# 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:** 35
**Author:** ![DSeid](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/dseid/32/20194_2.png) [@DSeid](https://boards.straightdope.com/u/DSeid)\ **Post date:** [November 3, 2024, 12:34am UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/686 "2024-11-03T00:34:12Z")
> [@Jas09](#): > > What if, for example, Harris has weakened marginally with black and hispanic men, but strengthened significantly with white women (of all education levels) relative to Biden 2020? Care to participate here? > [@Prediction for specific demographic changes compared to 2020](https://boards.straightdope.com/t/prediction-for-specific-demographic-changes-compared-to-2020/1009511): > > Up down for total turnout to which side using 2020 as the baseline. Sources: for [younger voters subgroups](https://circle.tufts.edu/latest-research/2020-youth-voter-turnout-raceethnicity-and-gender); and [other groupings](https://www.pewresearch.org/politics/2021/06/30/behind-bidens-2020-victory/). Younger apparently gets defined as 18 to 29. My short version: Younger voters have an increase in female voter turnout and D preference across all subgroups. Biggest increase in young Black women, increasing from 47 to 55 or even 60%. The younger male vote remains an unchanged total turnout for Latino and Black men (39 and 31%) with a marginal but real improvement to… ---
**Author:** ![DSeid](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/dseid/32/20194_2.png) [@DSeid](https://boards.straightdope.com/u/DSeid)\ **Post date:** [November 3, 2024, 12:36am UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/687 "2024-11-03T00:36:30Z")
It is good to have such divergent polls at the same time. I’m expected Emerson to be humiliated. ---
**Author:** ![Munch](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/munch/32/5281_2.png) [@Munch](https://boards.straightdope.com/u/Munch)\ **Post date:** [November 3, 2024, 12:36am UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/688 "2024-11-03T00:36:34Z")
What Im curious about is the correlation between Iowa and Wisconsin. I can certainly see a big swing in Iowa being attributed to the abortion ban there (8 weeks), and that’s not entirely dissimilar to Wisconsin (abortion banned at 22 weeks). But there are massive ad buys in Wisconsin right now - how much of that bleeds over into bordering Iowa? ---
**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 3, 2024, 12:43am UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/689 "2024-11-03T00:43:44Z")
> [@Smapti](#): > > If the Supreme Court winds up being the reason Trump loses I will consider it adequate revenge for 2000. But is it worth the actual loss of Roe? Interesting question (and not really answerable, because we’ll never for sure know if Trump would have won with Roe intact). Best outcome is that the loss of Roe led to Trump’s defeat, AND Roe is then restored fully. ---
**Author:** ![DSeid](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/dseid/32/20194_2.png) [@DSeid](https://boards.straightdope.com/u/DSeid)\ **Post date:** [November 3, 2024, 12:56am UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/690 "2024-11-03T00:56:59Z")
> [@Munch](#): > > how much of that bleeds over into bordering Iowa The border area is not huge and not near significant population centers. Dubuque, which narrowly went Trump in 2020, borders, but no major media market near on the WI side. Platteville maybe? ---
**Author:** ![DSeid](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/dseid/32/20194_2.png) [@DSeid](https://boards.straightdope.com/u/DSeid)\ **Post date:** [November 3, 2024, 1:21am UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/691 "2024-11-03T01:21:43Z")
> [@DSeid](#): > > OTOH I’m not sure how much harm it does? Basically it results in more a lean to a median than average result? Tossing away outlier findings before they even get published. The unpublished outlier results _should_ be equally distributed in both directions if the house is otherwise The Selzer poll demonstrates to me the harm done by adopting a median approach that I was questioning. She may be wrong, outliers sometimes are, but boy this single outlier, really raises the probability of the herd having systemic error overestimating Trump support _dramatically_, and by itself virtually eliminates the possibility of a systemic error that is underestimating his support again. She may be wrong but she won’t be _that_ wrong. ---
**Author:** ![PhillyGuy](https://avatars.discourse-cdn.com/v4/letter/p/ed655f/32.png) [@PhillyGuy](https://boards.straightdope.com/u/PhillyGuy)\ **Post date:** [November 3, 2024, 2:17am UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/692 "2024-11-03T02:17:06Z")
> [@Jas09](#): > > There is obviously a big risk in doing that when you have sub-1% response rates . . . Are you sure she has that kind of response rate? I would have guessed that the phrase “Iowa Poll” is well known in Iowa as a state point of pride, so lots of people answer. Or am I hopelessly naive? ---
