MURDER, POLITICS, AND THE END OF THE JAZZ AGE
by Michael Wolraich
Order today at Barnes & Noble / Amazon / Books-A-Million / Bookshop
MURDER, POLITICS, AND THE END OF THE JAZZ AGE by Michael Wolraich Order today at Barnes & Noble / Amazon / Books-A-Million / Bookshop |
The old saw "statistics lie, and liars statistic..." wait, that's not it, "fool me once....won't get fooled again"...mmm, overused and off kilter..... maybe.... "It's just 3dB error", e.g. off by half - that's it.
Polls are popular - search hard enough, you're bound to find one that supports your cause. And if not, you can always complain about accuracy, number of people, leading questions and push polls, land lines vs. mobile, etc. But whether pro or con, polls typically sample 1/1000th or less of the final population. I.e. "a spit in the ocean" in modern parlance.
When they say "within a sampling error of plus or minus 5 points", I don't think that means what you think it means - I believe it's "our process is 95% repeatable", and sure as hell don't mean "accurate to within 5%", judging from deviances of 20% or more between polls
Let's look at the polls. 10 days before South Carolina, so-called respectable CNN/ORC had Clinton up by 18%. The RealClearPolitics average was 27.5%. Hillary won by 47.5%. The *average* of polls taken over the prior two weeks was *twenty percentage points off*. It would have been much worse, but Clemson published a lone poll predicting a 50 point spread, only slightly high.
The Virginia GOP polls had Trump up by a spread of 14.5% - he won by only 2.8%, about 12% difference.
So I'm supposed to make a decision based on hypothetical polls 8 months in advance, when they have trouble with elections 2 weeks away?
Worse, since Gallup and Harris and other big names dropped out of the election polling business, it's gotten more and more inaccurate, with fewer extensive polls to go by. Everyone's copying 538 by averaging now, though not necessarily weighting them, but there's a great acceptance of "no matter how inaccurate we are, they'll forgive us and forget after the actual election". So little drive-by hit pollsters can flourish among the more dedicated and believable.
More I have trouble with many people preaching "electability" - they typically don't know what they're talking about, or at least their arguments point in the wrong direction. Even if the mood of 1000 people reflected the vote of 150 million voters 8 months later, *THAT'S NOT HOW US ELECTIONS WORK*.
Obama beat McCain by 7.4% of the vote. He won the electoral vote by over 2 to 1. That latter figure is what matters. Al Gore won the popular vote over George W. Bush. But the election tilted to Bush not over a dispute about raw numbers, but electoral votes in one particular state.
In 2008, only 14 states were within 10% difference, only 6 states were within 5%.
In 2012, Obama won with 3.8% vote total, but a 3:2 electoral vote advantage: 62% to 38%, 24% difference. Only 3 states were within 5%, 17 within 10%.
There's a good chance that in states with more than 10% difference they could put a comatose pedophile with a skin disease and he'd still win. The only thing that matters are likely 10 to maybe 17 swing states, max 1/3 of the nation. Realistically, much less.
As for "likely voters" in the electability argument, I don't get it - Bernie isn't able to convince more conservative & moderate Democratic southern voters, but I should be happy that he "gets 25% of the republican vote in his state and attracts the largest share of independents" - are those people I want to attract? I thought they were tea baggers and survivalists and racial supremacists and other kinds of cysts. In that case, why not put Trump on our ballot too, so we can win at all costs - yay!!!
Here's Obama, looking cooler than ever (think he's taking on Bill Cosby's standup routine since he won't be needing it anymore), laying out the difference between Democratic wishes & achievements and Republican delusions. It'd be nice if we tried to keep that divide while unifying the party. We're supposed to be the reality-based side of the aisle - whether global warming, healthcare, gun control, foreign policy or election possibilities. Fight hard, dream hard, but keep your feet on the ground - don't succumb to the Madness of Crowds nor the "Science of Big Impressive Sounding Numbers".
Comments
I hereby render unto Peracles the Dayly Line of the Day Award for this here Dagblog Site, given to all of him from all of me for this gem:
HERE'S OBAMA LOOKING COOLER THAN EVER...
hahahahahahaah
You know Peracles, I still think that Trump got into this mess all because Obama made fun of him. hahahahah
DID WE FAKE THE MOON LANDING?
This has nothing to do with nothing. I just recall the '80's when I was a n active contributor to this fine society.
hahahahah
by Richard Day on Thu, 03/03/2016 - 10:38am
Yes, we faked the moon landing. Even now people get weirded out when astronauts say they heard weird music. How could they have handled the truth?
by PeraclesPlease on Thu, 03/03/2016 - 11:49am
Another part of the forest.
Now comes Nate Silver who tells us that "Trump has a fairly high floor , about 30% ". He has these people who show up in exit polls saying they had "decided months ago". And they'll probably stick with him.. But, says Nate, when Trump starts earning higher numbers those supporters "will be relatively new to him " folks who used to be listed as supporters of other candidates. And, still according to Nate, there's a rule in polling called LIFO-last in first out. Which as a one time accountant I can confirm is an accounting "convention" (don't ask).
But these "newbies" will in fact be "new". They're the "last in". And could be the first to be seduced "out"..
Sounds as if it might be encouraging. But....
While that may be the experience with Trump's "floor" within the closed world of Republican candidates is that same phenomenon dependable when the question is Trump vs the democratic candidates. Logic might say yes or no but I'm not interested in logic, I'm interested in history.
Sooner or later I'll find Nate's position on that and I'll report back here.
I'm sure you'll be holding your breath.
by Flavius on Fri, 03/04/2016 - 1:01am
Language gets to me more and more as I find myself closer to death. hahahahah
High floor. I dunno.
Low ceiling?
Nate was sooooooooo much fun in '08 and '12?
A high floor should only be attributed to the posh apartment at the top?
Anyway, I am holding my breath for you Flavius.
by Richard Day on Fri, 03/04/2016 - 1:33am