Nov 2023 Mod notice:
There may be other, more specific, threads about some aspects of US politics, but this one tends to act as a hub for all sorts of related news and information, so it's usually one of the busiest OTC threads.
If you're new to OTC, it's worth reading the Introduction to On-Topic Conversations
and the On-Topic Conversations debate guidelines
before posting here.
Rumor-based, fear-mongering and/or inflammatory statements that damage the quality of the thread will be thumped. Off-topic posts will also be thumped. Repeat offenders may be suspended.
If time spent moderating this thread remains a distraction from moderation of the wiki itself, the thread will need to be locked. We want to avoid that, so please follow the forum rules
when posting here.
In line with the general forum rules, 'gravedancing' is prohibited here. If you're celebrating someone's death or hoping that they die, your post will get thumped. This rule applies regardless of what the person you're discussing has said or done.
Edited by Mrph1 on Nov 30th 2023 at 11:03:59 AM
270 to Win normally works on mobile, you can play around with a number of different combinations on there.
You’re correct that the Clinton states plus the three mid-west states is enough for a Biden win, he could’ve even afford to loose a small Clinton state (so Nevada, New Hampshire or Maine) in such a scenario.
There are also other paths, like picking up Michigan, Arizona and North Carolina. Or Florida and any other swing state.
“And the Bunny nails it!” ~ Gabrael “If the UN can get through a day without everyone strangling everyone else so can we.” ~ CyranSpeaking of 270towin, since we're only 40 Days away from Election day, I think we should start sharing our maps from it.
Here is mine
, with a Biden win of 353-185. I think (as of today) Biden will flip: Arizona, Nebraska-2, Wisconsin, Ohio, Michigan, Pennsylvania, North Carolina and Florida (and Maine-2). Until Ginsburg died I was thinking he would also flip Iowa and Georgia, but I think the Evangelicals of these 2 will come home to roost (IE Come out in droves to vote for their guy).
Edited by DingoWalley1 on Sep 23rd 2020 at 11:33:31 AM
I didn't think 538 could be trusted. Weren't they way off in 2016?
BTW, the place I saw Trump was ahead in some states was this rolling update thread from Reddit:
https://old.reddit.com/r/politics/comments/iy8ken/discussion_thread_2020_general_election_daily/
To my recollection, they gave Trump a 30% or so chance by the end. Which is a much more substantial chance than most people think. The fact that trump won does not mean they were wrong, and to insist otherwise is a misunderstanding of statistics.
Like the day before the election 538 literally put out an article titled “Trump is one normal sized polling error away from winning”, they also gave him the best chance of any of the professional election forecasters (30%ish).
Reddit is certainly not to be trusted.
“And the Bunny nails it!” ~ Gabrael “If the UN can get through a day without everyone strangling everyone else so can we.” ~ CyranThey repeatedly and constantly said that Trump was within normal deviation of winning. They were practically the only place to suggest that Trump could win based on actual empirical models rather than wishy washy feelings.
So no, they are perfectly trustworthy. They even give a full range of results and probabilities for every statewide race ffs.
They were closer than most other polling aggregates.
Now, Nate Silver in particular has made some embarrassing errors, in particular tweeting 6 hours apart that pundits were overstating Trump's chances of trying to soft coup and having an Oh, Crap! reaction to Trump refusing to say he'd let a peaceful transition of power happen and implying the opposite, but as far as polling aggregates go they're pretty good.
Edited by AzurePaladin on Sep 23rd 2020 at 11:36:57 AM
The awful things he says and does are burned into our cultural consciousness like a CRT display left on the same picture too long. -FighteerBy factoring them into a generic numerical term which basically widens or narrows the range that the simulation is going to be sampling outcomes from. This is no different than polls trying to estimate how their sample represents the population and adjust to compensate.
Like... this is how you model things with uncertain variables. You can't just pretend they're not there.
Sure, you can complain that the things being estimated in don't make sense, but it's not like he's saying "a big NY Times headline is a 1% shift to Trump!"
2012 Nate Silver was the guy saying "Why does anyone bother with data that's not polls?"
2020 Nate Silver is just "It could be this or it could be that, the conclusion is that there's no conclusion." And no, factoring the size of headlines into his model based on nothing more than their physical size is not sound, just because Silver claims it "creates uncertainty."
![]()
That was my point yesterday. In 2012 he'd been entirely focused on sports models and had turned that to political polling.
He's had 8 years of political experience and two presidential elections to consider how other terms might factor into things, and what's easy to abstract into a minor inclusion to polling terms.
Honestly the font size of the NY Times headline being in there makes bugger all difference practically. It's probably, what, a *1.01 modifier on the uncertainty term which already includes much bigger things?
Except he hasn't, because he's not ascribing them with definite positive or negative effects, only using them to tweak the variance.
Edited by RainehDaze on Sep 23rd 2020 at 5:23:09 PM

Real Clear Politics isn't that great because they're biased in the polls that they include in their averages.