jt512
Philosopher
- Joined
- Sep 24, 2011
- Messages
- 5,089
Yeah, you're right. After that is accounted for, his model must be assuming that there is more uncertainty in the polls than the other forecasters' models are.
Ed FiveThirtyEight GE Forecast
- Thread starter Tony Stark
- Start date
Yeah, you're right. After that is accounted for, his model must be assuming that there is more uncertainty in the polls than the other forecasters' models are.
ddt
Mafia Penguin
Just to be sure I'm not dreaming, I just reloaded the 538 page.
The last run, as of this post, is 8 hours old.
In the Now-cast, both AZ and OH are red (AZ: Trump 50.8%; OH: Trump 54.8%).
In the Polls-only-cast, they're both blue
In the Polls-plus-cast, they're again both red
Tony Stark
Philosopher
- Joined
- Nov 22, 2014
- Messages
- 9,626
gypsyjackson
Master Poster
I stayed up to watch the 2008 results roll in, but I can't recall when they began. I intend to do so for this one, as some sort of catharsis to ease the horror and sick fascination/frustration this election has produced in me.
So can anyone tell me roughly when the first states will start to declare, or when the networks will start calling them? Either Eastern or GMT will be fine. Ta.
So can anyone tell me roughly when the first states will start to declare, or when the networks will start calling them? Either Eastern or GMT will be fine. Ta.
Tony Stark
Philosopher
- Joined
- Nov 22, 2014
- Messages
- 9,626
Probably around the time polls close on the east coast. So like 8pm Eastern.
Foolmewunz
Grammar Resistance Leader, TLA Dictator
If you look at the schedule, they apparently have strange hopes/needs.
Friday for Trump - NH and ME in the afternoon. He's not stealing NH but must believe the noise out of the libertarian and FOTL camps is promising. If they think it's close, that single vote in ME2 could be important, but there was someone on his team saying they thought they could take all of ME. Frankly, it sounds like they had a day to kill and are trying to show Donald as indefatigable. That image helped a hoarse Bill Clinton and helped Harry Truman. I doubt it works but their polling may show they're close enough that it's worth sending His Orageness.
Friday for Trump - IA. They're worried. Clinton is ahead in one poll and within the MoE in others. There's a solid Dem base in the state - enough to steal it if the GOP falls asleep.
Saturday for Trump - CO and AZ. He has no support from the CO GOP and not much hope. Maybe he's planning to incite a riot, fight with the fire marshals again, create some noise. Seems a filler on his way to AZ. It's scheduled for noon, so it'll be rushed as he has to get in and out of there and down to Phoenix for 3:00.
AZ? Yep. Clinton's team head-faked a big week there last week. He's now having to show up to keep AZ in the red column.
Pence is in the important/must win states - PA and NC. No surprise there - sensible moves in terms of the states' importance, but Pence gets no media time compared to Trump. It's the Trump schedule that's puzzling. I'd have Pence in AZ, CO, NH, ME and put Trump in PA and NC and even another run down to Friendly Florida (that's a different state from Florida Florida).
Last edited:
gypsyjackson
Master Poster
Sweet. 1am I can handle. Hopefully they'll be up to 270 by 3am (10ET)! Thanks very much.
Tsukasa Buddha
Other (please write in)
- Joined
- Sep 10, 2006
- Messages
- 15,302
For those following 538 numbers, keep in mind "polls-plus" has been worse than "polls-only". At least in last proper analysis I read. Google just shows him saying plus has been better 57% of the time.
Last edited:
Tony Stark
Philosopher
- Joined
- Nov 22, 2014
- Messages
- 9,626
If they call Florida, Ohio, North Carolina, or probably even New Hampshire for Hillary it will basically be over. Not technically perhaps but losing any of those states would make it almost impossible for him to win. And if Pennsylvania is called, he would basically have to win every swing state to win.
This is, I think, a very conservative map. Basically best case scenario for Trump
[IMGw=640]http://www.270towin.com/presidential_map_new/maps/pyoL6.png[/imgw]
gypsyjackson
Master Poster
Good to know - useful that the swing states are broadly early closers.
But the election isn't for a couple of weeks. No one wins or loses until then. Is the map showing what will happen in two weeks time? Are they allowed to change the map next week?
This is all very confusing.
Tony Stark
Philosopher
- Joined
- Nov 22, 2014
- Messages
- 9,626
If the polls swing to Donald bigly, I think that is the best be can hope for. He has no realistic path. Which additional blue state(s) do you think he can swing?
I have no idea. The only thing I've heard of interest is that every president since X (some god-awful long time ago) has won Ohio.
Tony Stark
Philosopher
- Joined
- Nov 22, 2014
- Messages
- 9,626
And if Donald does win Ohio, he still won't win. Unless he swings a bunch of other states too. As the map shows.
It would be more convincing if the election were tomorrow. It's like we are talking about going on vacation in a couple weeks and I'm obsessing about today's weather report. I get that it's the only weather report we have, it's just not the one I'm really interested in.
Tony Stark
Philosopher
- Joined
- Nov 22, 2014
- Messages
- 9,626
Again, this is a very conservative map not one I think is likely. And Hillary still wins with it. Unless you can name an additional blue state that Donald can flip I'm not even sure why you are replying.
Because I'm interested in the meaning of probability. I even started a thread on it.
Is it your contention that the map will not (or cannot) change in the next two weeks? Did it change in the last two weeks?
If the election were some kind of sum over previous polls, I think I'd get it. But I don't think that's how it works. I bet you could even find different maps posted in this very thread that don't match the current map. And if the map has changed, why do we think it is now stable?
This is like the stock market, where they warn you that "past performance is not necessarily indicative of future results" - critical thinking 101.
Tony Stark
Philosopher
- Joined
- Nov 22, 2014
- Messages
- 9,626
Yes it is my contention that the map will not change to the point that the very conservative map that still has Donald losing plus at least one additional blue state ends up being the real one. Telling that you can't even suggest which state this might be. Because it doesn't exist.
bonzombiekitty
Graduate Poster
- Joined
- Jun 23, 2011
- Messages
- 1,848
The map posted is the most conservative Clinton win scenario. Based on the current polling, in all the blue sections in that map, Clinton has a pretty solid lead.
The state that Trump has the best chance of flipping is PA, and Clinton leads in polls such that it would take a really wild swing in public perception of her or a systemic failure in polling for her to lose. There's less than 2 weeks left in the election - this close to the day, there's not much room for big swings. Not only is there just not much time left, by this point there's not many people left that can be swayed (granted, this election there seems to be more than normal) so something like that is very unlikely.
If Clinton takes PA, she's won unless something crazy happens.
Last edited:
I can't suggest it because I have no way to gather such data and I don't have any hunches to replace the missing data.
Just to make a point though - would you be surprised if the map went even more blue than now, that Hillary picked up more states (on the same source's map) in the next couple of weeks?
In other words, can you imagine (realistically) a major swing toward Hillary?
Last edited: