- 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: .
Posted on 11/11/16 at 12:24 pm to DelU249
Posted on 11/11/16 at 12:24 pm to DelU249
quote:Well maybe it's just a big fish small pond phenomenon on that field, but he was again the best polling forecaster by a large margin.
He spent 4 years being treated like a legend and figuratively blown at every turn
quote:I think it's just his personality, but we should be able to seperate that from his competency as a professional when evaluating his professional work.
I don't know if that made him a prick or just exacerbated all of his personality traits that make him one
3
Posted on 11/11/16 at 12:25 pm to RobbBobb
quote:I'm just addressing your lack of understanding, and at times ignorance, regarding polling and forecasting.
Even ole Nate has a CTR project going.
Its the lib go to move, in the Internet age
Posted on 11/11/16 at 12:25 pm to BigPerm30
quote:
I came within 48 points of 270 and spent $500 million less than Hillary.
So you spent $700m, wow
Posted on 11/11/16 at 12:31 pm to buckeye_vol
quote:Statistics show that Bill Gates and I have an average combined net worth of over $40 billion.....
show little understanding of survey theory, and probability/statistics.
Posted on 11/11/16 at 12:42 pm to LSURussian
quote:Sure. It's also true that wealth in not normally disrupted so it should be modeled on something like a Pareto Distribution, plus using a far larger sample than N = 2.
Statistics show that Bill Gates and I have an average combined net worth of over $40 billion.....
This post was edited on 11/11/16 at 12:43 pm
Posted on 11/11/16 at 12:45 pm to More&Les
quote:
So you spent $700m, wow
Did she really spend 1.2 billion?
Posted on 11/11/16 at 12:45 pm to buckeye_vol
quote:
I'm just addressing your lack of understanding, and at times ignorance, regarding polling and forecasting.
OK. lol. I've known for some time that Nate was just wishcasting. "Wishcasting is when the evidence that supports a strong desire is given more weight than it deserves."
Posted on 11/11/16 at 12:46 pm to buckeye_vol
So salty
Are you a Kasich Cuck or Hill femme?
Are you a Kasich Cuck or Hill femme?
Posted on 11/11/16 at 12:47 pm to buckeye_vol
As an experimental and survey design statistician I think that political survey methodology is really hackish...with something so inherently biased as politics there is really no way to believe most polls to a certain degree.
You cant take the politics out of it.
The problem with all these state surveys is they are taking a random sample, not a stratified sample, and they don't know what to do with non-response bias
You cant take the politics out of it.
The problem with all these state surveys is they are taking a random sample, not a stratified sample, and they don't know what to do with non-response bias
This post was edited on 11/11/16 at 12:48 pm
Posted on 11/11/16 at 12:48 pm to DelU249
Posted on 11/11/16 at 12:50 pm to DelU249
Nate was off by like 10 points in Wisconsin and Michigan so he needs to STFU.
Posted on 11/11/16 at 12:57 pm to gaetti15
quote:Of course, especially since the population it's trying to sample is not entirely known (actual voters) rather than a representative sample of the a known population.
As an experimental and survey design statistician I think that political survey methodology is really hackish...with something so inherently biased as politics there is really no way to believe most polls to a certain degree.
quote:Agreed.
The problem with all these state surveys is they are taking a random sample, not a stratified sample, and they don't know what to do with non-response bias
However, the are obviously all flaws of the polling itself, but those flaws don't justify the amount of criticisms that are directed toward 538 since they don't have any involvement in that.
Now Nassim Taleb had had some scathing criticisms of the mathematical foundations of the process; those are valid, although I wish he would avoid making so many personal insults with his criticisms (Sam Harris is another one seems to despise for some reason).
Posted on 11/11/16 at 1:01 pm to DelU249
Did he run the same projection IF Trump had 2 more points?
Posted on 11/11/16 at 1:05 pm to buckeye_vol
I'd like to put he intertwined them, not me 
Posted on 11/11/16 at 1:07 pm to Catman88
quote:
Did he run the same projection IF Trump had 2 more points?
Of course not. Doesn't fit the narrative.
Posted on 11/11/16 at 1:10 pm to Catman88
quote:No, but he would have won two more states (New Hampshire & Minnesota) and 14 electoral college votes.
Did he run the same projection IF Trump had 2 more points?
That's a lot less interesting that 2 points the other way which would have given Hillary Florida, Michigan, Pennsylvania, and Wisconsin for a total of 75 Electoral Votes.
Posted on 11/11/16 at 1:12 pm to Vandyrone
quote:Or maybe the 14 EC vote increase and 28 point swing with the same election outcome is a lot less note worthy than a 75 vote decrease and 150 point swing with a completely different election outcome.
Of course not. Doesn't fit the narrative.
Posted on 11/11/16 at 1:16 pm to buckeye_vol
quote:
I feel the need to defend 538 against these these ridiculous criticisms so can you really blame him? In addition, it's clear that many people still don't understand probability, and how a minor variations continuous metric (votes on a state) can have HUGE ramifications on a binary metric (win-loss) with a multiplicative transformation (win*EC votes in each state).
He didn't predict anything. Like Allan Lichtman said, he's a clerk, he isn't a scientist. What he does is hokum. Even his plus forecast isn't meaningful in any scientific sense.
This post was edited on 11/11/16 at 2:23 pm
Posted on 11/11/16 at 1:32 pm to therick711
quote:Regardless of accuracy, if you predict an outcome, how can you say it's not a prediction?
He didn't predict anything. Like Allan Lichtman said, he's a clerk, he isn't a scientist. What he does is hokum. Even his pluls forecast isn't meaningful in any scientific sense.
And you can criticize the methodology, like anything else, but with the use of statistical modeling using distributions probability distributions based on empirical results of past elections (e.g., modeling trend, a fat tailed distribution) and making two falsifisble predictions (win probability and vote shares) across all 54 different races (50 states + DC + NE And ME CDs+ DC + national totals), it's a perfect example of science.
In fact, given the small sample size for presidential election, there is probably more scientific value in this election than when things end up as expected (I.e., like Black Swan theory).
Posted on 11/11/16 at 1:33 pm to Blizzard of Chizz
quote:
Dio isn't a metal band
What? They were definitely under the heavy metal category back then. I was a teen at that time.
Page 3 of 5
Popular
Back to top
Follow TigerDroppings for LSU Football News