Nate Silver 538

MayDay10

Lerxst1992 said:

Lerxst1992 said:

mrussel1 said:

Lerxst1992 said:

Jearlpam0925 said:

Here we go - the good stuff:

https://fivethirtyeight.com/features/is-joe-biden-toast-if-he-loses-pennsylvania/

interesting article but it’s putting a chill in my burgeoning optimism. According to article, Biden’s chances drop about 60% if he loses PA and he is only up 5% there (which 538 and RCP both dropped 1.5% in the last week,almost 25% of Bidens PA lead). Just does not have the feel of a nearly 90% race, but I am done questioning that.

On the optimistic side, nearly 70 million have voted and about 25% who did not vote in 2016. That has to be a good sign for Biden. 

Considering Democratic leaners are less likely to vote AND Trump disparages voting by mail, this would be a very good sign.  Where did you get that info on the 25%?

I failed to mention that was for N.C. Very important state but could be a good news.

Texas youth vote up 600%, which is important because overall Texas vote is already almost 90% of 2016 total, so,it’s a big sample.

another canary in a coal mine?

https://www.ksat.com/vote-2020/2020/10/27/the-youth-vote-in-texas-is-up-by-more-than-600-from-last-presidential-election/

Nate Silver has alluded to it.... but it is equally as just possible that there is a 'magic bump' in Biden's turnout as there was for Trump's.  IMO, common sense tells me that is more likely.  I truly think Trump had his large pocket of motivated voters.  VERY motivated, and they turned out in 2016 in very high percentages, which tipped the polls off.  Polls have adjusted to that turnout.  Trump has not earned any significant voter bloc, and if anything has sloughed off voters.  Now, Trump's base has gone from VERY motivated to INSANELY motivated, but their vote still counts the same.  We just have to see childish parades, floatillas, and giant flags on pickup trucks.  But roughly its the same number.

I do think there is a wave of voters in support of Biden (probably moreso a rejection of Trump) that could be under the radar and unprecedented.  

Jearlpam0925

MayDay10 said:

Lerxst1992 said:

Lerxst1992 said:

mrussel1 said:

Lerxst1992 said:

Jearlpam0925 said:

Here we go - the good stuff:

https://fivethirtyeight.com/features/is-joe-biden-toast-if-he-loses-pennsylvania/

interesting article but it’s putting a chill in my burgeoning optimism. According to article, Biden’s chances drop about 60% if he loses PA and he is only up 5% there (which 538 and RCP both dropped 1.5% in the last week,almost 25% of Bidens PA lead). Just does not have the feel of a nearly 90% race, but I am done questioning that.

On the optimistic side, nearly 70 million have voted and about 25% who did not vote in 2016. That has to be a good sign for Biden. 

Considering Democratic leaners are less likely to vote AND Trump disparages voting by mail, this would be a very good sign.  Where did you get that info on the 25%?

I failed to mention that was for N.C. Very important state but could be a good news.

Texas youth vote up 600%, which is important because overall Texas vote is already almost 90% of 2016 total, so,it’s a big sample.

another canary in a coal mine?

https://www.ksat.com/vote-2020/2020/10/27/the-youth-vote-in-texas-is-up-by-more-than-600-from-last-presidential-election/

Nate Silver has alluded to it.... but it is equally as just possible that there is a 'magic bump' in Biden's turnout as there was for Trump's.  IMO, common sense tells me that is more likely.  I truly think Trump had his large pocket of motivated voters.  VERY motivated, and they turned out in 2016 in very high percentages, which tipped the polls off.  Polls have adjusted to that turnout.  Trump has not earned any significant voter bloc, and if anything has sloughed off voters.  Now, Trump's base has gone from VERY motivated to INSANELY motivated, but their vote still counts the same.  We just have to see childish parades, floatillas, and giant flags on pickup trucks.  But roughly its the same number.

I do think there is a wave of voters in support of Biden (probably moreso a rejection of Trump) that could be under the radar and unprecedented.  

Absolutely. I feel like it's a near majority the people voting for Biden are doing it more in support of ousting Trump than for their full support and admiration of Joseph Biden.

The Juggler

Shy Biden voters

https://thehill.com/homenews/campaign/523076-democrats-see-signs-of-hidden-biden-voters-flipping-from-gop

Poncier

Jearlpam0925 said:

Yeah, but he'd be the youngest of the 78's. He's actually healthier than 20 year olds, people have told me. 

