- My Forums
- Tiger Rant
- LSU Recruiting
- SEC Rant
- Saints Talk
- Pelicans Talk
- More Sports Board
- Fantasy Sports
- Golf Board
- Soccer Board
- O-T Lounge
- Tech Board
- Home/Garden Board
- Outdoor Board
- Health/Fitness Board
- Movie/TV Board
- Book Board
- Music Board
- Political Talk
- Money Talk
- Fark Board
- Gaming Board
- Travel Board
- Food/Drink Board
- Ticket Exchange
- TD Help Board
Customize My Forums- View All Forums
- Show Left Links
- Topic Sort Options
- Trending Topics
- Recent Topics
- Active Topics
Started By
Message
re: Excellent Analysis of the Fake News process, Re: NYT, Perry
Posted on 1/19/17 at 10:05 pm to RobbBobb
Posted on 1/19/17 at 10:05 pm to RobbBobb
quote:Look maybe you had some exclusive possession to important polling data (I doubt it), but you're arguments regarding this have shown a complete lack of understanding of survey theory and methodology and statistical analysis.
I told you from Day 1, that the polls being touted were not the ones I had in my possession.
This is highlighted by you citing the LaTimes poll as accurate, WHEN in fact they were one of the least accurate national polls.
Polling has some issues, and it's not immune to systematic biases (in the midwest in this election), but they weren't too inaccurate elsewhere.
quote:Charges that they were fake are unsupported. In fact, I contend that they their biggest misses were due to a superior campaign strategy by Trump.
But keep believing that the charge is unsupported
Regardless, hackish pundits aside, any person familiar with polling and data would have known that Trump winning was not some outrageous outcome. Despite it being less likely, the probability was still well within an area of a reasonable chance of victory.
In other words, a 1/3 or 1/4 chance happens a lot. I mean flipping two heads or two tails in a row (assuming a fiar coin) is less likely than a Trump victory (using 538's model with the strongest methodology). Yet, it would be illogical to argue that the odds are wrong when that happens.
1
Posted on 1/19/17 at 10:09 pm to USMCTiger03
I've work on undisclosed projects for DOE surrounding nuclear security and I think it's far to say that most Americans don't realize DOE's role in protecting the nuclear arsenal.
Posted on 1/19/17 at 10:13 pm to buckeye_vol
So your argument boils down to we cannot prove its fake because nobody was there to witness it, tape it, and have a signed confession that the event did not happen as represented?
Yeesh. You're charging Normandy Beach with your dick as your only weapon.
Yeesh. You're charging Normandy Beach with your dick as your only weapon.
Posted on 1/19/17 at 10:19 pm to Jcorye1
quote:No my argument is when you criticize a conclusion that is not supported by the evidence, don't resort to the opposite conclusion in absolute when it lacks the evidence.
So your argument boils down to we cannot prove its fake because nobody was there to witness it, tape it, and have a signed confession that the event did not happen as represented?
It's like global warming. I agree that the absolute conclusions (causality, warming predictions, only extremely negative future outcomes, etc.) that are presented are not supported by the evidence. Yet, it's illogical to then conclude the opposite (not warming, no impact, etc.) must be true.
Posted on 1/19/17 at 10:46 pm to CptBengal
quote:
how long until they are just printing straight up lies?
Y
(Yeah, I know, not quite the same.)
Posted on 1/19/17 at 11:14 pm to buckeye_vol
quote:
This is why the "fake news" concept was stupid. It may be poor conclusions based on speculation, but "fake news" to me is something that is completely made up (tabloids, Infowars). In my opinion, there is a major distinction between subpar journalism and fake news.
What you are assuming is that the work product is the result of laziness or bias.
Coming to a conclusion that is not supported by sufficient evidence because of your laziness or bias is poor journalism. That is not the same thing as fabricating a narrative you are determined to produce when you know it is not supported by evidence. That's fake news.
Posted on 1/20/17 at 2:12 am to buckeye_vol
quote:
Yet, it would be illogical to argue that the odds are wrong when that happens.
I know you like to tout your knowledge in this area, but you need to focus on what was being sold, as to what the real numbers actually were
Pollsters were not doing due diligence. They were using estimations based on previous polling cycles. I would be willing to bet that they were not making the calls they claimed, but relying on modeling, while intentionally feeding in false data. And they didn't care if they were off. This election was not going to determine whether they ever were hired again. Most simply didn't care, nor want, to be accurate (Hillary with a 14% lead within days of the election. C'mon?)
One example, Penn: Final polls had Hillary 48%, Trump 42%
Penn:
6,115,402 total votes cast for president
2,970,863 actual votes for Trump
2,568,469 votes for Trump if 42% was accurate
402,394 difference (their poll missed more voters in one state than cast votes in AK, DEL, HI, MT, ND, RI, SD, or VT)
They were off by 2-4x the MOE, in important states. Which means it really wasn't the MOE, then was it? Which is all the proof you need that the poll was flawed
in Penn, Ohio, Wisc, Mich, Maine, Iowa, N. Car, Fla. You cannot be that far outside the MOE in that many states, by accident, or coincidence. Its an intentional skew. And if its not, then polling should be shut down, because its a worthless enterprise
Except those same pollsters nailed Hillarys vote in each of those states. Hillary was only off 0.1 to 1.4% in polling from her actual votes. Hmmmmmm
Page 3 of 3
Popular
Back to top
Follow TigerDroppings for LSU Football News