Princeton Election Consortium - Critique of Nate Silver as Being Too Cautious About Obama's Chances!

[TomCADem]

Sam Wang predicts an electoral total of:

Obama: 305
Romney: 233

And he predicts the probability of Obama re-election: Random Drift 93%, Bayesian Prediction 98%. The key difference is that Sam Wang argues that Nate introduces other factors that Sam does not believe have any predictive relevance:

http://election.princeton.edu/2012/10/29/nerds-under-attack/#more-8151

I have my own technical beefs with FiveThirtyEight (for example, see here, here, and here). I believe Silver doesn’t extract all the information and tends to add unnecessary factors, which leads to blurry probabilities and poor time resolution. However, his intuitions about the data are excellent and he is very concerned with getting things right. For purposes of popular consumption, he is a fine and honest nerd.

So, from a purely nerdy point of view, Sam argues that Nate is playing it too safe.

[geek tragedy]

1. I tend to side with Silver. Krugman has

noted that Wang possibly understates correlations--which seems a valid critique.

[tbennett76]

2. How do we get to 305?

Not mapped that one out

[muriel_volestrangler]

8. I think it's the median EC result, not the most likely

Though it is possible - if both sides win the ones they're expected to, with Obama taking Ohio (though I think Silver is correct in saying that's not really a toss-up any more), Nevada, Colorado, Iowa and North Carolina, while Romney takes Florida and Virginia.

That's not a particularly likely combination, though - it's unlikely Obama would get NC, but Romney VA - as Wang's histogram for today shows:



As of 30 Oct 8pm EDT, that shows 303 the 2nd most likely (as for the 305 scenario, but Obama gets VA, and Romney NC), and 290 the most likely (Romney takes both VA and NC).

[gravity]

3. I rather trust the guy who is being more cautious

There are still many factors that can affect the election that aren't picked up in polls.

[Third Doctor]

4. I prefer the cautious approach.

[DemocratSinceBirth]

5. I Wish I Had Professor Wang's Curriculum Vitae

http://synapse.princeton.edu/SWang-CV-current.pdf

And if you read Professor Wang's site he says Silver is one of the good guys.

[defacto7]

6. Caution is good! Caution is reasonable!

I would like to think Wang is right, but then I am a bit prejudiced about the election.

[TomCADem]

7. I Agree, But It Also Suggests The RW Attack On Nate If Full of S*&t

If anything, Nate is being too cautious.