More to the point, I’m arguing against the perception that Trump is running a losing campaign deliberately in order to set himself up for a media empire.
This beef with the entire fucking state of Nevada would hurt his media empire.
And I still feel a tiny little frisson of wonder from that sentence.
Everything Trump says is right. If it’s not right, he simply doesn’t say it anymore, and then nobody can ever say he said it. And if you do say he’s wrong, well, then, you’re being unfair to him.
Besides, Nevada wouldn’t let him open a casino in Vegas, so they’re obviously haters who’ve been being not very nice to him, and everyone agrees that they’re being unfair to him, because he says they are, and everything he says is right.
Sure, and it’s worth acknowledging that often simpler models can perform better than more complex ones that try to “take everything into account”, since they’re less susceptible to overfitting.
What’s interesting is their previous two endorsements. The first was Abraham Lincoln based on his anti-slavery position, but the second, which strangely parallels this election, was in 1964 where another reasonably successful Democrat (LBJ) was running against another candidate The Atlantic felt was actively dangerous (Goldwater).
The bit about fantasy baseball is because Silver first came to prominence as the stats cruncher for The Baseball Prospectus. There, he and others took the beginnings of Sabermetrics - the application of statistical analysis to baseball which Bill James started - and applied rigor to the math.
538 updated again, and now Arizona, which had been various shades of red, is essentially 50-50 in the Now-Cast and shaded the very lightest of light pink. My girl is going to college there, and has a chance to cast a vote that really counts in her first Presidential election!
It’s important to note that the disagreements between Wang’s and Silver’s models are, in the broad scheme of political projections, REALLY minor. You’ve got two honest to God statisticians who use a logical, fact-based approach based on mathematics. They have some methodological differences but both arrive at pretty much the same conclusions and keep getting it right. The majority of other “forecasts” are either based on something other than math, or apply math to something other than polls, or “unskew” the polls, or are based on some other variety of unicorns, hopes and dreams.
I do find this comment in the Wikipedia weird, though:
[QUPOTE=Wiki article on Sam Wang]His statistical analysis in 2012 correctly predicted the presidential vote outcome in 49 of 50 states and even the popular vote outcome of Barack Obama’s 51.1% to Mitt Romney’s 48.9%.
[/QUOTE]
Uh… that was not the popular vote outcome, though. It ended up being 51.1 to 47.2. Wang did predict Obama would win 51.1-48.9 if you just counted Obama and Romney, so he was a tiny bit off there - certainly not by a lot.
Holy cow - on the Nowcast Trump has dropped below 15%, and even in the polls-plus (the most favourable forecast for him) he’s just breaking 25%. Roll on the next debate…
In the Polls Only, Colorado just moved above 80% for Hillary, which on my 270towin board, moves CO out of “swing” territory and solid blue. And puts Hillary at 269 (relatively) safe EVs. And with AZ moving officially into “swing” territory, momentum is not on Don’s side right now.
Course, I won’t believe anything until the votes are all actually cast.