# Polling is broken, right? Or is it the news? Or all of the US? **URL:** \ **Category:** Politics & Elections\ **Tags:** trump\ **Created:** [November 27, 2023, 5:30pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779 "2023-11-27T17:30:18Z")\ **Posts on this page:** 20\ **Page:** 32
**Author:** ![Happy\_Lendervedder](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/happy_lendervedder/32/8488_2.png) [@Happy\_Lendervedder](https://boards.straightdope.com/u/Happy_Lendervedder)\ **Post date:** [November 1, 2024, 3:20pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/626 "2024-11-01T15:20:31Z")
[Politico story](https://www.politico.com/news/magazine/2024/11/01/shy-kamala-harris-voters-polling-00186653) about how Harris voters may be undercounted in polling this year. If they’re located in large numbers in swing states, that could mean this isn’t as close as the polls are showing. Summary: Haley voters who are voting for Harris may be greatly underrepresented in polling numbers, and those “Uncommitted” Gaza-supporting voters may be telling pollsters they’re uncommitted just to send a message but are smart enough to know that a vote for Harris is in their best interest. ---
**Author:** ![DigitalC](https://avatars.discourse-cdn.com/v4/letter/d/ebca7d/32.png) [@DigitalC](https://boards.straightdope.com/u/DigitalC)\ **Post date:** [November 1, 2024, 3:24pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/627 "2024-11-01T15:24:03Z")
> [@Happy\_Lendervedder](#): > > Summary: Haley voters who are voting for Harris may be greatly underrepresented in polling numbers, and those “Uncommitted” Gaza-supporting voters may be telling pollsters they’re uncommitted just to send a message but are smart enough to know that a vote for Harris is in their best interest. From what I keep hearing there has been zero mention of any R voters for Harris swing in any single polls. If it is happening more than the normal few %s that happen every election it is not reflected at all any where. ---
**Author:** ![bordelond](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/bordelond/32/150_2.png) [@bordelond](https://boards.straightdope.com/u/bordelond)\ **Post date:** [November 1, 2024, 3:46pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/628 "2024-11-01T15:46:54Z")
> [@DigitalC](#): > > From what I keep hearing **there has been zero mention of any R voters for Harris swing in any single polls**. If it is happening more than the normal few %s that happen every election it is not reflected at all any where. (my emphasis) **[No, such polls are out there.](https://www.newsweek.com/kamala-harris-winning-republicans-pennsylvania-1970137)** They are few in number and don’t get repeated periodically like the horse-race polls. This one is specific to Pennsylvania: > A _New York Times_ /Siena College/_Philadelphia Inquirer_ poll found Harris winning 12 percent of Republicans in the state. It surveyed 857 likely voters from October 7 to 10 and had a margin of error of plus or minus 4 percentage points. ---
**Author:** ![bordelond](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/bordelond/32/150_2.png) [@bordelond](https://boards.straightdope.com/u/bordelond)\ **Post date:** [November 1, 2024, 3:52pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/629 "2024-11-01T15:52:31Z")
Late to the party, but the _New York Times_ is finally acknowledging that red-wave polling exists and is essentially known to be a tool of Republican supporters. As to be expected, the NYT is much more equivocal than liberal Substackers like Simon Rosenberg (who is one of the few Democratic voices quoted in the piece) and Jay Kuo. The Times takes care to note that red-wave polling doesn’t much change the big analysts’ averages – but at least recognizes that gamed polling has other deleterious effects. > **[Why the Right Thinks Trump Is Running Away With the Race](https://www.nytimes.com/2024/10/31/us/politics/trump-harris-partisan-polls.html)** > > Skewed polls and anonymous betting markets are building up Republicans’ expectations. Donald Trump could use that to challenge the result. _(Link is paywalled. I will see whether another outlet like MSN or Yahoo picks up the story.)_ ---
**Author:** ![Hawkeyeop](https://avatars.discourse-cdn.com/v4/letter/h/958977/32.png) [@Hawkeyeop](https://boards.straightdope.com/u/Hawkeyeop)\ **Post date:** [November 1, 2024, 4:07pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/630 "2024-11-01T16:07:05Z")
A lot of polls have the partisan breakdown per candidate, so you should be able to see that data. I also don’t see any reason why it wouldn’t be picked up in a normal polling average. I’m hopeful that Harris beats her polling a bit at least in swing states where voter engagement is high, but I think a lot of this is just grasping at straws. ---
**Author:** ![Hawkeyeop](https://avatars.discourse-cdn.com/v4/letter/h/958977/32.png) [@Hawkeyeop](https://boards.straightdope.com/u/Hawkeyeop)\ **Post date:** [November 1, 2024, 4:09pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/631 "2024-11-01T16:09:21Z")
Betting lines have recently moved towards Harris, so there is now little difference between polling and betting lines at this point. ---
