Maybe the Polls are Rigged?

[NatMorton]

Ok, I know when the Republicans say "the media's polls are rigged!" it sounds like mindless Trumpism, but more and more a rational mind has to agree there is something to these complaints.

A case in point, The Cook Political Report bills itself as "... an independent, non-partisan newsletter that analyzes elections and campaigns for the US House of Representatives, US Senate, Governors and President as well as American political trends." Their studies are cited widely in the media as "non-partisan." (a quick Google will confirm this).

Here's the thing, on November 2nd, the day before the election, they published a report labeling 27 House races as "toss ups."

https://cookpolitical.com/ratings/house-race-ratings

... yet the GOP won every single one of those 27 races. How can that happen? I realize it's a bit simplistic to liken a close race to a coin flip, but the etymology of the term "toss-up" does come from the act of flipping a coin, so a 50/50 chance is clearly what is meant by the term. The odds of flipping a coin 27 times and having it come up heads 27 times in a row is 1 in 134,217,728 (i.e. 2 to the 27th power). A quick search of the web tells me those chances are roughly 268 times more remote than the chances of any one of you reading this message being struck by lightning this year.

So are we really to believe Cook was this wrong merely by chance?

 

[Emily L]

Nat, I honestly believe that conservatives tend to be hesitant to answer polls honestly, due to fear of some kind of backlash. I think that's why conservative candidates tend to out-perform their poll numbers.

 

[CaughtInThe]

I bunch of people were massively embarrassed to tell someone on the phone that they were going to vote for Trump again.

 

[NatMorton]

I bunch of people were massively embarrassed to tell someone on the phone that they were going to vote for Trump again.

Um, these polls weren't forecasting votes for Trump.

 

[NatMorton]

Nat, I honestly believe that conservatives tend to be hesitant to answer polls honestly, due to fear of some kind of backlash. I think that's why conservative candidates tend to out-perform their poll numbers.

While that explanation fits the facts, why is it a more likely explanation than media bias?

 

[Phys251]

Ok, I know when the Republicans say "the media's polls are rigged!" it sounds like mindless Trumpism, but more and more a rational mind has to agree there is something to these complaints.

A case in point, The Cook Political Report bills itself as "... an independent, non-partisan newsletter that analyzes elections and campaigns for the US House of Representatives, US Senate, Governors and President as well as American political trends." Their studies are cited widely in the media as "non-partisan." (a quick Google will confirm this).

Here's the thing, on November 2nd, the day before the election, they published a report labeling 27 House races as "toss ups."

https://cookpolitical.com/ratings/house-race-ratings

... yet the GOP won every single one of those 27 races. How can that happen? I realize it's a bit simplistic to liken a close race to a coin flip, but the etymology of the term "toss-up" does come from the act of flipping a coin, so a 50/50 chance is clearly what is meant by the term. The odds of flipping a coin 27 times and having it come up heads 27 times in a row is 1 in 134,217,728 (i.e. 2 to the 27th power). A quick search of the web tells me those chances are roughly 268 times more remote than the chances of any one of you reading this message being struck by lightning this year.

So are we really to believe Cook was this wrong merely by chance?

Your next-to-last paragraph implies that all 27 of the House races in question were statistically independent, when that is most certainly not the case.

 

[NatMorton]

Your next-to-last paragraph implies that all 27 of the House races in question were statistically independent, when that is most certainly not the case.

I realize it's not a perfect comparison. The point stands: 27 "toss-ups" all going in only one direction by mere chance? That's a bit much to believe.

 

[bearpoker]

Um, these polls weren't forecasting votes for Trump.

Republicans in congress have been embarrassing as well. The dirty little secret vote was bigger than anyone thought.

 

[bearpoker]

While that explanation fits the facts, why is it a more likely explanation than media bias?

Republicans are natural born liars.

 

[Phys251]

I realize it's not a perfect comparison. The point stands: 27 "toss-ups" all going in only one direction by mere chance? That's a bit much to believe.

It was a terrible comparison. Do you even know how correlation works?

 

[CaughtInThe]

Um, these polls weren't forecasting votes for Trump.

they were still embarrassed Trump voters. why would they partake in any political poll? are there not Trump supporters running for office?

