Playing around with this, really my main take away is they seem significantly more overconfident in their model than I thought. I've been watching their main forecast for a while, and that seemed reasonable enough, if probably overestimating Biden's chances. It's obviously making a million assumptions like any political model would need to, but with enough care and self awareness about its strengths and weaknesses by the people making it, that can be a pretty useful tool.
But now I see they've chosen to make this tool, which is basically designed to explore scenarios where the model breaks down, and then make predictions using the same model that just failed to predict that hypothetical. It's likely that something that doesn't fit into their assumptions will cause some states to go against their predictions, and in real life they may use that info to update their model in the future, but here they're still trying to apply the exact same assumptions again. It really makes me think they're not being careful enough about the limitations of their model.
Edit: Oh wow, I originally saw the "?cid=abcnews" in the url and thought maybe it was only a small project made for like a specific news story or something, and I might be reading more into it than it warranted, but no, it's a main, attention grabbing part of their homepage, and they seem to want everybody to play with it.
Great comment, you've made me really re-consider Trump's chances after I began playing with this map too.
Five Thirty Eight are usually spot on with their stuff, but they have gotten extremely arrogant this past year. The podcast called "Trump's chances to still win" mostly is them just bragging about how they think Democrats are doing so well, then turn around and criticize Biden for his lack of message on policies and such saying that might hurt him, but they keep banging the drum "it's going to be a blow out! Dems seem to know this and are pouring their money in red states!"
The whole reason they waited until now to do a map is cause they've been criticized for the past 4 years over how Hillary lost even though they gave Trump better chances then.
Now that I think about it, even their model polling Trump is pretty flawed. Every time Trump goes up a bit in approval, they don't add that in with how it stacks up vs Biden's approval.
In one of their recent podcasts, Five Thirty Eight's crew literally talked about that, and they also started going off about how Texas could turn blue, which is a fucking joke to anyone who actually lives out there.
I honestly don't think the moderate Republicans or Anti-Trump Republicans are that big of a group. I haven't checked in a while, but Trump going into this year had an over 90% approval rating with the GOP's voter base. That approval rating managed to increase after all the riots in June. He is by far the most popular Republican president since Reagan. The Republicans love him. It just so happens that all the anti-Trump ones are in the media and mainstream news has an extreme liberal bias.
In the event of Republicans actually voting for Biden and the down ballot as you mention, you raise a great point. The GOP could actually win back seats in the House and keep a large chunk of the Senate.
Liberals are idiots and don't think about this at all with the Never Trumpers. Just cause they might not like Trump don't mean they're going to betray their own party.
This has been quite a day.... Up until this thread with the Five Thirty Eight map and now looking at all this, I am beginning to truly wonder if he has a real chance of winning. On the map that Five Thirty Eight has, if you play around with it and just give Trump one of the swing states, the entire thing shifts from Biden at 88% to Trump at over 60%. Color in Florida for Trump and watch the entire map flip for him.
Maybe Biden's lead really is overrated and Five Thirty Eight themselves got so much criticism for the 2016 models that they are running a very flawed model here making it seem Biden has it in the bag.
They recently have been running articles about the potential of Trump winning Minnesota and Wisconsin, which more or less goes to show how the Democrats are running a crappy campaign to the point they could lose one of those states if not both of them. Five Thirty Eight's defense of this is to just yell "these were never safe blue states anyway!" but it still proves a point there that their calculations and models are not reflecting the greater reality. Both of those states opened up after the riots and Dems horrible response to all of the police brutality.
This race might not really be a blow out then. I can actually see Trump taking it now as I am reminded that the enthusiasm gap is also very real. I live in a deep red state but it don't matter, I never see any signs for Biden or hear anything positive about him locally. Trump on the other hand? His base is louder and more rabid than they were in 2016.
I agree with you. It's made up by the liberal media.
I live in a deep red state and I can tell you back last year, I started noticing more Trump signs popping up than all I had seen in 2016. This was before the primaries, yet I was seeing Trump crap in places I didn't see it whatsoever back in 2016. Now? It's tripled. If there is one thing I have learned about this election is that the enthusiasm gap is real. I didn't see a single Clinton sign in all of 2016 and I have yet to see a single Biden sign anywhere. Libs will smugly remind me "that's cause you live in a red state!" but that's entirely true. I am old enough to have lived through Kerry's failed 2004 run and there were tons of Kerry signs out here back then, and for both 2008 and 2012, I seen LOTS of Obama signs despite him losing my state.
I have known conservatives who hate Trump and guess who they were most interested in? Bernie Sanders. Definitely not Joe Biden. These conservative leaning people were on board with the idea of M4A and programs like that. They fucking hate Biden and see him as part of the status quo they are against.
Every "moderate" Republican I know is voting for Trump. They talk about how Biden was Obama's VP and they absolutely hated Obama and don't want to go back to that era. Some of them will smartly bring up how Biden and Obama were responsible for the 2009 recession. Usually the go-to argument though is that they despised Obama, or they'll say Biden has dementia. Lately what I have been hearing locally is about Biden's tax plan and his gun control platform that makes them seethe in rage.
It is so amusing to me seeing the Democratic party trot out these old neo-cons to try and reach out to conservative voters. Talk to conservatives, they fucking hate McCain, Romney, Bush, and all these other has-beens. When a Republican politician reaches across the aisle to the Dems or when they leave office, they become RINOs to their voters. Bush has been widely hated by GOP voters since the day he left the oval office, and yet here are Dems pushing Biden and trying to get Bush to endorse him.
