538 Model Megathread
       |           

Welcome, Guest. Please login or register.
Did you miss your activation email?
March 29, 2024, 05:25:04 AM
News: Election Simulator 2.0 Released. Senate/Gubernatorial maps, proportional electoral votes, and more - Read more

  Talk Elections
  Election Archive
  Election Archive
  2016 U.S. Presidential Election
  538 Model Megathread
« previous next »
Pages: 1 ... 27 28 29 30 31 [32] 33 34 35 36 37 ... 49
Author Topic: 538 Model Megathread  (Read 82820 times)
Gustaf
Moderators
Atlas Star
*****
Posts: 29,770


Political Matrix
E: 0.39, S: -0.70

Show only this user's posts in this thread
« Reply #775 on: October 17, 2016, 10:10:57 AM »

One thing that always stood out to me as weird in the 538 is their scenarios stuff at the end. Currently in the Nowcast the models says:

Clinton wins at least one state Mitt Romney won in 2012           86.4%
Trump wins at least one state President Obama won in 2012   54.1%

This seems really weird to me. It implies that they assign at least a 40% probability of Trump winning an Obama state while Clinton wins a Romney state. Generally these numbers have always struck me as too high throughout the campaign.
Logged
Middle-aged Europe
Old Europe
Atlas Icon
*****
Posts: 17,178
Ukraine


Show only this user's posts in this thread
« Reply #776 on: October 17, 2016, 10:11:33 AM »

Pulls-plus forecast is at a new record high with a 84.3% chance for Hillary to win the election, beating the previous record of 83.3% from last Wednesday.

Polls-only at 87.3% for Hillary, which is the highest number since August 17.

Correction, pulls-plus now at new record of 84.8% (+0.5) and polls-only at 87.7% (+0.4).
Logged
GeorgiaModerate
Moderators
Atlas Superstar
*****
Posts: 32,354


Show only this user's posts in this thread
« Reply #777 on: October 17, 2016, 10:16:46 AM »

One thing that always stood out to me as weird in the 538 is their scenarios stuff at the end. Currently in the Nowcast the models says:

Clinton wins at least one state Mitt Romney won in 2012           86.4%
Trump wins at least one state President Obama won in 2012   54.1%

This seems really weird to me. It implies that they assign at least a 40% probability of Trump winning an Obama state while Clinton wins a Romney state. Generally these numbers have always struck me as too high throughout the campaign.


Well, they're not independent events, so the odds of both happening are not quite that straightforward.  But it does seem like something that could easily happen, e.g. Trump winning Iowa while Clinton wins NC.
Logged
Blackacre
Spenstar3D
Sr. Member
****
Posts: 2,172
United States


Political Matrix
E: -5.35, S: -7.22

Show only this user's posts in this thread
« Reply #778 on: October 17, 2016, 10:17:58 AM »

One thing that always stood out to me as weird in the 538 is their scenarios stuff at the end. Currently in the Nowcast the models says:

Clinton wins at least one state Mitt Romney won in 2012           86.4%
Trump wins at least one state President Obama won in 2012   54.1%

This seems really weird to me. It implies that they assign at least a 40% probability of Trump winning an Obama state while Clinton wins a Romney state. Generally these numbers have always struck me as too high throughout the campaign.


Those numbers actually make a little sense, at least if you isolate them.

Clinton winning a Romney state? She's currently favoured in NC and Arizona and has an outside chance of winning Georgia or Alaska, and actually there's no EV where she has a >0.1% chance of winning, unlike Trump or either Romney or Obama.

Trump winning an Obama state? Every one of his winning scenarios counts on that. Plus he has a 35% shot in Iowa, a 31% shot in Ohio, a 22.5% shot in Nevada, and a 21.8% shot in Florida. Plus, if they count ME-2, he has a 46% shot there as well. States correlate, but not 100%
Logged
Gustaf
Moderators
Atlas Star
*****
Posts: 29,770


Political Matrix
E: 0.39, S: -0.70

Show only this user's posts in this thread
« Reply #779 on: October 17, 2016, 10:27:33 AM »

One thing that always stood out to me as weird in the 538 is their scenarios stuff at the end. Currently in the Nowcast the models says:

Clinton wins at least one state Mitt Romney won in 2012           86.4%
Trump wins at least one state President Obama won in 2012   54.1%

This seems really weird to me. It implies that they assign at least a 40% probability of Trump winning an Obama state while Clinton wins a Romney state. Generally these numbers have always struck me as too high throughout the campaign.


Well, they're not independent events, so the odds of both happening are not quite that straightforward.  But it does seem like something that could easily happen, e.g. Trump winning Iowa while Clinton wins NC.

