• Quick note - the problem with Youtube videos not embedding on the forum appears to have been fixed, thanks to ZiprHead. If you do still see problems let me know.

Did Nate Silver nail it or what?

Puppycow

Penultimate Amazing
Puppycow
Joined
Jan 9, 2003
Messages
32,083
Location
Yokohama, Japan
There was a lot of hating and criticizing of Nate Silver of the 538 blog by Republican critics before the election, but looking at the results and comparing them to his forecast, I don't see how he could have been any righter than he was.

This funky thing shows the probability of Obama getting a certain number of electoral votes according to his forecast:
picture.php


(Well, you can see it a lot better on his blog where it's in context)

Anyway, those two largest spikes are for 303 and 332 electoral votes, respectively. The one for 332 is the largest one and 303 is the second largest. IOW, if Obama's lead in FL holds up, he will have picked every single state correctly. There's more counting to do before we know what the final popular vote will be, but that too looks like it will be very close to Nate's forecast.

So I think he's been vindicated in a very big way.
 
Puppycow said:
There was a lot of hating and criticizing of Nate Silver of the 538 blog by Republican critics before the election, but looking at the results and comparing them to his forecast, I don't see how he could have been any righter than he was.

This funky thing shows the probability of Obama getting a certain number of electoral votes according to his forecast:
[qimg]http://www.internationalskeptics.com/forums/picture.php?albumid=200&pictureid=6892[/qimg]

(Well, you can see it a lot better on his blog where it's in context)

Anyway, those two largest spikes are for 303 and 332 electoral votes, respectively. The one for 332 is the largest one and 303 is the second largest. IOW, if Obama's lead in FL holds up, he will have picked every single state correctly. There's more counting to do before we know what the final popular vote will be, but that too looks like it will be very close to Nate's forecast.

So I think he's been vindicated in a very big way.
Click to expand...

He did a pretty good job, though just averaging in-trade with polls would have worked as well.
 
I actually have his book and was waiting to see how the election turned out before I started reading it. I would have passed on it if he'd been way off hehe, but since he ended up being so accurate I will give it a go.

I think there's a lot of doubletalk and gamesmanship in the days leading up to the election in an effort to get the voters out. I'm sure a lot of the doubt and vitriol cast his way was just for show.
 
Puppycow said:
There was a lot of hating and criticizing of Nate Silver of the 538 blog by Republican critics before the election, but looking at the results and comparing them to his forecast, I don't see how he could have been any righter than he was.

This funky thing shows the probability of Obama getting a certain number of electoral votes according to his forecast:
[qimg]http://www.internationalskeptics.com/forums/picture.php?albumid=200&pictureid=6892[/qimg]

(Well, you can see it a lot better on his blog where it's in context)

Anyway, those two largest spikes are for 303 and 332 electoral votes, respectively. The one for 332 is the largest one and 303 is the second largest. IOW, if Obama's lead in FL holds up, he will have picked every single state correctly. There's more counting to do before we know what the final popular vote will be, but that too looks like it will be very close to Nate's forecast.

So I think he's been vindicated in a very big way.
Click to expand...
For clarity, I think it would be better to phrase is as "the biggest spike is for 332, the second biggest spike is 303".

If the 3% of votes not yet counted in Florida don't change who appears to be winning there, the number of electoral college votes in the November 6 forecast seems to have always been leaning conservative. And still leaning conservative most of the time if Romney gets Florida. Beautiful. :)
 
Last edited:
Alan said:
If the 3% of votes not yet counted in Florida don't change who appears to be winning there, the number of electoral college votes in the November 6 forecast seems to have always been leaning conservative.
Click to expand...

I don't know about that. That number was like the average of all probabilities on the distribution, not the most likely single outcome. The most likely single outcome in his model has been 332 for months, and 303 was the second most likely.
 
Anyone else thinking of the ending of Dirty Dancing? When the hotel owner finally gets it that this new dancing has got all his customers up dancing in the aisles and he turns to the band leader and asks, "You got sheet music for this?"

I can hear the conversation from the Bush compound on conference call to Karl Rove and Michael Barone, now....

Barbara Bush: "So, this poll averaging with statistical weighting and historical analysis.... You boys got algorithms for this?"

Next election cycle, I'm betting there will be lots of (what did that a**hat at Unskewed Polls call him...) "skinny effeminate geeks" working with a lot of polling organizations.
 
Puppycow said:
I don't know about that. That number was like the average of all probabilities on the distribution, not the most likely single outcome. The most likely single outcome in his model has been 332 for months, and 303 was the second most likely.
Click to expand...
Yes, but people critical of Silver saying that Romney would be doing better tended to discuss the numbers at the top. But you are correct.

I had been thinking that it would be good if the one showing the frequency of various eventualities was higher up on the page so that it got more attention.
 
ronin688 said:
I actually have his book and was waiting to see how the election turned out before I started reading it. I would have passed on it if he'd been way off hehe, but since he ended up being so accurate I will give it a go.

I think there's a lot of doubletalk and gamesmanship in the days leading up to the election in an effort to get the voters out. I'm sure a lot of the doubt and vitriol cast his way was just for show.
Click to expand...

Ya, a lot of it was from talking heads of both colors going at him because he is doing to them what he helped do to hack baseball writers. This time around, the right wing mouth pieces were just really mad because they didn't like his results.

He's helping to cut through the pretentious crap and pointless drama to give us actual quanitifable results.
 
I'm curious how this would have played out if the predictions and results were reversed; meaning if Silver had been predicting a Romney victory this whole time and had been proven correct.

Leading up to the election, Silver made several appearances on left-leaning shows like Realtime with Bill Maher, The Rachel Maddow Show, and The Colbert Report. Would he have been asked to appear if he had been projecting a 80% likelihood of a Romney victory?

I think Rachel would have had him on regardless, given that she's such an unabashed geek. Colbert too, probably (it would have played to his schtick). Probably not Maher, though. And I can almost hear Ed Schultz's bloviating ("I don't care about your numbers and fancy computers, I know the American people, and blah blah blah...").
 
There was another guy who totally nailed it, but I think he was using standard 'guesswork' and gut feeling rather than any kind of statistical voodoo.
 
I think Colbert and Stewart would still have him on. Maddow I don't watch or follow so Ican't say.

Bill Maher only likes science-y thigns when they support far left causes.

I want to know how Rush and Dean Chambers would have handled it.
 
You must log in or register to reply here.

Back
Top Bottom