Overtime Politics Thread (WARNING: Possible fraud)

[ProgressiveCanadian]

Are we really still entering Overtime polls? ugh.

Yes troll.

[I spent the winter writing songs about getting better]

Are we really still entering Overtime polls? ugh.

Yes troll.

OMG YPU ACTUALLY SERIOUSLY BELIEVE THESE ARE REAL POLLS THAT THESE ARE REAL POLLS WHAT THE F[INKS]

[PeteB]

A poll  by Overtime Politics done in VA from the 12th to 14th Feb. is showing interesting results.

Donald Trump – 33%
Marco Rubio – 19%
John Kasich – 14%
Ted Cruz – 12%
Jeb Bush – 8%
Ben Carson – 8%
Unknown – 6%

Not only does Trump lead by 14% but Cruz has dropped to 4th and Jeb Bush is at Ben Carson's levels, at the bottom.  Kasich, who previously had 1% received a boost of 13% and is challenging Rubio for second.

http://overtimepolitics.com/pollingdata/OvertimePolitics.comFeb12-Feb14RepublicanPrimaryPoll-Virginia.pdf

[I Will Not Be Wrong]

Good for Rubio!

[EliteLX]

This is an Overtime Politics poll, AKA a literal blog style most-likely single employee "pollster company" dropping intricate statewide polls day in and day out, so I'm taking it not even with a grain of salt.

Once we get some actual VA primary results, you will probably see a very close gap between Rubes and Donald. Jeb will probably also beat out Ted or maybe Kasich.

[I spent the winter writing songs about getting better]

THIS IS NOT A REAL POLL. THIS IS SOME BLOGGER IDIOT WHO JUST PULLS NUMBERS OUT OF HIS ASS.

[Starbucks Union Thug HokeyPuck]

THIS IS NOT A REAL POLL. THIS IS SOME BLOGGER IDIOT WHO JUST PULLS NUMBERS OUT OF HIS ASS.

So, this guy doesn't even poll anybody?  I have no idea what the deal is.

[I spent the winter writing songs about getting better]

THIS IS NOT A REAL POLL. THIS IS SOME BLOGGER IDIOT WHO JUST PULLS NUMBERS OUT OF HIS ASS.

So, this guy doesn't even poll anybody?  I have no idea what the deal is.

Many have proven that some of his numbers are literally impossible. For example Adam Griffin noted he had a Michigan poll with a subsample of the "Other" race showing Hillary 46-Bernie 45. However the percentage of voters in Michigan who aren't white, black or Hispanic is only 3%. To have a one percentage point difference like that, you'd need a sample of around 100, even with rounding it'd probably require more than 70. The TOTAL sample size of the poll was about 400, so it would require that almost 20% of respondents be not white, black or Hispanic for those numbers to even be conceivable.

Just one example. I'm sure you can find countless other examples of implausible nonsense if you page through the archives. The fact that so many dumbs think this crap is real is unbelievable.

It is also worth noting that this operation simply posts polls for free with no way to monetize them, yet the expenses of conducting as many polls as it does would have to be at least $10k/week.

[This account no longer in use.]

THIS IS NOT A REAL POLL. THIS IS SOME BLOGGER IDIOT WHO JUST PULLS NUMBERS OUT OF HIS ASS.

So, this guy doesn't even poll anybody?  I have no idea what the deal is.

Many have proven that some of his numbers are literally impossible. For example Adam Griffin noted he had a Michigan poll with a subsample of the "Other" race showing Hillary 46-Bernie 45. However the percentage of voters in Michigan who aren't white, black or Hispanic is only 3%. To have a one percentage point difference like that, you'd need a sample of around 100, even with rounding it'd probably require more than 70. The TOTAL sample size of the poll was about 400, so it would require that almost 20% of respondents be not white, black or Hispanic for those numbers to even be conceivable.

Just one example. I'm sure you can find countless other examples of implausible nonsense if you page through the archives. The fact that so many dumbs think this crap is real is unbelievable.

It is also worth noting that this operation simply posts polls for free with no way to monetize them, yet the expenses of conducting as many polls as it does would have to be at least $10k/week.

This is completely true. I honestly don't see why you're getting so worked up over this, though.

[The Other Castro]

THIS IS NOT A REAL POLL. THIS IS SOME BLOGGER IDIOT WHO JUST PULLS NUMBERS OUT OF HIS ASS.

So, this guy doesn't even poll anybody?  I have no idea what the deal is.

Many have proven that some of his numbers are literally impossible. For example Adam Griffin noted he had a Michigan poll with a subsample of the "Other" race showing Hillary 46-Bernie 45. However the percentage of voters in Michigan who aren't white, black or Hispanic is only 3%. To have a one percentage point difference like that, you'd need a sample of around 100, even with rounding it'd probably require more than 70. The TOTAL sample size of the poll was about 400, so it would require that almost 20% of respondents be not white, black or Hispanic for those numbers to even be conceivable.

Just one example. I'm sure you can find countless other examples of implausible nonsense if you page through the archives. The fact that so many dumbs think this crap is real is unbelievable.