I also believe that he and McConnell will never die.

Mitch looks like he won't make Thanksgiving...

OnWis97

Poncier said:

Jearlpam0925 said:

Yeah, but he'd be the youngest of the 78's. He's actually healthier than 20 year olds, people have told me. 

I also believe that he and McConnell will never die.

Mitch looks like he won't make Thanksgiving...

Imagine if that was true. Spending his last days owning the libs and working towards a one-party system. I don’t quite understand what makes him tick.

HughFreakingDillon

Poncier said:

Jearlpam0925 said:

Yeah, but he'd be the youngest of the 78's. He's actually healthier than 20 year olds, people have told me. 

I also believe that he and McConnell will never die.

Mitch looks like he won't make Thanksgiving...

insert barf emoji

tbergs

HughFreakingDillon said:

Poncier said:

Jearlpam0925 said:

Yeah, but he'd be the youngest of the 78's. He's actually healthier than 20 year olds, people have told me. 

I also believe that he and McConnell will never die.

Mitch looks like he won't make Thanksgiving...

insert barf emoji

Typical Mitch, pretends nothing is happening to be concerned about.

Halifax2TheMax

OnWis97 said:

Poncier said:

Jearlpam0925 said:

Yeah, but he'd be the youngest of the 78's. He's actually healthier than 20 year olds, people have told me. 

I also believe that he and McConnell will never die.

Mitch looks like he won't make Thanksgiving...

Imagine if that was true. Spending his last days owning the libs and working towards a one-party system. I don’t quite understand what makes him tick.

The robe and hood hanging in his closet and the giant metal cross in his barn.

Jearlpam0925

Poncier said:

Jearlpam0925 said:

Yeah, but he'd be the youngest of the 78's. He's actually healthier than 20 year olds, people have told me. 

I also believe that he and McConnell will never die.

Mitch looks like he won't make Thanksgiving...

I think we're both wrong and that he's already dead. He's just caring less about wearing the costume anymore. What a ghoul.

Spiritual_Chaos

Poncier said:

Jearlpam0925 said:

Yeah, but he'd be the youngest of the 78's. He's actually healthier than 20 year olds, people have told me. 

I also believe that he and McConnell will never die.

Mitch looks like he won't make Thanksgiving...

YOU

DO

NOT

TALK

ABOUT

FIGHT

CLUB

dignin

What To Make Of That New Wisconsin Poll That Has Biden Way Ahead

https://fivethirtyeight.com/features/what-to-make-of-that-new-wisconsin-poll-that-has-biden-way-ahead/

17 points ahead in an A+ poll. Believe the hype.

JimmyV

I want that Maine congressional district to flip so bad. It's the only electoral vote in all of New England that Trump "won" in 2016. 

Jearlpam0925

I'm really enjoying these people:

https://twitter.com/gelliottmorris/status/1321484610388271104?s=20

The Juggler

Jearlpam0925 said:

I'm really enjoying these people:

https://twitter.com/gelliottmorris/status/1321484610388271104?s=20

Yeah that's funny.

Did you know that Nate Silver blocked this guy on twitter? The two of them had a nerdy beef a couple months back about their models and stuff. I like them both but trust Silver's model more. 

HughFreakingDillon

teeing up the crown royal for my epic return from sober october to sit and enjoy trump's epic destruction. 

Jearlpam0925

The Juggler said:

Jearlpam0925 said:

I'm really enjoying these people:

https://twitter.com/gelliottmorris/status/1321484610388271104?s=20

Yeah that's funny.

Did you know that Nate Silver blocked this guy on twitter? The two of them had a nerdy beef a couple months back about their models and stuff. I like them both but trust Silver's model more. 

Yeah, I saw that, but I don't think this guy is The Model Guy. The Model Guy would be Andrew Gelman.

For shits and giggles - but how and why do you trust Silver's model more? Because it's been around? FWIW it looks like Gelman is no slouch and Economist's model included 538 as part of it.

I think the difference is Silver's may be overly conservative for nothing else but the sake of not wanting to perceived as "wrong" in a worse cast scenario. And this isn't me trying to shit on 538. I trust both models, I think I honestly just like how overtly aggressive The Econ's model is.

