I have to say, the data is displayed beautifully. I hope for people in the us that Biden wins also seeing how republicans just outright deny science. Not just with the corona virus, but also...
I have to say, the data is displayed beautifully. I hope for people in the us that Biden wins also seeing how republicans just outright deny science. Not just with the corona virus, but also climate change.
And I have to say: almost 50 percent voting for Trump is something I find hard to grasp. I can’t imagine how people can still support this insanity
The mascot character is really amusing to me, because a lot of what they're using it for is emphasizing that their forecast is just that, a forecast. It's based on their best models and the best...
The mascot character is really amusing to me, because a lot of what they're using it for is emphasizing that their forecast is just that, a forecast. It's based on their best models and the best data available, but it's still just a forecast. They said all of those things in their 2016 forecasts too, but it was usually buried in the articles accompanying the fancy graphics.
You have to remember that 50% means 50% of people who actually vote, which is not even a majority of Americans. Still way too many people supporting Trump, but at least it's not half of the whole...
You have to remember that 50% means 50% of people who actually vote, which is not even a majority of Americans. Still way too many people supporting Trump, but at least it's not half of the whole population.
I would also note the article linked in that snapshot: "Its way too soon to count Trump out". It's more that people are realllly bad at probability. Flip a coin twice. If you get two heads, that's...
I would also note the article linked in that snapshot: "Its way too soon to count Trump out". It's more that people are realllly bad at probability.
Flip a coin twice. If you get two heads, that's less likely than the chance this model gives Trump of winning.
People see it as like a score, or vote, not as a bernoulli
Yeah they said those things constantly leading up to the 2016 election, and emphasized that the polls were within the margin or error and if they broke toward trump it could flip a bunch of...
Yeah they said those things constantly leading up to the 2016 election, and emphasized that the polls were within the margin or error and if they broke toward trump it could flip a bunch of important states. And, lo and behold, that's exactly what happened.
Not sure what point you're trying to make here? If you're suggesting that FiveThirtyEight is inaccurate because Trump won, then it's important to remember that a 28% chance is not a 0% chance....
Not sure what point you're trying to make here? If you're suggesting that FiveThirtyEight is inaccurate because Trump won, then it's important to remember that a 28% chance is not a 0% chance. That page never said Trump wouldn't win; just that it was less likely.
To tack onto this, the popular vote wasn't that far off, interestingly enough. In fact, Hillary's popular vote prediction is pretty much spot on. As for Trump, it seems that they predicted more...
To tack onto this, the popular vote wasn't that far off, interestingly enough. In fact, Hillary's popular vote prediction is pretty much spot on.
As for Trump, it seems that they predicted more people would vote 3rd party and steal votes from Trump by doing so. Instead, more predicted 3rd party voters cast their ballot for Trump. I suspect a lot of these were protest/chaos votes personally, but who knows.
Really just shows the power of state delegates either way. They miscalculated Trumps popular vote by only 1.2%, but his delegates by 69.
No? I'm saying projections don't equate to winning an election. People have to go out and vote, and at that, not toss their votes to the third party this time.
No? I'm saying projections don't equate to winning an election. People have to go out and vote, and at that, not toss their votes to the third party this time.
With that said, one shouldn’t get too carried away with the comparisons to four years ago. In 2016, the reason Trump had a pretty decent chance in our final forecast was mostly just because the polls were fairly close (despite the media narrative to the contrary), close enough that even a modest-sized polling error in the right group of states could be enough to give Trump a victory in the Electoral College.
The uncertainty in our current 2020 forecast, conversely, stems mostly from the fact that there’s still a long way to go until the election. Take what happens if we lie to our model and tell it that the election is going to be held today. It spits out that Biden has a 93 percent chance of winning. In other words, a Trump victory would require a much bigger polling error than what we saw in 2016.
I wish they would acknowledge the potential for election fraud, voter suppression, etc., even if you can't really account for it. But it is nice to see Biden is favored regardless.
I wish they would acknowledge the potential for election fraud, voter suppression, etc., even if you can't really account for it. But it is nice to see Biden is favored regardless.
I find this wonderfully reassuring. As an international mafia tries to ramrod the Apocalypse through my country’s election system, it’s nice having a cheerful cartoon fox in wonky glasses remind...
I find this wonderfully reassuring. As an international mafia tries to ramrod the Apocalypse through my country’s election system, it’s nice having a cheerful cartoon fox in wonky glasses remind me how they could easily win with popular support. There he is brandishing a jar of red balls and blue balls, like it’s a state lotto draw, or a kindergarten sorting game. It’s not that such representations are unique; it’s how fantastically incongruous they are to the present moment that chills me. Then, perhaps the ability to treat any topic to a serious, measured discussion has been un-educated out of Americans, and is part of the reason we will soon get to live in a lazy-Nazi dystopia.
The point of the page is to talk about probability and statistics. Probability does come from games of chance, so it seems thematically appropriate? It emphasizes uncertainty and a certain kind of...
The point of the page is to talk about probability and statistics. Probability does come from games of chance, so it seems thematically appropriate? It emphasizes uncertainty and a certain kind of unpredictability, that the outcome is unknown and they aren't "calling" it.
And that's good as far as it goes, but there is the problem that the world is not a casino.
I'm willing to forgive this, because it's still better than the fatalism you often see, where we are doomed and so nothing we do matters, or the future depends on huge, impersonal, inevitable forces, and so nothing we do matters.
I love 538. That said, for me at least, it feels frankly disingenuous at this point to make predictions like this, without at least trying to include some estimate for how much the Trump Admin's...