It is also worth noting that this operation simply posts polls for free with no way to monetize them, yet the expenses of conducting as many polls as it does would have to be at least $10k/week.

That's why we have the megathread, so they don't get their own threads in the primary board.

[Ebsy]

The Overtime crosstabs on the Democratic side are comically bad, especially in relation to age. They have 18-30 and 60+ making up equal shares of the electorate, when the 2008 exit polls have 60+ at 25% and 18-29 at 15%. The race and gender crosstabs are also pretty far off, though not as egregious as some of the other s[Inks] that they have put out.

[Joe Republic]

Arkansas D:

Clinton 51%
Sanders 47%
Undecided 4%





Arkansas R:

Trump 32%
Cruz 14%
Rubio 12%
Kasich 12%
Carson 10%
Bush 7%

[Attorney General & PPT Dwarven Dragon]

Arkansas D:

Clinton 51%
Sanders 47%
Undecided 4%





Arkansas R:

Trump 32%
Cruz 14%
Rubio 12%
Kasich 12%
Carson 10%
Bush 7%

This doesn't actually exist

[I spent the winter writing songs about getting better]

OK someone explain to me how this Virginia poll isn't full of s[inks]: http://overtimepolitics.com/pollingdata/OvertimePolitics.comFeb12-Feb14DemocraticPrimaryPoll-Virginia.pdf

It has Sanders winning 18-30s 75-21, a very plausible number. But let's look at the income subsamples. Over 10% of respondents are in the <$10,000 category. It's probably safe to assume almost anyone in that category is either a student or young still living with their parents (even that's pretty low, I made over $10k working part time in college over 10 years ago), so they'd probably be strong for Sanders right? Nope! 57-38 for Hillary! Oh and Sanders only wins the $10k-$25k bracket 52-43, and the $25k-45k bracket 53-40. In fact by my calculations Hillary actually wins the under $25k, which makes up 36% of respondents in a state with a per capita income of $44k which is known for tons of affluent Democrats.

But just by the numbers, the poll cites 96 18-30s giving Sanders 75%, and yet 176 respondents with under $45k income giving Sanders only about 49% and a narrow plurality. Anyone with basic math skills can tell you that unless 18-30s are just as likely to make over $45k than the other groups and have a distribution amongst the income brackets comparable to that of the poll at large, this is simply not possible.

[Joe Republic]

Arkansas D:

Clinton 51%
Sanders 47%
Undecided 4%





Arkansas R:

Trump 32%
Cruz 14%
Rubio 12%
Kasich 12%
Carson 10%
Bush 7%

This doesn't actually exist

Are you accusing me of making up polling numbers?

[Attorney General & PPT Dwarven Dragon]

Arkansas D:

Clinton 51%
Sanders 47%
Undecided 4%





Arkansas R:

Trump 32%
Cruz 14%
Rubio 12%
Kasich 12%
Carson 10%
Bush 7%

This doesn't actually exist

Are you accusing me of making up polling numbers?

Yes. You're no better than the sh**tposters over in the random internal polls thread.

[I spent the winter writing songs about getting better]

You know what would be amazing? If any of the believers in this "polling company" would actually respond to any of the debunkings of how bad or implausible their numbers look that have appeared here, and on many other sites, both of the left wing and right wing, and issue a defense of the numbers based in actual statistics and mathematical concepts instead of just saying "oh but the guy did an AMA on Reddit, this must be totally legit!"

[○∙◄☻¥tπ[╪AV┼cVê└]

Lets just see how they do with NV, SC, and the Super Tuesday states.

[#TheShadowyAbyss]

I can't believe this thread is 24 pages long...

[ProgressiveCanadian]

Are we really still entering Overtime polls? ugh.

Yes troll.

OMG YPU ACTUALLY SERIOUSLY BELIEVE THESE ARE REAL POLLS THAT THESE ARE REAL POLLS WHAT THE F[INKS]

HEY, these polls are getting entered not by me. I'm just tired of the 24 pages of b*tching about it. I just pointed out an obvious troll no less.

[Joe Republic]

Arkansas D:

Clinton 51%
Sanders 47%
Undecided 4%





Arkansas R:

Trump 32%
Cruz 14%
Rubio 12%
Kasich 12%
Carson 10%
Bush 7%

This doesn't actually exist

Are you accusing me of making up polling numbers?

Yes. You're no better than the sh**tposters over in the random internal polls thread.

How dare you.  I typed out those poll numbers, and therefore they exist.  I will now enter them into the database.

[Gustaf]

Haha, I had totally missed all this stuff about Overtime. I thought they were just a bad pollster, not ARG-style frauds.

[Joe Republic]

California D:

Sanders 47%
Clinton 44%

Not sure why they bothered to poll this last in the nation, rubber stamp primary, though I guess it's a little surprising nonetheless.

[The Other Castro]

Joevertime Politics is a highly respected pollster. He included percentage symbols and everything.

[RBH]

the most suspicious thing about OTP polls?

their site has open directories!!