 

[Kreton]

Nat, I honestly believe that conservatives tend to be hesitant to answer polls honestly, due to fear of some kind of backlash. I think that's why conservative candidates tend to out-perform their poll numbers.

I disagree. For this to make sense it would have to imply some sort of backlash exists. Do those on the right really believe the pollster on the end of a phone is going to come after them? That is pretty dumb if they do. And if they are so hesitent to answer an anonymous poll online due to fear of backlash why would they be so vocal on social media where people know who they are? With giant signs, t-shirts, and hats. Bumper stickers and walk around praising the gospel of Trump? That doesn't add up.

More likely is that like Trump his supporters simply like to lie. They feed pollsters bad info because they don't like pollsters.

 

[Emily L]

While that explanation fits the facts, why is it a more likely explanation than media bias?

You might be right about the media bias, Nat, but I was just applying that old scientific guideline, "The simplest explanation is the probably the correct one." Media bias is a little more complicated (for me, anyway) than simple fibbing to a pollster.

 

[Emily L]

I disagree. For this to make sense it would have to imply some sort of backlash exists. Do those on the right really believe the pollster on the end of a phone is going to come after them? That is pretty dumb if they do. And if they are so hesitent to answer an anonymous poll online due to fear of backlash why would they be so vocal on social media where people know who they are? With giant signs, t-shirts, and hats. Bumper stickers and walk around praising the gospel of Trump? That doesn't add up.

More likely is that like Trump his supporters simply like to lie. They feed pollsters bad info because they don't like pollsters.

You make good points, Kreton. From what I've been told, it's kind of a paranoia thing. For example, they might fear that their telephone number might be put on a list of some kind--or maybe fear that the caller isn't really a pollster but a nearby vandal trying to find out which house to egg. You're certainly right about that social media thing, which indicates that not all conservatives have that paranoia thing going. But anyway, it's just a theory, so I might be wrong (as usual).

 

[Taylor]

If you ask the pollsters, they're quick to tell you that "the polls were right" - of course, their livelihood depends on that lie.

 

[Kreton]

You make good points, Kreton. From what I've been told, it's kind of a paranoia thing. For example, they might fear that their telephone number might be put on a list of some kind--or maybe fear that the caller isn't really a pollster but a nearby vandal trying to find out which house to egg. You're certainly right about that social media thing, which indicates that not all conservatives have that paranoia thing going. But anyway, it's just a theory, so I might be wrong (as usual).

I don't think they'd need a phone call to find the houses. The big Trump signs and statues of Trump would be a pretty good give away.

 

[Emily L]

I don't think they'd need a phone call to find the houses. The big Trump signs and statues of Trump would be a pretty good give away.

"Statues of Trump"? If that's a real thing, then you've just given me a great idea of what to send my liberal friends as gag gifts for Christmas!

 

[Eriech]

I was told by the media that Susan Collins would lose to Sarah Gideon. Susan won; it was not close. We were told Graham and McConnell could lose, it wasn't close. Maine, South Carolina and Kentucky do not use Dominion voting systems. We were told McSally and Gardiner and James would lose......you guessed it, Dominion.

Polls fit the narratives the media wants to spin. The Real Clear average is a devious device used to convey legitimacy to garbage polling

 

[ttwtt78640]

Ok, I know when the Republicans say "the media's polls are rigged!" it sounds like mindless Trumpism, but more and more a rational mind has to agree there is something to these complaints.

A case in point, The Cook Political Report bills itself as "... an independent, non-partisan newsletter that analyzes elections and campaigns for the US House of Representatives, US Senate, Governors and President as well as American political trends." Their studies are cited widely in the media as "non-partisan." (a quick Google will confirm this).

Here's the thing, on November 2nd, the day before the election, they published a report labeling 27 House races as "toss ups."

https://cookpolitical.com/ratings/house-race-ratings

... yet the GOP won every single one of those 27 races. How can that happen? I realize it's a bit simplistic to liken a close race to a coin flip, but the etymology of the term "toss-up" does come from the act of flipping a coin, so a 50/50 chance is clearly what is meant by the term. The odds of flipping a coin 27 times and having it come up heads 27 times in a row is 1 in 134,217,728 (i.e. 2 to the 27th power). A quick search of the web tells me those chances are roughly 268 times more remote than the chances of any one of you reading this message being struck by lightning this year.