**Author:** ![RickJay](https://avatars.discourse-cdn.com/v4/letter/r/bb73d2/32.png) [@RickJay](https://boards.straightdope.com/u/RickJay)\ **Post date:** [November 1, 2024, 4:36pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/632 "2024-11-01T16:36:10Z")
> [@Happy\_Lendervedder](#): > > Haley voters who are voting for Harris may be greatly underrepresented in polling numbers, and those “Uncommitted” Gaza-supporting voters may be telling pollsters they’re uncommitted just to send a message but are smart enough to know that a vote for Harris is in their best interest. It’s an interesting hypothesis and one hopes it’s true, but it’s just a hypothesis and no evidence is presented that it’s happening. There’s alos a huge logical leap here, one the author tries to sneak by us: that people who were “uncommitted” in the Demcoratic primary are, somehow, assumed to be answering “undecided” in polls about the general election. That claim is not supported by any evidence. > [@Hawkeyeop](#): > > Betting lines have recently moved towards Harris, so there is now little difference between polling and betting lines at this point. The betting lines clearly went irrationally far towards Trump on some platforms, especially in lines on the popular vote. Not ridiculously so - no legitimate site is giving you 5-1 or anything like that. IRL I’m still trying to line up some high odds bets with fools who say they’re SURE who will win, but when you ask people to put their money with their mouth is, they sure get quiet. ---
**Author:** ![bordelond](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/bordelond/32/150_2.png) [@bordelond](https://boards.straightdope.com/u/bordelond)\ **Post date:** [November 1, 2024, 5:17pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/633 "2024-11-01T17:17:50Z")
> [@RickJay](#): > > It’s an interesting hypothesis and one hopes it’s true, but it’s just a hypothesis and **no evidence is presented that it’s happening**. **[A few posts up?](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/628)** EDIT: **[Here’s some more.](https://www.politico.com/news/magazine/2024/11/01/shy-kamala-harris-voters-polling-00186653)** This kind of data is hard to find in media, but some researchers are pursuing it nevertheless. > A [**national survey**](https://blueprint2024.com/polling/haley-voters-10-9/) of Nikki Haley’s primary voters conducted in early October by the Democratic polling firm Blueprint has charted this group’s slow but significant shift away from supporting Trump and their increasing willingness to support Democratic presidential candidates over time. (Notably, this survey excludes registered Democrats who voted for Nikki Haley in primaries to ensure the results truly reflect Republican-leaning voter sentiments.) The survey showed that while 66 percent of Haley primary voters supported Trump in 2016, that number dropped to 59 percent in 2020 and is expected to drop even further to 45 percent in this year’s election. Meanwhile, their support for the Democratic presidential nominee has nearly tripled from only 13 percent supporting Hillary Clinton in 2016 to 36 percent indicating an intent to vote for Kamala Harris. EDIT2: Heh – I just posted the same link as @Happy_Lendervedder. But there’s also a link to the original polling if anyone wants to dive deeper. ---
**Author:** ![JKellyMap](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/jkellymap/32/15880_2.png) [@JKellyMap](https://boards.straightdope.com/u/JKellyMap)\ **Post date:** [November 1, 2024, 5:28pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/634 "2024-11-01T17:28:07Z")
> [@bordelond](#): > > (Notably, this survey excludes registered Democrats who voted for Nikki Haley in primaries to ensure the results truly reflect Republican-leaning voter sentiments.) Interesting – but note that this WOULD include people like me: living in a state where there isn’t official party _registration_, yet you can choose to pick a candidate from either party (but not from both parties) in any given primary. I am as Democrat as they come, but I voted for Haley. Does the survey you cited somehow count me as “Republican-leaning”? I don’t have time to read the details now from you link (maybe tomorrow), but I assume they asked people in states like mine something other than “are you a registered Democrat?”. Maybe “Do you usually vote for Democrats?” That _would_ have eliminated the likes of me. ---
