You should keep it in mind that 538 != Nate Silver. 538 had the election in favor of Biden. No one under gods blue heaven knows what kind of crack their model was smoking.
It is notable that Morris’s model was the only one that was giving Biden the edge because it was assuming events in the lead up to the election would benefit Biden. As we get closer to the election, the model would put more emphasis on polls than those (presumed) events.
Yeah the main reason I posted it was because it still is decent, free poll aggregate and does a good job of visually showing a path to victory. The interactive map also does a good job of showing which states are the key swing states.
But I agree Morris’s model is very flawed; in the past week alone it showed several reputable polls as having Trump in the lead (some by +5 and +6 points) and none showing Biden in the lead. Despite that, it still said Biden had around a 50% chance or more to win. To make it even more questionable, both The Economist and Nate Silver believed Biden would lose 2/3rds of the time.
Basically, it seems to rely heavily on the assumption that future events would play out extremely well for Biden and not at all for Trump, which in my opinion is a major flaw (especially given how crazy this election cycle has been already).
I guess we will see how much 538’s forecast changes as Morris emphasizes the polls more than “events” closer to the election. Perhaps it will be more realistic, because right now I agree it is not.
Yeah the tools that were built up under Nate are great. I use them all the time.
I’m extremely interested to see how Kamala’s polling shifts over the next few weeks. I’m always surprised by how long it takes the news to actually make it into the polling. I don’t expect things to be fully cooked for at least 2 weeks, more like 3. She just got in the race, lets see how she does, lets see what she says. I think Bidens poll numbers were also dragging Kamala down.
I see a lane for her to step left (oppose) Gaza/ Israel, and get back into the game in the upper mid-west, and to also put both NC and GA back into play. It makes a Josh Shapiro VP pick super obvious, because he can work Penn and thats the ball game.
You should keep it in mind that 538 != Nate Silver. 538 had the election in favor of Biden. No one under gods blue heaven knows what kind of crack their model was smoking.
Nate Silver had Biden to lose at %70+.
Nates most recent blog on the matter: https://www.natesilver.net/p/biden-and-democrats-make-the-rational
tldr: 538 is crap any more.
edit: new post from Nate https://www.natesilver.net/p/harris-is-the-clear-choice-of-democratic
edit edit: fun one. Nate throwing 538 under the bus https://www.natesilver.net/p/why-i-dont-buy-538s-new-election
Yes, Nate Silver has his own website now which unfortunately (but understandably) has a paywall to access some content.
G. Elliot Morris is now heading up 538’s model.
It is notable that Morris’s model was the only one that was giving Biden the edge because it was assuming events in the lead up to the election would benefit Biden. As we get closer to the election, the model would put more emphasis on polls than those (presumed) events.
Yeah I think any one with two eyes and any level of understanding of politics was both laughing at and shitting on the 538 predictions.
Like, so bad as to be considered in the realm of propaganda bad.
Yeah the main reason I posted it was because it still is decent, free poll aggregate and does a good job of visually showing a path to victory. The interactive map also does a good job of showing which states are the key swing states.
But I agree Morris’s model is very flawed; in the past week alone it showed several reputable polls as having Trump in the lead (some by +5 and +6 points) and none showing Biden in the lead. Despite that, it still said Biden had around a 50% chance or more to win. To make it even more questionable, both The Economist and Nate Silver believed Biden would lose 2/3rds of the time.
Basically, it seems to rely heavily on the assumption that future events would play out extremely well for Biden and not at all for Trump, which in my opinion is a major flaw (especially given how crazy this election cycle has been already).
I guess we will see how much 538’s forecast changes as Morris emphasizes the polls more than “events” closer to the election. Perhaps it will be more realistic, because right now I agree it is not.
Yeah the tools that were built up under Nate are great. I use them all the time.
I’m extremely interested to see how Kamala’s polling shifts over the next few weeks. I’m always surprised by how long it takes the news to actually make it into the polling. I don’t expect things to be fully cooked for at least 2 weeks, more like 3. She just got in the race, lets see how she does, lets see what she says. I think Bidens poll numbers were also dragging Kamala down.
I see a lane for her to step left (oppose) Gaza/ Israel, and get back into the game in the upper mid-west, and to also put both NC and GA back into play. It makes a Josh Shapiro VP pick super obvious, because he can work Penn and thats the ball game.