how stupid, evil, and wrong Trump is, and how stupid and racist all his supporters are. We've seen the media twist the facts to support this narrative. (Yes, Trump does so also. That's not the point.) I would be astonished if there aren't a bunch of Trump supporters who aren't going to tell pollsters what they think. (No, I can't prove it. Even if it's true, I can't quantify it. But I bet that they're out there, and that there are a lot of them.)
dillondoyle 6y ago
The polls aren't wrong and the state polls in 2016 weren't on the whole wrong either - one problem is there were too few. Right now there are a tons of Florida polls and they are basically all within the margin right now.
- dragontamer 6y ago
  https://www.realclearpolitics.com/epolls/latest_polls/ https://www.realclearpolitics.com/epolls/latest_polls/ There are plenty of polls. The 2016 issue was that those polls were __correlated__. They were NOT independent polls, but somehow all the polls made the same mistake across the board. https://fivethirtyeight.com/features/election-update-why-our-model-is-more-bullish-than-others-on-trump/ https://fivethirtyeight.com/features/election-update-why-our... > Assumption No. 4: State outcomes are highly correlated with one another, so polling errors in one state are likely to be replicated in other, similar states. > Basically, this means that you shouldn’t count on states to behave independently of one another, especially if they’re demographically similar. If Clinton loses Pennsylvania despite having a big lead in the polls there, for instance, she might also have problems in Michigan, North Carolina and other swing states. What seems like an impregnable firewall in the Electoral College may begin to collapse. -------- Note that this post was written well before the results of 2016 were known. This was Nate Silver challenging the poll's assumptions back then. It was clear that "correlation" needs to be accounted for in people's models. That's the #1 issue regards to modern polls. Lo and behold: Pennsylvania swung right towards Trump, and all the polls that were predicting a Pennsylvania win for Clinton were wrong. But not only that, it proved a bunch of other polls wrong in other states. Such is the nature of Bayesian statistics: P(A, B) = P(A given B) * P(B). P(A given B) should equal P(A) in an ideal world. But we live in a messy world with correlations: understanding B will change our understanding of other probabilities.
  - dillondoyle 6y ago
    sure. but their post analysis says that national polling in 2016 was CLOSER than 2012. and state polls were even closer - but there weren't as many as this cycle https://fivethirtyeight.com/features/the-real-story-of-2016/ https://fivethirtyeight.com/features/the-real-story-of-2016/ https://fivethirtyeight.com/features/the-polls-are-all-right/ https://fivethirtyeight.com/features/the-polls-are-all-right...
Simulacra 6y ago
What will be more interesting is when the polls are wrong again on election day,
greenpizza13 6y ago
That’s precisely how I viewed it as well. What I was saying is it’s terrifying that a swing like that implies a larger swing.

rubidium 6y ago

The electoral college is designed (from the start) to prevent the larger, more populous states from dominating the country. Seems to be working as designed. And doing a good job as a system. Politics is extremely broken, yes. But I don’t think the electoral college is the fault.

bobthepanda 6y ago
What is broken is not the number of electoral votes per state but the winner takes all allocation system of the EC. As it stands today winner takes all makes big states like California and small states like Wyoming equally uncompetitive since there’s a snowball’s chance in hell they will ever flip majority the other direction. Proportionally allocating a state’s electoral votes would level the state by state playing field and align with popular vote, but you’d need either every state to agree or an amendment.
- bryanlarsen 6y ago
  Another solution with potentially a clearer path: https://en.wikipedia.org/wiki/National_Popular_Vote_Interstate_Compact https://en.wikipedia.org/wiki/National_Popular_Vote_Intersta... What's really annoying is how close we came to abolishing the electoral college, that 3 segregationists were able to halt almost univeral consensus: https://en.wikipedia.org/wiki/Electoral_College_abolition_amendment#Bayh%E2%80%93Celler_amendment https://en.wikipedia.org/wiki/Electoral_College_abolition_am...
  - AnthonyMouse 6y ago
    The biggest problem with the National Popular Vote Interstate Compact is that in order to get a majority of the electoral college delegates into it, you would need states to sign on who benefit from the existing arrangement. You can't get the swing states because the existing system makes them the focus of national politics and you can't get always-red states because the party that controls them benefits from the extra two delegates per state which go predominantly to that party. And together they're the majority of the electoral college delegates. What you could do is do something like national popular vote weighted by electoral college delegates, because then you could get everyone but the swing states, and that would represent a majority. And then you're not disenfranchising half of both California and Texas.

rsynnott 6y ago

538's model has the (probably reasonable) belief that polling errors don't happen in a vacuum; if there's a polling error in one direction in a state, then other similar states probably have similar errors. Trump winning Florida would imply at least a 3.5 point error based on current polls, so not insubstantial.