Playing around with this, really my main take away is they seem significantly more overconfident in their model than I thought. I've been watching their main forecast for a while, and that seemed reasonable enough, if probably overestimating Biden's chances. It's obviously making a million assumptions like any political model would need to, but with enough care and self awareness about its strengths and weaknesses by the people making it, that can be a pretty useful tool.
But now I see they've chosen to make this tool, which is basically designed to explore scenarios where the model breaks down, and then make predictions using the same model that just failed to predict that hypothetical. It's likely that something that doesn't fit into their assumptions will cause some states to go against their predictions, and in real life they may use that info to update their model in the future, but here they're still trying to apply the exact same assumptions again. It really makes me think they're not being careful enough about the limitations of their model.
Edit: Oh wow, I originally saw the "?cid=abcnews" in the url and thought maybe it was only a small project made for like a specific news story or something, and I might be reading more into it than it warranted, but no, it's a main, attention grabbing part of their homepage, and they seem to want everybody to play with it.
Great comment, you've made me really re-consider Trump's chances after I began playing with this map too.
Five Thirty Eight are usually spot on with their stuff, but they have gotten extremely arrogant this past year. The podcast called "Trump's chances to still win" mostly is them just bragging about how they think Democrats are doing so well, then turn around and criticize Biden for his lack of message on policies and such saying that might hurt him, but they keep banging the drum "it's going to be a blow out! Dems seem to know this and are pouring their money in red states!"
The whole reason they waited until now to do a map is cause they've been criticized for the past 4 years over how Hillary lost even though they gave Trump better chances then.
Now that I think about it, even their model polling Trump is pretty flawed. Every time Trump goes up a bit in approval, they don't add that in with how it stacks up vs Biden's approval.
deleted by creator
In one of their recent podcasts, Five Thirty Eight's crew literally talked about that, and they also started going off about how Texas could turn blue, which is a fucking joke to anyone who actually lives out there.
deleted by creator
I honestly don't think the moderate Republicans or Anti-Trump Republicans are that big of a group. I haven't checked in a while, but Trump going into this year had an over 90% approval rating with the GOP's voter base. That approval rating managed to increase after all the riots in June. He is by far the most popular Republican president since Reagan. The Republicans love him. It just so happens that all the anti-Trump ones are in the media and mainstream news has an extreme liberal bias.
In the event of Republicans actually voting for Biden and the down ballot as you mention, you raise a great point. The GOP could actually win back seats in the House and keep a large chunk of the Senate.
Liberals are idiots and don't think about this at all with the Never Trumpers. Just cause they might not like Trump don't mean they're going to betray their own party.
deleted by creator
Holy shit, I didn't know it was that high.
This has been quite a day.... Up until this thread with the Five Thirty Eight map and now looking at all this, I am beginning to truly wonder if he has a real chance of winning. On the map that Five Thirty Eight has, if you play around with it and just give Trump one of the swing states, the entire thing shifts from Biden at 88% to Trump at over 60%. Color in Florida for Trump and watch the entire map flip for him.
Maybe Biden's lead really is overrated and Five Thirty Eight themselves got so much criticism for the 2016 models that they are running a very flawed model here making it seem Biden has it in the bag.
They recently have been running articles about the potential of Trump winning Minnesota and Wisconsin, which more or less goes to show how the Democrats are running a crappy campaign to the point they could lose one of those states if not both of them. Five Thirty Eight's defense of this is to just yell "these were never safe blue states anyway!" but it still proves a point there that their calculations and models are not reflecting the greater reality. Both of those states opened up after the riots and Dems horrible response to all of the police brutality.
This race might not really be a blow out then. I can actually see Trump taking it now as I am reminded that the enthusiasm gap is also very real. I live in a deep red state but it don't matter, I never see any signs for Biden or hear anything positive about him locally. Trump on the other hand? His base is louder and more rabid than they were in 2016.
deleted by creator
deleted by creator
I agree with you. It's made up by the liberal media.
I live in a deep red state and I can tell you back last year, I started noticing more Trump signs popping up than all I had seen in 2016. This was before the primaries, yet I was seeing Trump crap in places I didn't see it whatsoever back in 2016. Now? It's tripled. If there is one thing I have learned about this election is that the enthusiasm gap is real. I didn't see a single Clinton sign in all of 2016 and I have yet to see a single Biden sign anywhere. Libs will smugly remind me "that's cause you live in a red state!" but that's entirely true. I am old enough to have lived through Kerry's failed 2004 run and there were tons of Kerry signs out here back then, and for both 2008 and 2012, I seen LOTS of Obama signs despite him losing my state.
I have known conservatives who hate Trump and guess who they were most interested in? Bernie Sanders. Definitely not Joe Biden. These conservative leaning people were on board with the idea of M4A and programs like that. They fucking hate Biden and see him as part of the status quo they are against.
Every "moderate" Republican I know is voting for Trump. They talk about how Biden was Obama's VP and they absolutely hated Obama and don't want to go back to that era. Some of them will smartly bring up how Biden and Obama were responsible for the 2009 recession. Usually the go-to argument though is that they despised Obama, or they'll say Biden has dementia. Lately what I have been hearing locally is about Biden's tax plan and his gun control platform that makes them seethe in rage.
It is so amusing to me seeing the Democratic party trot out these old neo-cons to try and reach out to conservative voters. Talk to conservatives, they fucking hate McCain, Romney, Bush, and all these other has-beens. When a Republican politician reaches across the aisle to the Dems or when they leave office, they become RINOs to their voters. Bush has been widely hated by GOP voters since the day he left the oval office, and yet here are Dems pushing Biden and trying to get Bush to endorse him.