Welcome, Guest. Please login or register.
Did you miss your activation email?
June 17, 2019, 02:49:36 pm
News: Please delete your old personal messages.

  Atlas Forum
  Other Elections - Analysis and Discussion
  Gubernatorial/Statewide Elections
  2019 & Odd Year Gubernatorial Election Polls (Moderators: Brittain33, Virginiá)
  VA-Fox: Northam +4
« previous next »
Pages: [1] Print
Author Topic: VA-Fox: Northam +4  (Read 822 times)
heatcharger
YaBB God
*****
Posts: 4,165
Sweden


Political Matrix
E: -1.04, S: -0.24

Show only this user's posts in this thread
« on: September 19, 2017, 05:26:22 pm »

Source.

Northam 42%
Gillespie 38%

Well that polling drought was ended with a hurricane.
Logged
Dom. Pol. Councilor Dwarven Dragon
Wulfric
YaBB God
*****
Posts: 21,175
United States


Political Matrix
E: -1.68, S: 1.22

P P P

Show only this user's posts in this thread
« Reply #1 on: September 19, 2017, 05:59:56 pm »

This makes Quinnipiac even more of a ridiculous outlier. They have to be beyond embarrassed right now. Anyways, obvious Lean D race is obvious.
Logged
Gass3268
Atlas Politician
YaBB God
*****
Posts: 19,176
United States


Show only this user's posts in this thread
« Reply #2 on: September 19, 2017, 06:08:42 pm »

This makes Quinnipiac even more of a ridiculous outlier. They have to be beyond embarrassed right now. Anyways, obvious Lean D race is obvious.

Why would they be embarrassed for a poll that makes statistical sense if the race is currently Northam +4?
Logged
Dom. Pol. Councilor Dwarven Dragon
Wulfric
YaBB God
*****
Posts: 21,175
United States


Political Matrix
E: -1.68, S: 1.22

P P P

Show only this user's posts in this thread
« Reply #3 on: September 19, 2017, 06:11:38 pm »

This makes Quinnipiac even more of a ridiculous outlier. They have to be beyond embarrassed right now. Anyways, obvious Lean D race is obvious.

Why would they be embarrassed for a poll that makes statistical sense if the race is currently Northam +4?

Showing Northam up 10 when all other polls are showing a far closer race is a great sign that Quinnipiac is sh**t. That this poll also shows a close race just emphasizes how dumb Quinnipiac is with their Northam +10 finding.
Logged
Castro
Castro2020
YaBB God
*****
Posts: 10,794
United States


Show only this user's posts in this thread
« Reply #4 on: September 19, 2017, 06:16:34 pm »

This makes Quinnipiac even more of a ridiculous outlier. They have to be beyond embarrassed right now. Anyways, obvious Lean D race is obvious.

Why would they be embarrassed for a poll that makes statistical sense if the race is currently Northam +4?

Showing Northam up 10 when all other polls are showing a far closer race is a great sign that Quinnipiac is sh**t. That this poll also shows a close race just emphasizes how dumb Quinnipiac is with their Northam +10 finding.

That is not at all how polls and MOEs work. You should absolutely expect to see polls showing Northam +10, tied, and Northam +4 if it's approximately a 5 point race.
Logged
MT Treasurer
IndyRep
YaBB God
*****
Posts: 16,812
United States


P P P
Show only this user's posts in this thread
« Reply #5 on: September 19, 2017, 06:17:14 pm »

New Poll: Virginia Governor by Fox News on 2017-09-18

Summary: D: 42%, R: 38%, U: 18%

Poll Source URL: Full Poll Details
Logged
Devout Centrist
Atlas Politician
YaBB God
*****
Posts: 6,511
United States


Political Matrix
E: -99.99, S: -99.99

Show only this user's posts in this thread
« Reply #6 on: September 19, 2017, 08:28:05 pm »

This makes Quinnipiac even more of a ridiculous outlier. They have to be beyond embarrassed right now. Anyways, obvious Lean D race is obvious.

Why would they be embarrassed for a poll that makes statistical sense if the race is currently Northam +4?

Showing Northam up 10 when all other polls are showing a far closer race is a great sign that Quinnipiac is sh**t.
And this logic never fails
Logged
Gustaf
Moderators
YaBB God
*****
Posts: 29,357


Political Matrix
E: 0.39, S: -0.70

P P

Show only this user's posts in this thread
« Reply #7 on: September 20, 2017, 09:05:58 am »

Lol, why would you actively promote herding? What idiocy.
Logged
superbudgie1582
YaBB God
*****
Posts: 2,685


Show only this user's posts in this thread
« Reply #8 on: September 20, 2017, 09:16:01 am »

This makes Quinnipiac even more of a ridiculous outlier. They have to be beyond embarrassed right now. Anyways, obvious Lean D race is obvious.