Gern Blansten

HughFreakingDillon said:

teeing up the crown royal for my epic return from sober october to sit and enjoy trump's epic destruction. 

Let's hope....I'll probably do an edible and have some bourbon until I either pass out with joy or sorrow

Jearlpam0925

Ah, very timely for this tweet from Gelman. Seems worth a read:

https://twitter.com/StatModeling/status/1321439083869196291?s=20

Jearlpam0925

I like these two comments from the excerpt:

"Marc says:

October 28, 2020 at 11:52 am

For all the (understandable) discuss of the differences between the 538 and Economist models, I think that we are loosing how much they have in common.

Both now have a close to 99% chance of Biden winning the popular vote, with a bit lower for electoral college. Note that this is very different from both the betting markets and the tone of most pundits, so that fact that you agree is significant.

Both have the same basic message about the history of the race. Polls have been steady for months, and thus the likelihood of the candidate who is ahead (Biden) to win has increased as the time left for additional movement has decreased. You had this from mid 80’s to high 90’s, 538 from low 80’s to low 90’s.

You have very similar list and ranking of the swing states and pivot state.

Some of the differences that you talk about above — e.g. the vote share in DC or WY, have virtually no consequences. Biden will win DC, Trump WY, and the popular vote in these are too small to move the national total by much. These predictions are based on very little data, as no one has bothered to survey. Feels like we are getting to angels on a pin territory.

At this point, it seems to me that the biggest uncertainty is about the process of voting — how many can vote early, who might be deterred from voting by COVID scares or long lines, what happens with counting the votes. Neither of you have made any attempt to model these — understandable given that there is no historical data.

Net, I’d say let’s all leave these models be now, and look again when we have the actual vote counts to compare.

Now get out and VOTE."

From Merlin, also a part of building this model. And amazingly has the name Merlin:
"We have been quite explicit in stating that we are only modeling vote intentions. Accounting for events such as those is fraught with uncertainty as it is not even clear where one would start. Elliott and I made adjustments for absentee ballot rejection rates (https://www.economist.com/graphic-detail/2020/10/24/postal-voting-could-put-americas-democrats-at-a-disadvantage) but compared to the local effect on voter intimidation that issue is at least tractable."

All fair. 

If you can work Election Day.....DO IT.

Jearlpam0925

Better still - right from Gelman's mouth (fingers?) re: the 538 model. Really enjoyed it.

https://statmodeling.stat.columbia.edu/2020/10/24/reverse-engineering-the-problematic-tail-behavior-of-the-fivethirtyeight-presidential-election-forecast/

"P.P.P.S. Let me say it again: I see no rivalry here. Nate's doing his best, he has lots of time and resource constraints, he's managing a whole team of people and also needs to be concerned with public communication, media outreach, etc.

My guess is that Nate doesn't really think that, a NJ win for Trump would make it less likely for him to win Alaska; it's just that he's really busy right now and he's rather reassure himself that his forecast is directionally the right approach than worry about where it's wrong. As I well know, it can be really hard to tinker with a model without making it worse. For example, he could increase the between-state correlations by adding a national error term, or by adding national and regional error terms, but then he'd have to decrease the variance within each state to compensate, and then there are lots of things to check, lots of new ways for things to go wrong---not to mention the challenge of explaining to the world that you've changed your forecasting method. Simpler, really, to just firmly shut that Pandora's box and pretend it had never been opened.

I expect that sometime after the election's over, Nate and his team will think about these issues more carefully and fix their model in some way. I really hope they go open source, but even if they keep it secret, as long as they release their predictive simulations we can look at the correlations and try to help out.

Similarly, they can help out with us. If there are any particular predictions from our model that Nate thinks don't make sense, he should feel free to let us know, or post it somewhere that we will find it. A few months ago he commented that our probability of Biden winning the popular vote seemed too high. We looked into it and decided that Nate and other people who'd made that criticism were correct, and we used that criticism to improve our model; see the "Updated August 5th, 2020" section at the bottom of this page. And our model remains improvable.

Let me say this again: the appropriate response to someone pointing out a problem with your forecasts is not to label the criticism as a "quibble" that is "more debatable than wrong" or to say that you're "directionally right," whatever that means. How silly that is! Informed criticism is a blessing! You're lucky when you get it, and use that criticism as an opportunity to learn and to do better."