So are we really to believe Cook was this wrong merely by chance?

Polls referring to a race as being a “toss up” usually means that the predicted (projected?) winner/loser is within the poll’s expected margin of error.

 

[Guyzilla]

Ok, I know when the Republicans say "the media's polls are rigged!" it sounds like mindless Trumpism, but more and more a rational mind has to agree there is something to these complaints.

A case in point, The Cook Political Report bills itself as "... an independent, non-partisan newsletter that analyzes elections and campaigns for the US House of Representatives, US Senate, Governors and President as well as American political trends." Their studies are cited widely in the media as "non-partisan." (a quick Google will confirm this).

Here's the thing, on November 2nd, the day before the election, they published a report labeling 27 House races as "toss ups."

https://cookpolitical.com/ratings/house-race-ratings

... yet the GOP won every single one of those 27 races. How can that happen? I realize it's a bit simplistic to liken a close race to a coin flip, but the etymology of the term "toss-up" does come from the act of flipping a coin, so a 50/50 chance is clearly what is meant by the term. The odds of flipping a coin 27 times and having it come up heads 27 times in a row is 1 in 134,217,728 (i.e. 2 to the 27th power). A quick search of the web tells me those chances are roughly 268 times more remote than the chances of any one of you reading this message being struck by lightning this year.

So are we really to believe Cook was this wrong merely by chance?

It all boils down to what I have been saying for YEARS. Changes in PHONE habits, have thrown polling into chaos. I stopped working the phones for Dems, due to this. And I couldnt get anyone to listen to me.

 

[Cardinal]

Ok, I know when the Republicans say "the media's polls are rigged!" it sounds like mindless Trumpism, but more and more a rational mind has to agree there is something to these complaints.

A case in point, The Cook Political Report bills itself as "... an independent, non-partisan newsletter that analyzes elections and campaigns for the US House of Representatives, US Senate, Governors and President as well as American political trends." Their studies are cited widely in the media as "non-partisan." (a quick Google will confirm this).

Here's the thing, on November 2nd, the day before the election, they published a report labeling 27 House races as "toss ups."

https://cookpolitical.com/ratings/house-race-ratings

... yet the GOP won every single one of those 27 races. How can that happen? I realize it's a bit simplistic to liken a close race to a coin flip, but the etymology of the term "toss-up" does come from the act of flipping a coin, so a 50/50 chance is clearly what is meant by the term. The odds of flipping a coin 27 times and having it come up heads 27 times in a row is 1 in 134,217,728 (i.e. 2 to the 27th power). A quick search of the web tells me those chances are roughly 268 times more remote than the chances of any one of you reading this message being struck by lightning this year.

So are we really to believe Cook was this wrong merely by chance?

Nobody is helped by the polls being wrong.

 

[ttwtt78640]

It all boils down to what I have been saying for YEARS. Changes in PHONE habits, have thrown polling into chaos. I stopped working the phones for Dems, due to this. And I couldnt get anyone to listen to me.

Yep, having incoming caller ID and the option to not even have the phone ring for unknown callers hinders unknown callers. Please leave a (likely to be ignored) message at the beep.

 

[bearpoker]

"Statues of Trump"? If that's a real thing, then you've just given me a great idea of what to send my liberal friends as gag gifts for Christmas!

No statues in my neighborhood but two neighbors have Trump banners still flying. The Biden signs have all been taken down since the election, but the Trump ones are still up.

 

[Kreton]

Yep, having incoming caller ID and the option to not even have the phone ring for unknown callers hinders unknown callers. Please leave a (likely to be ignored) message at the beep.

I don't even have voicemail anymore so there isn't even a beep. I hear when someone calls, if I want to talk I'll pick up. If I don't, I don't want to hear about it in a message either.

 

[MamboDervish]

While that explanation fits the facts, why is it a more likely explanation than media bias?

What on earth makes media bias a likely explanation under any circumstances? Polls may be reported by the media, but they're not always conducted by the media. In fact, they're not even all domestic. Just as an example, the Reuters/Ipsos polls are European. Ipsos is a market research company based in Paris, and Reuters an international wire service based in London. They have no skin in the game of American politics. And yet their polling numbers are never grossly out of line with other domestic polling services, including media polls.