**Author:** ![Jas09](https://avatars.discourse-cdn.com/v4/letter/j/d07c76/32.png) [@Jas09](https://boards.straightdope.com/u/Jas09)\ **Post date:** [November 3, 2024, 2:20am UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/693 "2024-11-03T02:20:16Z")
> [@DSeid](#): > > She may be wrong but she won’t be _that_ wrong. Well, I mean, she _could be_, right? It would be catastrophic for her, but it’s certainly possible with her methods to get a sample so unrepresentative that you get a result significantly outside the MOE. > [@PhillyGuy](#): > > Are you sure she has that kind of response rate? That was a quote from Morris form 538 - I don’t have the tweet on hand. Here’s a quote from an MSNBC article: " Nate Cohn of The New York Times reported a [0.4%](https://www.nytimes.com/2022/10/12/upshot/midterms-polling-phone-calls.html)response rate to his polls in 2022." ETA: But yeah, her response rate could definitely be better. ---
**Author:** ![DSeid](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/dseid/32/20194_2.png) [@DSeid](https://boards.straightdope.com/u/DSeid)\ **Post date:** [November 3, 2024, 2:34am UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/694 "2024-11-03T02:34:12Z")
> [@Jas09](#): > > Well, I mean, she _could be_, right? It would be catastrophic for her, but it’s certainly possible with her methods Her reputation for being right when her result was the outlier is such that even being that wrong once wouldn’t be catastrophic for her. But yes it is _possible_. Looking at the two polls, and throw in that she had the Biden Trump matchup in June at Trump up 18, what combination of wrong can we come up with? The odds that there is not a real trend line in this population seems small. Silver apparently still gives Iowa to Trump by 3.4. I for one would not bet against Selzer. ---
**Author:** ![DSeid](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/dseid/32/20194_2.png) [@DSeid](https://boards.straightdope.com/u/DSeid)\ **Post date:** [November 3, 2024, 2:45am UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/695 "2024-11-03T02:45:26Z")
This 2015 interview with her may be of interest to some here. > **[Iowa’s Ann Selzer on what journalists need to know about polling](https://www.cjr.org/united_states_project/ann_selzer_iowa_poll_interview.php)** > > The profession of political polling is “teetering on the edge of disaster,” Jill Lepore writes in this week’s New Yorker. It’s a stark diagnosis, but not a fringe opinion. As voters continue their migration from landlines to... ---
**Author:** ![PhillyGuy](https://avatars.discourse-cdn.com/v4/letter/p/ed655f/32.png) [@PhillyGuy](https://boards.straightdope.com/u/PhillyGuy)\ **Post date:** [November 3, 2024, 2:47am UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/696 "2024-11-03T02:47:15Z")
> [@DSeid](#): > > Silver apparently still gives Iowa to Trump by 3.4. I for one would not bet against Selzer. I would not bet against Selzer either. But I also would not bet that as Iowa goes, so goes the nation. It may be that the timing of their 6 week abortion ban (effective late July 2024), plus something about the mood in Iowa, hurts Trump there in a unique way. All that may not translate to other states. ---
**Author:** ![RickJay](https://avatars.discourse-cdn.com/v4/letter/r/bb73d2/32.png) [@RickJay](https://boards.straightdope.com/u/RickJay)\ **Post date:** [November 3, 2024, 5:34am UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/697 "2024-11-03T05:34:15Z")
> [@Smapti](#): > > IMO, it’s had an even larger impact than the advent of the Internet did in the '90s. You probably have to go back to the dawn of the telegraph to find an invention that produced so much change so fast. A lot of it bad. I’m quite convinced it gave us Donald Trump quite possibly being re-elected. ---
**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 3, 2024, 11:56am UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/698 "2024-11-03T11:56:35Z")
> [@PhillyGuy](#): > > I would not bet against Selzer either. > > But I also would not bet that as Iowa goes, so goes the nation. It may be that the timing of their 6 week abortion ban (effective late July 2024), plus something about the mood in Iowa, hurts Trump there in a unique way. All that may not translate to other states. Good points. I think it raises\* Harris’ chances in Wisconsin modestly; in Michigan, a tiny bit; and that’s about it. But that’s still pretty important. (\*Better to say “it exposes what have, for a while, been higher-than-most-thought chances…”) ---
**Author:** ![TeroSunbear](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/terosunbear/32/14748_2.png) [@TeroSunbear](https://boards.straightdope.com/u/TeroSunbear)\ **Post date:** [November 3, 2024, 12:03pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/699 "2024-11-03T12:03:42Z")
With their fixed democrat republican and independent percentages, all the polls get to say: we were right, it was close! ---
