For the first time since 538 published our presidential election forecast for Vice President Kamala Harris and former President Donald Trump, Trump has taken the lead (if a very small one) over Harris. As of 3 p.m. Eastern on Oct. 18, our model gives Trump a 52-in-100 chance of winning the majority of Electoral College votes. The model gives Harris a 48-in-100 chance.
I’ve seen the reports. I highly doubt any of the arm chair statisticians (who have never taken a day of mathematical or proof-level stats) have a clue what they are talking about. The polls’ histories and lean are factored into 538’s averages. They are not new to this.
And how many polls are left leaning? The graph posted a couple of days ago on midwest claims 35% are right leaning, and a correlation with the drop in support for Harris. What it doesn’t say is the proportion of democratic polls, and there really isn’t a correlation over the length of history shown.
Hard to make informed decisions when half the information is hidden. (But arm chair statisticians don’t recognize the issue do they?)
Left or right leaning isn’t the problem. Deliberately sleeping the averages is. I don’t care for Nate Silver, but in his own words: https://www.natesilver.net/p/are-republican-pollsters-flooding
A .5% shift in bullshit data explains all of this right here. Polling is useless anymore anyway, but when your Trump’s little circus of assclowns who want to form a pretext of something like a stolen election, it sure as fuck serves a purpose.
Edit: and then you have shit like this: https://www.rawstory.com/donald-trump-2669441230/ https://www.reuters.com/world/us/large-bets-election-prediction-market-are-overseas-source-says-2024-10-18/
Not suspicious or coordinated at all.
I’ll give you your references. Will have to read later.