2016 Postmortem

andym

(6,048 posts)

12. His arguments for his models having higher uncertainty do seem plausible

Sat Nov 5, 2016, 04:05 PM

Nov 2016

http://fivethirtyeight.com/features/election-update-why-our-model-is-more-bullish-than-others-on-trump/

Here are his assumptions:
Assumption No. 1: The high number of undecided and third-party voters indicates greater uncertainty.
Assumption No. 2: The FiveThirtyEight model is calibrated based on general elections since 1972.
Assumption No. 3: The FiveThirtyEight model uses a t-distribution with “fat tails,” which gives a greater likelihood of rare events.
Assumption No. 4: State outcomes are highly correlated with one another, so polling errors in one state are likely to be replicated in other, similar states.

He uses polling data going back to 1972-- not sure that is a great idea, since the voting population has changed so much. But he claims that is useful.

Linking similar states is useful, but tricky on how states are grouped.

The good news from all of this is that Hillary is likely going to win, but it could be close, so GOTV and we CAN make a difference.

Edit history

Please sign in to view edit histories.

Recommendations

0 members have recommended this reply (displayed in chronological order):

58 replies

= new reply since forum marked as read

Highlight:

Nate silver is rage tweeting right now because people are calling [View all] SDANation Nov 2016 OP

This message was self-deleted by its author NightWatcher Nov 2016 #1

So now he is dropping F-bombs on Twitter. Must be taking advice from Trumpenstein. RBInMaine Nov 2016 #3

You didn't have that reaction when he said, Ms. Toad Nov 2016 #27

If you gotta cuss, you lost. If you ask me. nt SusanCalvin Nov 2016 #56

He's frustrated I think sharp_stick Nov 2016 #5

Nate is getting a little more abuse than he should Cosmocat Nov 2016 #6

Well if you read Ryan Grim's original article, it does bring across some very good points MyNameIsKhan Nov 2016 #8

I was ready to give Nate the benefit of the doubt... Chichiri Nov 2016 #10

His defensiveness maybe because his model never anticipated what is happening: BlueStreak Nov 2016 #19

I would get off his case he may be right nobody knows until the actual votes are counted. doc03 Nov 2016 #11

In your dreams it is a "tossup." duffyduff Nov 2016 #18

Well, that's the thing. Nate's been right and has come to think his rightness means.... Moonwalk Nov 2016 #35

And, he WAS completely wrong about Trump in the primaries, so his skin is thin now. manicraven Nov 2016 #53

His arguments for his models having higher uncertainty do seem plausible andym Nov 2016 #12

You can't base anything on previous elections. That is why he will likely be seen as full of it. duffyduff Nov 2016 #22

This bothers me, "He uses polling data going back to 1972." manicraven Nov 2016 #57

He Has Been Attacked Before The Great Escape Nov 2016 #13

Makes you wonder why. Buzz Clik Nov 2016 #33

Eh, leave the man alone Dem2 Nov 2016 #15

Not attacking he man personally, but his results... Those are questionable. manicraven Nov 2016 #59

I've been posting some of the bizarre illogical adjustments for several days now Dem2 Nov 2016 #60

He's angry because they are attacking his pride and joy: his model. andym Nov 2016 #16

The problem is he nudged the numbers for no reason. Buzz Clik Nov 2016 #34

The article doesn't say that he changed the numbers for "no reason" andym Nov 2016 #38

He trolls Sam wang? SDANation Nov 2016 #20

I think it is reflective of the general jitteriness that any American with a firing synapse has to hlthe2b Nov 2016 #21

I generally trust Nate ... LenaBaby61 Nov 2016 #24

His prediction numbers have actually helped democrats panic enough to get out bushisanidiot Nov 2016 #25

Some folks drive me nuts. Adrahil Nov 2016 #26

That really is not the point. Buzz Clik Nov 2016 #32

According to Harry Enten.... Adrahil Nov 2016 #39

Wang and Silver were close about a week ago. No longer. Buzz Clik Nov 2016 #41

Far too many people lose it when they read or hear anything that doesn't conform to their politics tymorial Nov 2016 #37

I agree. Adrahil Nov 2016 #40

538 deserves the criticism. Buzz Clik Nov 2016 #30

Silver's explanation as to why 538 is more bullish toward Trump... Garrett78 Nov 2016 #45

Silver is facing competition for the first time. geek tragedy Nov 2016 #43

The guy is in meltdown mode... nt. Blue Idaho Nov 2016 #46

Talking about kneeing people in the balls? budkin Nov 2016 #49

He has Clinton at 65% Capt. Obvious Nov 2016 #50

Yeah guess you're right. budkin Nov 2016 #55

He's getting grief because his results vastly differ from 4 other highly regarded pollsters. AgadorSparticus Nov 2016 #52

I agree... manicraven Nov 2016 #62

Well, he needs to knock it off. SusanCalvin Nov 2016 #54

The Cubs knocked him off his stride kurt_cagle Nov 2016 #61

If he projected that HRC would lose by a landslide I'd be all good if it were an objective analysis. Buzz Clik Nov 2016 #65

This personage, and whatever he does treestar Nov 2016 #71

Dec 1969 #