**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 3, 2024, 12:13pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/700 "2024-11-03T12:13:09Z")
I don’t understand how VP Harris can be deadlocked with him in PA (per the last NY Times / Siena poll), but in the PA Senate, Casey (D) is polling consistently ahead of McCormick (R) by 6-ish points. It seems odd to me that there could be that much “ticket splitting” going on. > **[Toplines: October 2024 Times/Inquirer/Siena Poll of Registered Voters in...](https://www.nytimes.com/interactive/2024/11/03/us/elections/times-siena-pennsylvania-poll-crosstabs.html)** > > Results of a New York Times/Philadelphia Inquirer/Siena College poll conducted among 1,527 voters in Pennsylvania from Oct. 29 to Nov. 2, 2024. ---
**Author:** ![Fretful\_Porpentine](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/fretful_porpentine/32/2873_2.png) [@Fretful\_Porpentine](https://boards.straightdope.com/u/Fretful_Porpentine)\ **Post date:** [November 3, 2024, 12:34pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/701 "2024-11-03T12:34:39Z")
> [@Win\_Place\_Show](#): > > I don’t understand how VP Harris can be deadlocked with him in PA (per the last NY Times / Siena poll), but in the PA Senate, Casey (D) is polling consistently ahead of McCormick (R) by 6-ish points. > > It seems odd to me that there could be that much “ticket splitting” going on. Honestly, that sounds like more or less exactly what I’d expect, considering that Casey is 1) male; 2) white; and 3) a longtime incumbent who, as far as I know, hasn’t been involved in any major scandals or done anything exceptionally likely to piss people off. Harris was always going to lose some votes on account of her race and gender, and I expect that dynamic would be especially strong in a state with a lot of white working-class voters who are trending away from the current incarnation of the Democratic party, but might still be willing to vote for an old-school Democrat that they’ve voted for several times before. ---
**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 3, 2024, 2:21pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/702 "2024-11-03T14:21:33Z")
FWIW, **[NYT/Siena dropped another set of battleground state polls](https://scri.siena.edu/2024/11/03/new-york-times-siena-college-final-battleground-polls/)**: - TOO CLOSE TO CALL!! - Arizona: Trump 49% – Harris 45%; Gallego up by 5 Over Lake - Georgia: Harris 48% – Trump 47% - Michigan: Trump 47% – Harris 47%; Slotkin Leads Rogers 48-46% - North Carolina: Harris 48% – Trump 46%; Stein Leads by 17 Points - Nevada: Harris 49% – Trump 46%; Rosen by 9 Over Brown - Pennsylvania: Presidential Vote Tied; Casey 50% – McCormick 45% - Wisconsin: Harris 49% – Trump 47%; Baldwin 50% – Hovde 46% ---
**Author:** ![Ludovic](https://avatars.discourse-cdn.com/v4/letter/l/7ab992/32.png) [@Ludovic](https://boards.straightdope.com/u/Ludovic)\ **Post date:** [November 3, 2024, 2:51pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/703 "2024-11-03T14:51:37Z")
Wow, no breathing room with those results. If Trump wins the two tied states as well as the ones with a lead, Harris can’t lose any others (unless she loses Nevada and flips Iowa.) ---
**Author:** ![DSeid](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/dseid/32/20194_2.png) [@DSeid](https://boards.straightdope.com/u/DSeid)\ **Post date:** [November 3, 2024, 4:03pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/704 "2024-11-03T16:03:28Z")
Focusing on the degrees to which polling is broken. Selzer may be a special case. The Iowa population is less demographically diverse, and it is widely believed that voters there are more likely to answer their phones and participate in _that_ poll than other states. She famously does NOT model what she thinks the electorate _should_ look like, but lets the responses tell her what it does look like. Her line in that linked interview was that history is a great predictor until it isn’t, or something like that. I don’t think that approach works in more diverse voter demographics with different participation tendencies. I’m not sure which would amuse me more - Harris winning with Florida or Iowa being the tipping point. Iowa being the less improbable. NYT/Siena changes from where they had been strikes me as odd. Did voters really move like that? Sunbelts significantly better for Harris and Blue Walls to ties? Or is this just their having one set of funky numbers. That much real shift this late? Really? ---
**Author:** ![RickJay](https://avatars.discourse-cdn.com/v4/letter/r/bb73d2/32.png) [@RickJay](https://boards.straightdope.com/u/RickJay)\ **Post date:** [November 3, 2024, 5:10pm UTC](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779/705 "2024-11-03T17:10:09Z")
I’ve found a sort of possible map where we have, horribly enough, a tie: > **[270toWin - 2024 Presidential Election Interactive Map](https://www.270towin.com/maps/ODdrQ)** > > Create your own forecast for the 2024 presidential election This requires Trump to win Maine-2 (likely) but Harris lose NE-2 (unlikely) but it’s possible. Were Harris to win NE-2 she wins 270-268, which, honestly, might be even MORE terrifying. For Trump, 269-269 is a victory. The state delegations will hand him the crown. But if it’s 270-268, I do not believe the Trumpists incapable of anything in their rush to fuck things up. [Previous page](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779.md?page=34) [Next page](https://boards.straightdope.com/t/polling-is-broken-right-or-is-it-the-news-or-all-of-the-us/993779.md?page=36)