Right, but them not being independent events makes the situation worse, because it makes it less likely Trump outperforms the polls to win Iowa while not at the same time maintaining North Carolina. 

My point was that it's at least 40% given those numbers. Which sounds high to me.

When the race was closer this was even more glaring.

So, to Zombie Spenstar - while states don't correlate perfectly, Clinton's win prob in NC is only marginally higher than that in IA and OH. It seems to me that there should be a fairly tiny slice of the prob dist in which she retains NC but loses one of the other ones.

I guess they have a lot of idiosyncratic uncertainty but these numbers indicate to me that maybe that uncertainty is too high. Is my point I suppose.
Logged
Blackacre
Spenstar3D
Sr. Member
****
Posts: 2,172
United States


Political Matrix
E: -5.35, S: -7.22

Show only this user's posts in this thread
« Reply #780 on: October 17, 2016, 10:34:38 AM »

So, to Zombie Spenstar - while states don't correlate perfectly, Clinton's win prob in NC is only marginally higher than that in IA and OH. It seems to me that there should be a fairly tiny slice of the prob dist in which she retains NC but loses one of the other ones.

I guess they have a lot of idiosyncratic uncertainty but these numbers indicate to me that maybe that uncertainty is too high. Is my point I suppose.

The states are also different in their elasticity and demographics. Should the race tighten again, Iowa will probably go atlas-blue before NC. And if Clinton's support falls among white voters without a college degree but remains steady with college-educated whites and nonwhites, that would be terrible for her in Iowa but wouldn't change too much of the math in NC.
Logged
Gustaf
Moderators
Atlas Star
*****
Posts: 29,770


Political Matrix
E: 0.39, S: -0.70

Show only this user's posts in this thread
« Reply #781 on: October 17, 2016, 10:47:43 AM »

So, to Zombie Spenstar - while states don't correlate perfectly, Clinton's win prob in NC is only marginally higher than that in IA and OH. It seems to me that there should be a fairly tiny slice of the prob dist in which she retains NC but loses one of the other ones.

I guess they have a lot of idiosyncratic uncertainty but these numbers indicate to me that maybe that uncertainty is too high. Is my point I suppose.

The states are also different in their elasticity and demographics. Should the race tighten again, Iowa will probably go atlas-blue before NC. And if Clinton's support falls among white voters without a college degree but remains steady with college-educated whites and nonwhites, that would be terrible for her in Iowa but wouldn't change too much of the math in NC.

Yeah, I'm aware of those aspects of the model. I guess I'm just skeptical of it being true to such an extent.
Logged
Sorenroy
Jr. Member
***
Posts: 1,697
United States


Political Matrix
E: -5.55, S: -5.91

P P
Show only this user's posts in this thread
« Reply #782 on: October 17, 2016, 01:34:13 PM »

Seems like the 538 model for Utah is automatically putting in all polls without McMuffin at 0.00 weight.
Logged
Gass3268
Moderators
Atlas Star
*****
Posts: 27,478
United States


Show only this user's posts in this thread
« Reply #783 on: October 17, 2016, 02:55:16 PM »

Every state in the blue wall is now over 85% for Clinton.
Logged
ApatheticAustrian
ApathicAustrian
Junior Chimp
*****
Posts: 6,603
Austria


Show only this user's posts in this thread
« Reply #784 on: October 17, 2016, 03:31:09 PM »

just wanted to share nate's take on utah with you.

http://fivethirtyeight.com/features/heres-how-were-forecasting-the-4-way-presidential-race-in-utah/
Logged
dspNY
Sr. Member
****
Posts: 2,802
United States


Show only this user's posts in this thread
« Reply #785 on: October 17, 2016, 05:41:25 PM »

HRC approaching 50% of the popular vote in the polls-only model
Logged
heatcharger
Junior Chimp
*****
Posts: 5,238
Sweden


Political Matrix
E: -1.04, S: -0.24

Show only this user's posts in this thread
« Reply #786 on: October 17, 2016, 06:25:34 PM »

Trump is back in single-digits in the now-cast, and NE-02 has flipped in the polls-only model.
Logged
Sorenroy
Jr. Member
***
Posts: 1,697
United States


Political Matrix
E: -5.55, S: -5.91

P P
Show only this user's posts in this thread
« Reply #787 on: October 17, 2016, 06:27:29 PM »

Trump is back in single-digits in the now-cast, and NE-02 has flipped in the polls-only model.