**Author:** ![bordelond](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/bordelond/32/150_2.png) [@bordelond](https://boards.straightdope.com/u/bordelond)\ **Post date:** [November 1, 2024, 5:38pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/635 "2024-11-01T17:38:10Z")
> [@JKellyMap](#): > > Maybe “Do you usually vote for Democrats?” **[From page 2 of Blueprint’s poll:](https://blueprint2024.com/polling/haley-voters-10-9/)** > **5. What best describes your voting pattern over the last several elections?** > > - All Republican 13% > - Mostly Republicans 42% > - A few more Republicans than Democrats 19% > - Evenly Republican and Democratic 12% > - A few more Democrats than Republicans 8% > - Mostly Democrats 3% > - All Democrats 1% > - Not sure 2% Their survey results are prefaced by the following statement: > Blueprint surveyed 781 Haley voters, defined as Republicans, Republican-leaning independents or true independents who selected Nikki Haley in the primary. Our sample also includes Republicans and Republican-leaning independents who said they preferred Haley as the Republican nominee. We did not include any Democratic-identifying voters or Democratic-leaning independents in our sample. Can’t speak for others … but to me, Harris peeling off around 1/10 (or a little better) of self-identified _Republican_ Haley supporters … that’s not crazy talk. What’s unreasonable about it? It really isn’t that far out there, and there _is_ a little evidence out there to support the notion. ---
**Author:** ![JKellyMap](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/jkellymap/32/15880_2.png) [@JKellyMap](https://boards.straightdope.com/u/JKellyMap)\ **Post date:** [November 1, 2024, 5:46pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/636 "2024-11-01T17:46:55Z")
Thanks. I agree, not unreasonable at all. There is a reason she outperformed some expectations in some primaries, and it ain’t just because of folks like me. ---
**Author:** ![Win\_Place\_Show](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/win_place_show/32/13469_2.png) [@Win\_Place\_Show](https://boards.straightdope.com/u/Win_Place_Show)\ **Post date:** [November 1, 2024, 7:02pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/637 "2024-11-01T19:02:42Z")
> [@bordelond](#): > > Late to the party, but the _New York Times_ is finally acknowledging that red-wave polling exists and is essentially known to be a tool of Republican supporters. As to be expected, the NYT is much more equivocal than liberal Substackers like Simon Rosenberg (who is one of the few Democratic voices quoted in the piece) and Jay Kuo. The Times takes care to note that red-wave polling doesn’t much change the big analysts’ averages – but at least recognizes that gamed polling has other deleterious effects. > > [https://www.nytimes.com/2024/10/31/us/politics/trump-harris-partisan-polls.html](https://www.nytimes.com/2024/10/31/us/politics/trump-harris-partisan-polls.html) > > _(Link is paywalled. I will see whether another outlet like MSN or Yahoo picks up the story.)_ I’ll call your “late to the party” and raise you two “lates”. Here’s a gift link that goes with the one you posted a few yours ago. > **[Why the Right Thinks Trump Is Running Away With the Race](https://www.nytimes.com/2024/10/31/us/politics/trump-harris-partisan-polls.html?unlocked_article_code=1.Wk4.WkpS.JhCrGvJw1ACT&smid=url-share)** > > Skewed polls and anonymous betting markets are building up Republicans’ expectations. Donald Trump could use that to challenge the result. ---
**Author:** ![bordelond](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/bordelond/32/150_2.png) [@bordelond](https://boards.straightdope.com/u/bordelond)\ **Post date:** [November 1, 2024, 9:10pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/638 "2024-11-01T21:10:53Z")
Thanks for the assist, @Win_Place_Show. @DigitalC posted another NYT piece to the “Stop Panicking” thread – this one by their pollster Nate Cohn. IMHO, a pretty damning account of the state of affairs of current polling. It’s interesting to me that Cohn writes like he’s outside of the fray, not making the mistakes other pollsters are making. > [@Stop Panicking!](https://boards.straightdope.com/t/stop-panicking/1003447/1294): > > According to Nate Conh: [quote]It’s hard to overstate how traumatic the 2016 and 2020 elections were for many pollsters. For some, another underestimate of Mr. Trump could be a major threat to their business and their livelihood. For the rest, their status and reputations are on the line. If they underestimate Mr. Trump a third straight time, how can their polls be trusted again? It is much safer, whether in terms of literal self-interest or purely psychologically, to find a close race than to… > **[So, Can We Trust the Polls?](https://www.nytimes.com/2024/11/01/upshot/so-can-we-trust-the-polls.html)** > > Recent changes add up to a case for cautious optimism on better accuracy, but that’s far from a guarantee. _(paywalled)_ ---
**Author:** ![RickJay](https://avatars.discourse-cdn.com/v4/letter/r/bb73d2/32.png) [@RickJay](https://boards.straightdope.com/u/RickJay)\ **Post date:** [November 1, 2024, 9:13pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/639 "2024-11-01T21:13:45Z")
"Skewed polls"Reminds me of Dean Chambers, the guy in 2012 who ran “Unskewed Polls” who insisted Romney was headed for a victory. Along the way he called Nate Silver a homophobic slur. Anyway, after Obama won, he briefly admitted he was wrong and Silver was right. Then, fewer than TWO WEEKS later, he started a new website claiming the Democrats cheated and only won thru election fraud. Sound familiar? ---
