How Nate Silver Missed Donald Trump

[I spent the winter writing songs about getting better]

Putting up hypotheses based on data and testing them isn't "wrong".

Since polls are generally very unreliable in primaries compared to general elections, trying to put in more data makes sense.

But I guess joining the latest lynch mob to shout in unison is more fun for some people.

Well, he has seemed awfully certain about this, showing little indications of considering arguments that ran counter to his conclusions (and making a lot of dubious arguments himself, such as the idea that Trump as a candidate was more or less equivalent to earlier "anti-establishment candidate" (none of whom shared Trump's level of support over time, his celebrity status, and probably lots of other factors). Now that Silver is starting to backtrack, at the very last moment, I hardly think he should be considered the victim of a "lynch mob".

And on the flipside, when he's right he is largely lauded as a genius, even when his statistical modeling barely differs from a simple polling average.

Those are different points. I agree Silver has seemed reluctant to admit Trump is doing well.

What I take issue with is the constant ridicule of him not just looking at polling data. I find that line of attack pretty silly.

I also think it's generally petty to hate on people because other people laud them as geniuses. I guess the saintly users on Atlas Forum would have refused lucrative contracts with the media out of principle but I harbour no such self-illusions. If people proclaimed me the Messiah for pointing out obvious things I'd cash that check.

It's not that he's not just looking at polling data, it's that his non-polls data this year is garbage. "Endorsement points" is an absurd premise worse than most pbrower stuff.

[I spent the winter writing songs about getting better]

It's not that he's not just looking at polling data, it's that his non-polls data this year is garbage. "Endorsement points" is an absurd premise worse than most pbrower stuff.

I think you might be looking at that the wrong way. If a candidate gets a whole bunch of endorsements, that tells you who the establishment is willing throw their weight behind,  which probably does tend to affect the result I hope we can agree? He's not saying necessarily that endorsements make voters to change their minds. Correlations can (emphasis can) be predictive even if there are not causal links.

And look at all the good that establishment support did for Jeb!...

[I spent the winter writing songs about getting better]

Here's why I can't put any value on the endorsement points model at all:

1-He never did this in the past. He may have tracked endorsements in 2008 but only because it mattered for superdelegates. I don't recall him ever doing so in 2012. And in 2008 Silver was actually pretty skeptical of the media's whenever they did one of those "Hey Politician X from state that's coming up to vote just endorsed Hillary/Obama DOES THIS CHANGE EVERYTHING?" stories.
2-Just look at it. You not only have Jeb! in first and Trump with zero, but Christie, Kasich and HUCKABEE are even beating Cruz. This has about zero relevance to the actual campaign.
3-He's assigning objective numerical values to endorsements like that can be done. It reminds me of how teenage posters like to say things"If *candidate* picks *Governor/Senator* as their running mate, then they will gain X% in that home state and Y% in neighboring state." Yes VP picks matter but not like that. And similarly it's kind of silly to say that the Governor of Idaho or Wyoming's endorsement is worth 10x as much as some influential Tea Party House Republican's, or Charlie Baker and Larry Hogan's too for that matter. I'll note this is the only reason why Christie is so high, he has two Governors. Kasich and Huckabee too benefit from Gubernatorial endorsements. The Governors for Christie are...oh yeah actually Larry Hogan! And Paul LePage. Huckabee has his own Governor of Arkansas. Kasich has the Governor of Alabama. who care.

So I think it's safe to say this model is quite flawed, even if you want to argue that Trump is just an anomaly.