Speaking of CDs, ME-02 has flipped back to Clinton in all three models.
Logged
Mr. Morden
Atlas Legend
*****
Posts: 44,073
United States


Show only this user's posts in this thread
« Reply #788 on: October 17, 2016, 09:02:40 PM »

Tipping point state is now PA (in polls only) and MN (in polls plus and nowcast).
Logged
Dr. Arch
Arch
Atlas Icon
*****
Posts: 12,452
Puerto Rico


Show only this user's posts in this thread
« Reply #789 on: October 18, 2016, 01:11:56 AM »

As of this post, AZ is less than 3% points from flipping in polls-plus and over 60% Clinton in the now-cast.
Logged
Mr. Morden
Atlas Legend
*****
Posts: 44,073
United States


Show only this user's posts in this thread
« Reply #790 on: October 18, 2016, 01:19:30 PM »

Tipping point state is now PA (in polls only) and MN (in polls plus and nowcast).


And now with the latest updates, MN is the tipping point in all three models.  Might not last very long though, as it's pretty much tied with PA.
Logged
dspNY
Sr. Member
****
Posts: 2,802
United States


Show only this user's posts in this thread
« Reply #791 on: October 19, 2016, 06:48:30 AM »

With the Bloomberg Clinton +9, the PRRI Clinton +15 and the Arizona Clinton +4 poll, 538 will most likely have Clinton at or slightly above 50% of the popular vote in at least one of their models, possibly two (polls-only and Nowcast)
Logged
Gass3268
Moderators
Atlas Star
*****
Posts: 27,478
United States


Show only this user's posts in this thread
« Reply #792 on: October 19, 2016, 07:51:57 AM »

With the Bloomberg Clinton +9, the PRRI Clinton +15 and the Arizona Clinton +4 poll, 538 will most likely have Clinton at or slightly above 50% of the popular vote in at least one of their models, possibly two (polls-only and Nowcast)

This actually resulted in Clinton's win % dropping by 0.1, huh?
Logged
Brittain33
brittain33
Moderators
Atlas Star
*****
Posts: 21,933


Show only this user's posts in this thread
« Reply #793 on: October 19, 2016, 07:56:10 AM »

Hallelujah, they're now posting the impact of each update on the headline numbers!
Logged
Gustaf
Moderators
Atlas Star
*****
Posts: 29,770


Political Matrix
E: 0.39, S: -0.70

Show only this user's posts in this thread
« Reply #794 on: October 19, 2016, 08:01:29 AM »

With the Bloomberg Clinton +9, the PRRI Clinton +15 and the Arizona Clinton +4 poll, 538 will most likely have Clinton at or slightly above 50% of the popular vote in at least one of their models, possibly two (polls-only and Nowcast)

This actually resulted in Clinton's win % dropping by 0.1, huh?

They haven't posted the PRRI yet, ftr. And the -0.1 effect comes from lumping in the other polls with the LA tracker etc.
Logged
Erich Maria Remarque
LittleBigPlanet
YaBB God
*****
Posts: 3,646
Sweden


Show only this user's posts in this thread
« Reply #795 on: October 19, 2016, 08:15:34 AM »

Hallelujah, they're now posting the impact of each update on the headline numbers!
I was asking for this. It is really nice, but little bit too late. When Trump is at <10% and is falling down, there is no space for changing... Cheesy
Logged
dspNY
Sr. Member
****
Posts: 2,802
United States


Show only this user's posts in this thread
« Reply #796 on: October 19, 2016, 09:22:51 AM »

Arizona actually has a slightly higher chance of going Democratic than Iowa in the Nowcast

AZ: 59.4%
IA: 57.3%
Logged
Lief 🗽
Lief
Atlas Legend
*****
Posts: 44,876


Show only this user's posts in this thread
« Reply #797 on: October 19, 2016, 09:38:43 AM »

So all of the landslide national polls this morning actually caused Clinton's chance of winning to tick DOWN slightly Huh

Especially weird because Bloomberg polls has caused huge shifts in the past (because Nate absurdly overrates Selzer outside of Iowa).
Logged
calvinhobbesliker
Rookie
**
Posts: 33


Show only this user's posts in this thread
« Reply #798 on: October 19, 2016, 12:37:44 PM »

LOL, all the double digit polls changed nothing but a single IBT poll dropped Clinton by 1%.
Logged
Speed of Sound
LiberalPA
Atlas Icon
*****
Posts: 14,166
United States


Show only this user's posts in this thread
« Reply #799 on: October 19, 2016, 12:42:55 PM »

For comparison, Princeton Election Consortium now has their odds of a Clinton victory at 96%/98% in their two models with an EV projection of 337/201.
Logged
Pages: 1 ... 27 28 29 30 31 [32] 33 34 35 36 37 ... 49  
« previous next »
Jump to:  


Login with username, password and session length

Terms of Service - DMCA Agent and Policy - Privacy Policy and Cookies

Powered by SMF 1.1.21 | SMF © 2015, Simple Machines

Page created in 0.052 seconds with 13 queries.