Why would they be embarrassed for a poll that makes statistical sense if the race is currently Northam +4?

Showing Northam up 10 when all other polls are showing a far closer race is a great sign that Quinnipiac is sh**t. That this poll also shows a close race just emphasizes how dumb Quinnipiac is with their Northam +10 finding.

For someone with 14,000 posts on a polling website, you don't seem to know about MoE.
Logged
swf541
YaBB God
*****
Posts: 1,144


Show only this user's posts in this thread
« Reply #9 on: September 20, 2017, 09:52:08 am »

This makes Quinnipiac even more of a ridiculous outlier. They have to be beyond embarrassed right now. Anyways, obvious Lean D race is obvious.

Why would they be embarrassed for a poll that makes statistical sense if the race is currently Northam +4?


Or even understand how outliers can often (not most of the time) be right.
Showing Northam up 10 when all other polls are showing a far closer race is a great sign that Quinnipiac is sh**t. That this poll also shows a close race just emphasizes how dumb Quinnipiac is with their Northam +10 finding.

For someone with 14,000 posts on a polling website, you don't seem to know about MoE.
Logged
Beet
Moderators
YaBB God
*****
Posts: 24,491


Show only this user's posts in this thread
« Reply #10 on: September 20, 2017, 10:02:50 am »

God, this site is just getting dumber and dumber over time. I hate to defend Wulfric but what passes for sophistication around here is juvenile. It's true in theory, the race could be Northam +4, and with a 3% MoE, if Gillespie's number is at the very bottom end of that range and Northam's error is at the very top end in the Quinnipiac, then the truth is within Quinnipiac's 95% confidence interval. But to say that it "makes statistical sense" is meaningless. There's no such thing as "makes statistical sense." What are the odds that the true margin is 4% AND Quinnipiac poll has Northam at the top of their confidence interval AND Gillespie at the bottom, versus the probability that there is some house effect?
Logged
UncleSam
YaBB God
*****
Posts: 1,322
Show only this user's posts in this thread
« Reply #11 on: September 20, 2017, 11:30:00 am »

If the race is Northam +4 then Quinnipiac would have achieved a p value of p~.05 with their Northam +10 and a MOE of +/- 3 points (though it is worth noting - is this MOE reported for the net top like or for the individual proportions? It's not exactly clear to me they are consistent and very rarely explain where their top line comes from) in the absolute best case scenario. This is a poor result no matter how you slice it.

What's probably worse for Quinnipiac, however, is the inaccuracy of their cross tabs (again. Assuming a Northam +4 result). They showed a gargantuan gender gap of 40 points net which hasn't been replicated anywhere else, and have no age bands or political affiliations.

In other words, Wulfric is more or less right. Quinnipiac looks like a complete joke of a pollster to anyone familiar with polling methodology right now.
Logged
Castro
Castro2020
YaBB God
*****
Posts: 10,794
United States


Show only this user's posts in this thread
« Reply #12 on: September 20, 2017, 11:45:55 am »

Nobody is saying Quinnipiac is without fault, just that it's a very bad idea to dismiss any poll on account of the result when such results are in line with what could be plausible toplines. Be skeptical, but not reflexively dismissive of polls.
Logged
superbudgie1582
YaBB God
*****
Posts: 2,685


Show only this user's posts in this thread
« Reply #13 on: September 20, 2017, 05:43:15 pm »

God, this site is just getting dumber and dumber over time. I hate to defend Wulfric but what passes for sophistication around here is juvenile. It's true in theory, the race could be Northam +4, and with a 3% MoE, if Gillespie's number is at the very bottom end of that range and Northam's error is at the very top end in the Quinnipiac, then the truth is within Quinnipiac's 95% confidence interval. But to say that it "makes statistical sense" is meaningless. There's no such thing as "makes statistical sense." What are the odds that the true margin is 4% AND Quinnipiac poll has Northam at the top of their confidence interval AND Gillespie at the bottom, versus the probability that there is some house effect?

That is completely untrue.
Logged
Figueira
84285
YaBB God
*****
Posts: 9,662


Show only this user's posts in this thread
« Reply #14 on: September 20, 2017, 09:10:12 pm »

Can't we wait until the election happens to decide which polls were and weren't accurate? Qunnipiac IIRC doesn't have a great track record but who knows who's right here?
Logged
Pages: [1] Print 
« previous next »
Jump to:  


Login with username, password and session length
Logout

Terms of Service - DMCA Agent and Policy - Privacy Policy and Cookies

Powered by SMF 1.1.21 | SMF © 2015, Simple Machines

© Dave Leip's Atlas of U.S. Elections, LLC