I love 538.
That said, for me at least, it feels frankly disingenuous at this point to make predictions like this, without at least trying to include some estimate for how much the Trump Admin's efforts at electoral fraud will influence the outcome.
This is not a prediction of how the election will go in the US. It is a prediction of how it might go, if the election happened in some fantasy-US where all eligible voters who want to vote actually get to, and then, where all those votes actually get counted.
It is obvious by now that this election will not be fair. I would love to start seeing some best-guess predictions — and fast — on just how unfair it will be.
The model is indirectly based on estimates of who is likely to vote, since polls do that. But election fraud is not something that can be forecast using the kind of model that 538 has, since the...
The model is indirectly based on estimates of who is likely to vote, since polls do that. But election fraud is not something that can be forecast using the kind of model that 538 has, since the model is calibrated based on what happened in previous years. You will have to combine their estimate with your own guesses about what else will happen this year that's new.
You might use a betting market to get an idea of other people's guesses, but betting markets don't have a great track record.
I have to say, the data is displayed beautifully. I hope for people in the us that Biden wins also seeing how republicans just outright deny science. Not just with the corona virus, but also climate change.
And I have to say: almost 50 percent voting for Trump is something I find hard to grasp. I can’t imagine how people can still support this insanity
I know right? All I can talk about is the design of this page. I love that they put in a mascot character to put footnotes on the graphs.
The mascot character is really amusing to me, because a lot of what they're using it for is emphasizing that their forecast is just that, a forecast. It's based on their best models and the best data available, but it's still just a forecast. They said all of those things in their 2016 forecasts too, but it was usually buried in the articles accompanying the fancy graphics.
You have to remember that 50% means 50% of people who actually vote, which is not even a majority of Americans. Still way too many people supporting Trump, but at least it's not half of the whole population.
Please, please, please, please, please make sure you keep in mind this is the page from the 2016 elections and we know how that turned out.
I would also note the article linked in that snapshot: "Its way too soon to count Trump out". It's more that people are realllly bad at probability.
Flip a coin twice. If you get two heads, that's less likely than the chance this model gives Trump of winning.
People see it as like a score, or vote, not as a bernoulli
Yeah they said those things constantly leading up to the 2016 election, and emphasized that the polls were within the margin or error and if they broke toward trump it could flip a bunch of important states. And, lo and behold, that's exactly what happened.
Not sure what point you're trying to make here? If you're suggesting that FiveThirtyEight is inaccurate because Trump won, then it's important to remember that a 28% chance is not a 0% chance. That page never said Trump wouldn't win; just that it was less likely.
To tack onto this, the popular vote wasn't that far off, interestingly enough. In fact, Hillary's popular vote prediction is pretty much spot on.
As for Trump, it seems that they predicted more people would vote 3rd party and steal votes from Trump by doing so. Instead, more predicted 3rd party voters cast their ballot for Trump. I suspect a lot of these were protest/chaos votes personally, but who knows.
Really just shows the power of state delegates either way. They miscalculated Trumps popular vote by only 1.2%, but his delegates by 69.
No? I'm saying projections don't equate to winning an election. People have to go out and vote, and at that, not toss their votes to the third party this time.
I would note the companion article, and this exercpt
https://fivethirtyeight.com/features/its-way-too-soon-to-count-trump-out/
I wish they would acknowledge the potential for election fraud, voter suppression, etc., even if you can't really account for it. But it is nice to see Biden is favored regardless.
They do, just not in this article. See my comment here.
Gotcha, thank you. I hadn't seen anything about that.
I find this wonderfully reassuring. As an international mafia tries to ramrod the Apocalypse through my country’s election system, it’s nice having a cheerful cartoon fox in wonky glasses remind me how they could easily win with popular support. There he is brandishing a jar of red balls and blue balls, like it’s a state lotto draw, or a kindergarten sorting game. It’s not that such representations are unique; it’s how fantastically incongruous they are to the present moment that chills me. Then, perhaps the ability to treat any topic to a serious, measured discussion has been un-educated out of Americans, and is part of the reason we will soon get to live in a lazy-Nazi dystopia.
The point of the page is to talk about probability and statistics. Probability does come from games of chance, so it seems thematically appropriate? It emphasizes uncertainty and a certain kind of unpredictability, that the outcome is unknown and they aren't "calling" it.
And that's good as far as it goes, but there is the problem that the world is not a casino.
I'm willing to forgive this, because it's still better than the fatalism you often see, where we are doomed and so nothing we do matters, or the future depends on huge, impersonal, inevitable forces, and so nothing we do matters.
Mildly offtopic but the cartoon fox doesn't pop up in mobile, even with Chrome's "version for PC" feature.
I love 538.
That said, for me at least, it feels frankly disingenuous at this point to make predictions like this, without at least trying to include some estimate for how much the Trump Admin's efforts at electoral fraud will influence the outcome.
This is not a prediction of how the election will go in the US. It is a prediction of how it might go, if the election happened in some fantasy-US where all eligible voters who want to vote actually get to, and then, where all those votes actually get counted.
It is obvious by now that this election will not be fair. I would love to start seeing some best-guess predictions — and fast — on just how unfair it will be.
The model is indirectly based on estimates of who is likely to vote, since polls do that. But election fraud is not something that can be forecast using the kind of model that 538 has, since the model is calibrated based on what happened in previous years. You will have to combine their estimate with your own guesses about what else will happen this year that's new.
You might use a betting market to get an idea of other people's guesses, but betting markets don't have a great track record.
Do the simulations take into account the election interference from Trump?
No.