**Author:** ![bordelond](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/bordelond/32/150_2.png) [@bordelond](https://boards.straightdope.com/u/bordelond)\ **Post date:** [November 1, 2024, 9:20pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/640 "2024-11-01T21:20:17Z")
> [@RickJay](#): > > "Skewed polls"Reminds me of Dean Chambers, the guy in 2012 … Stop right there: Regarding polling, 2012 might as well be 1912. The Nate Cohn article **DigitalC** posted details how pollsters have been casting about to regain at least a pre-Trump accuracy level and (probably) haven’t been able to do it. This Reddit thread about today’s Cohn article covers a lot of the same ground without a paywall (it is meta-discussion – the entire article is not quoted). [**Nate Cohn: “Pollsters are more willing to take steps to produce more Republican-leaning results”.**](https://www.reddit.com/r/fivethirtyeight/comments/1ghe2m3/nate_cohn_pollsters_are_more_willing_to_take/?rdt=57616) EDIT: I know not all Reddit rooms feature erudite, respectful discussion … but I thought this comment at the link above is well-considered: > I think this captures why, as a poll-truster, I’m still more optimistic on Harris than the polls this year. I know polls are conducted by humans, and the downsides of underestimating Harris are way way way way less bad than the downsides of underestimating Trump a 3rd time. There’s a lot of economic and psychological incentives for them to tinker with results that seem “too” D-leaning. Plus, they’ve spent 8 years designing methods to try to capture Trump voters, while no other potential source of error has gotten 5% of that attention. > > It seems likely to me that the dominant factor in poll error will be something different this year. > > EDIT: I’m not claiming malice. As a data scientist there’s hundreds of tiny decisions you need to make during analysis that seem arbitrary. But if one choice happens to give “safer” results, it’s hard to overcome that bias. ---
**Author:** ![RickJay](https://avatars.discourse-cdn.com/v4/letter/r/bb73d2/32.png) [@RickJay](https://boards.straightdope.com/u/RickJay)\ **Post date:** [November 1, 2024, 9:22pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/641 "2024-11-01T21:22:12Z")
> [@bordelond](#): > > Stop right there It was just a funny sidebar, man. ---
**Author:** ![bordelond](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/bordelond/32/150_2.png) [@bordelond](https://boards.straightdope.com/u/bordelond)\ **Post date:** [November 1, 2024, 9:26pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/642 "2024-11-01T21:26:22Z")
All good. Just seems like “skewed polls” has become something of a debate-ender … like a BS excuse for a given outcome. Reading both of today’s NYT polling articles back to back gives a fuller picture of what both Cohn and the other two NYT writers are trying to communicate. I see Cohn kind of slinking away from even what _his_ polls are saying. Maybe I’m being uncharitable. ---
**Author:** ![DigitalC](https://avatars.discourse-cdn.com/v4/letter/d/ebca7d/32.png) [@DigitalC](https://boards.straightdope.com/u/DigitalC)\ **Post date:** [November 1, 2024, 10:00pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/643 "2024-11-01T22:00:42Z")
Polls historically are systematically off about 2-3% in one direction, every time. It would be weird if the results were exactly what the polls are saying, specially with the ridiculous MOE they are having. It’s not dumb to go by “vibes” to try to figure out which way they might break. I’ve seen literally no reason to think Trump might be surging in the last month APART from the polls, if anything the wheels are coming off his campaign and they are on defense mode the entire time. ---
**Author:** ![PhillyGuy](https://avatars.discourse-cdn.com/v4/letter/p/ed655f/32.png) [@PhillyGuy](https://boards.straightdope.com/u/PhillyGuy)\ **Post date:** [November 2, 2024, 3:34am UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/644 "2024-11-02T03:34:37Z")
This Nate Silver article will be remembered, and linked to, for years. Decades. Maybe centuries. I hope it somehow is mistaken, because it justifies lack of social trust. But I have read [Retraction Watch](https://retractionwatch.com/) too often to dismiss it: > **[There’s more herding in swing state polls than at a sheep farm in the...](https://www.natesilver.net/p/theres-more-herding-in-swing-state)** > > Some pollsters aren’t telling you what their data really says. If Silver is correct, there needs to be formal retraction of some recent polls. ---
**Author:** ![RickJay](https://avatars.discourse-cdn.com/v4/letter/r/bb73d2/32.png) [@RickJay](https://boards.straightdope.com/u/RickJay)\ **Post date:** [November 2, 2024, 3:36am UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/645 "2024-11-02T03:36:44Z")
Or, a formal admission of OTHER polls. [Previous page](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779.md?page=31) [Next page](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779.md?page=33)