HomeLatest ThreadsGreatest ThreadsForums & GroupsMy SubscriptionsMy Posts
DU Home » Latest Threads » Forums & Groups » Main » General Discussion (Forum) » Nate Silver's best tweet ...

Thu Dec 6, 2018, 12:18 AM

Nate Silver's best tweet in 2018--changes call in NC 9th

We're changing our rating on this race to Lean Prison.



5 replies, 2227 views

Reply to this thread

Back to top Alert abuse

Always highlight: 10 newest replies | Replies posted after I mark a forum
Replies to this discussion thread
Arrow 5 replies Author Time Post
Reply Nate Silver's best tweet in 2018--changes call in NC 9th (Original post)
Julian Englis Dec 6 OP
Gothmog Dec 6 #1
Farmer-Rick Dec 6 #2
Awsi Dooger Dec 6 #3
Canoe52 Dec 6 #4
peggysue2 Dec 6 #5

Response to Julian Englis (Original post)

Thu Dec 6, 2018, 12:38 AM

1. Yeah for Nate

Reply to this post

Back to top Alert abuse Link here Permalink


Response to Julian Englis (Original post)

Thu Dec 6, 2018, 12:59 AM

2. You would think his statistical analysis would have picked up on this

Every time there has been a claim of election fraud, Nate has come out saying there was no fraud and his stats didn't indicate fraud.

So, now he changes his call but he never saw it coming.

Reply to this post

Back to top Alert abuse Link here Permalink


Response to Farmer-Rick (Reply #2)

Thu Dec 6, 2018, 01:36 AM

3. I hope you aren't serious

He is "changing his call" in jest.

Nate realizes it is fraud but he isn't going to change his methodology to run around looking for cynical fraud behind every pole and every poll.

Nate Silver is a compiler who does a magnificent job of not only weighting polls but also applying logical situational adjustments to them.

House races like NC-9 are not polled frequently and the polling is not reliable in House districts to begin with. There is zero chance Nate is going to focus on each district as opposed to providing a logical overview of the national environment and how it likely attaches to certain types of districts. 538 did that early this cycle and properly evaluated the suburban districts as most ripe for the picking. Nate Silver also emphasized that he thought Democratic pickup opportunities were overrated in the South but that substantial gains could be made in the Midwest and especially in California.

The generic ballot is by far the best way to evaluate the national House terrain and likely net change. Nate Silver and Co. did that brilliantly this year, projecting the generic ballot gap and how many seats Democrats figured to gain at each range...whether it was +6 or +7 or +8, etc. Their overall final estimate of high 30s was almost perfect.

I love the notion that a polling compiler is supposed to "pick up" on a swing of perhaps 1000 or a few thousand votes in a House district. That is not the real world. If that result had been 15,000 votes different either way it wouldn't have triggered anything. It is called normal distribution, just like Donald Trump defeating Hillary Clinton as a mild underdog. Anyone with a background in results and probability understands that 70/30 is a mild underdog.

Now, when it comes to bizarre variance in a specific county, then Silver and his employees can make an estimate of how likely that would be, under normal distribution and minus manipulation.

Reply to this post

Back to top Alert abuse Link here Permalink


Response to Julian Englis (Original post)

Thu Dec 6, 2018, 01:38 AM

4. Nate picks them better than anybody else, probably called this one correct also.

Reply to this post

Back to top Alert abuse Link here Permalink


Response to Julian Englis (Original post)

Thu Dec 6, 2018, 01:48 AM

5. LOL!!

Lean Prison. Score that one for Nate Silver.

Reply to this post

Back to top Alert abuse Link here Permalink

Reply to this thread