I assume you are posting this as an “lol America”.
Can’t read it
Here’s a gift link:
https://www.nytimes.com/2024/06/05/upshot/polling-trump-conviction-voters.html?unlocked_article_code=1.xk0.GKz0.Nhah0W_r3iGx&smid=url-share
2 Likes
But in a close election, losing 7 percent of your supporters can be decisive. In recent polls, Mr. Biden either leads or is within two points of Mr. Trump in states and districts worth the 270 electoral votes required to win the presidency. A potentially crucial sliver of Mr. Trump’s former supporters — 3 percent — now told us they’ll back Mr. Biden, while another 4 percent say they’re now undecided.
But the slight movement overall toward Mr. Biden is broadly in line with other recent surveys. In a Reuters/Ipsos survey taken immediately after the verdict, 10 percent of Republicans said Mr. Trump’s conviction made them less likely to support him in the fall.
Another recontacting study by Echelon Insights, a Republican firm, found Mr. Biden gaining two points compared with its previous survey.
That’s nice. Real nice.
Registered voter poll also. But still, FL is just an abusive ex trying with the “remember the good times.” (There were no good times.)
1 Like
Spoiler alert, they all came home to daddy trump.
3 Likes
There is not a big enough lol America on earth
This was pretty great, worth the watch!
These simulations are stupid.
It’s a high variance election, but not by that much.
Results like Biden 530-8 (where I guess he wins everything except WY, WV and NE-03) and Trump 492-46 (I can’t even get the math to work out for that one) are not 1 in 1000 events. They aren’t even 1 in a million events.
1 Like
I think I got it with Biden taking HI/DC/VT/MA/MD/DE/CT/RI and ME-01 with Trump winning everything else, including California and New York.
Just completely stupid.
And if you scroll down, it gives Trump a 3% chance of winning New York, a state he lost by over 20 points twice.
I wish these 538 idiots would be forced to take bets when they release obvious garbage probability forecasts like this. They’d go broke.
2 Likes
It makes more sense when you consider their motivation isn’t to accurately or successfully forecast the election. Their motivation is to drive engagement to their platform and increase advertising revenue.
1 Like
What do you specifically mean when you say it’s wrong ?
Are you saying you would give me 10,000:1 on one of the 1:1000 long tail events?
I wouldn’t, because I either can’t pay out or the gain is not worth it.
I am saying that I don’t believe that the 538 modelers actually believe their stuff. They would not offer $1 for every $99 bet that Biden loses New York, even though that is theoretically 3x better odds than their model says.
Think of what it would take for Trump to win California and New York, or for Biden to sweep the Dakotas, Idaho and Utah. I’m much more likely to roll a yahtzee.
1 Like
Don’t those long tail events include things like death?
2 Likes
I don’t think so? That would be tough to model (because there’s little if any polling data on that), and they specifically say “Trump” and “Biden”, not “R” and “D”.
But even if Biden drops dead and Kamala is the nominee, she’s not losing California. (She would, however, very likely lose a lot of the close states Biden won in 2020.)
1 Like
I’m not enough of an expert to fully understand these models but it seems reasonable that the long tail events are an artifact of small sample size. There have only been 59 elections ever and it looks like the model only includes those since 1920 so a sample size of like 20. Couldn’t this be a better explanation than 538 just making up data?
Is there really a difference between coming to conclusions off a ludicrous sample size and making up data?
2 Likes