Resolves after the AP calls the race.
Related questions
As nate silver was saying at manifest, polls these days are just models with poll-flavored inputs. Non-response rates keep going up and correcting for that is hard.
And even if you successfully correct for non-response, you don't know how many of the kennedy responders will switch for the actual election when the stakes are higher.
Outside view: nevada hasn't gone red since 2004 and it's added a zillion immigrants (from california and from latin america) since then. It's probably staying blue.
How do you explain polling results like this? Nevada Cross-Tabs: May 2024 Times/Siena Poll - The New York Times (nytimes.com).
In the crosstabs, Trump is winning in every group except for white college graduates. He's even +11 among Hispanics.
Of course, I expect this is a bit of an outlier, and Trump isn't actually going to win by double digits. But I don't see how it's consistent with Biden being favored to win.
The RFK spoiler effect, plus sampling bias.
Based on 2020 turnout they seem to be massively under-sampling people who rarely vote (or people are lying). Surely interest in politics correlates with willingness to waste 10 minutes taking a survey about politics.
They seem to also be massively undersampling people with low life satisfaction, especially democrats with low life satisfaction. I don't know how you square this with GSS showing huge rates of depression among liberals. Being depressed would make people less likely to respond.
Disaffected democrats might or might not show up to vote in greater numbers than this poll predicts, and RFK voters might or might not switch to Biden.
What's the outside view on how many votes a third party candidate polling at 12% in June actually gets in November?
In this poll, RFK is taking 11% of Trump 2020 voters and 8% of Biden 2020 voters, suggesting Trump would gain slightly by RFK voters dropping off.
If they are undersampling people who rarely vote, they are very likely underestimating Trump
My thinking is not mostly based on polls, but demographics. Nevada has moved to the right 3 presidential elections in a row, and migration effects are hugely overrated in general. It seems extremely likely that Biden will see a national drop with both hispanic voters and young voters- Nevada is the 5th most hispanic state and has a lot of young voters. And if the presidential election is ~tied in the popular vote, Nevada is highly likely going to Trump.
I see your point, but Obama was just a stronger candidate than Biden or Clinton. 2016 NV dem margin was 2.42% and 2020 NV dem margin was 2.39%. Biden won the popular vote by a much larger margin than Hillary, but that might have been a fluke because of the huge increase in mail voting in other states.
Nevada went, relative to the nation, (i.e. nevada margin minus national margin) from D+5.3 (08) to D+2.8 (12) to D+0.3 (16) to R+2.2 (20). RCP currently has Trump +5.7 in NV vs Trump +0.9 nationally, or R+4.8 relative to the nation. This is basically agnostic to candidates. An average of polling crosstabs show a 13 point shift to the right among hispanic voters after they already shifted a lot in 2020 and a 12 point shift among 18-29 voters.
It's of course entirely possible that NV stops trending red, that Biden wins the popular vote by a few points, that the crosstabs and state polls are all wrong, and these are dependent probabilities, but I certainly don't think it's a tossup. I think it's ~70/30 for Rs.
@gpt_news_headlines also, poly has 60/40 for dem/gop in nevada. https://polymarket.com/event/nevada-presidential-election-winner?tid=1714085675905 largely I think on the back of the above. Sharps are selling at 60c!
I can only assume that for the past 18 months a crack team of democratic party operatives appointed directly by Shadow President Gretchen Whitmer have been silently pulling strings across the nation to make sure that abortion rights initiatives are on the ballot in every swing state.
Now the dominos finally begin to fall.
Trump lost Nevada by 2.5% in 2016 and 2020. Nevada is only going to trend blue-ward because of immigration. Polls use complicated adjustments for who's less likely to pick up the phone and who's more likely to actually vote. Regime shift could thow off those models. And there's a lot of time between now and the election for trump to embarrass himself and for all the mainstream media to propagandize for Biden.
@JonathanRay minorities, especially males, are moving towards Trump:
Source: https://www.wsj.com/politics/elections/black-voters-show-signs-of-slipping-away-from-biden-in-2024-69ae78b0
Trump cannot embarrass himself more than he has done so far, his support is immune to embarrassment, however Biden has a lot of space for more gaffes.
@JonathanRay latest poll from WSJ has Trump +4:
The source has other interesting information that goes against Democrats, for example voters in Nevada are less inclined to vote, which tends to benefit Republicans: https://www.wsj.com/politics/elections/trump-biden-poll-swing-states-ad594acb
@egroj Doesn't lower turnout benefit Democrats now? Republicans no longer have the high-propensity base. Though maybe the Democratic base in Nevada is lower propensity than other states because they rely heavily on minority voters.
@PlasmaBallin I think Trump voters/fans are going to go vote no matter what, so I was assuming that the people uncertain to vote are overrepresented by Dems. But I don't have data to back-up the claim.
@PlasmaBallin The polls have been really weird this cycle when you look at crosstabs. It's stuff like Trump picking up substantial numbers of black voters and <25 voters. It's really bizarre.
@Najawin i think robocalls are so common nowadays that the vast majority of people have their phones set to block all unknown callers, which is probably a PITA for pollsters
@Najawin if you believe crosstabs in direction but not in magnitude (as you should) everything makes sense. it also fits with 2020 results and with primaries.
@SemioticRivalry I don't believe black people are trending towards Trump, no.
Also, no, things can't make sense without these magnitudes. These magnitudes are why Trump has the lead he does on Biden in these polls. You simply can't separate the two. If we believe that maybe Trump is picking up small amounts of support among these two demographics rather than large inroads, we simply don't believe that Trump is consistently +3-+5 above Biden, because that sort of gap requires us to care about the magnitudes of crosstabs. Otherwise we get a substantially closer race. (And, yes, yes, margin of error. But everyone is ignoring that in saying "Trump +4". If what we actually mean is "statistical dead heat", we should say that.)
@Najawin black voters shifted towards Trump by around 8 points in 2016 and then again in 2020 by around 6 points, whether you believe it or not. https://catalist.us/wh-national/ It doesn't seem hard for me to imagine that it will happen again. Crosstabs have Biden + 54 with black voters. If I had to place a guess, it'd probably be around ~Biden + 70, but that's still a loss of ten points relative to 2020. If you don't believe in crosstabs, please come bet with me.
things can definitely make sense even if you don't believe the magnitudes of black/youth shifts, because crosstabs also show massive shifts in Biden's direction. I also believe the direction (but not the magnitude) of the senior and white voter shift in Biden's favor. these shifts don't make eye catching headlines but are the reason that polls are even close right now. If you don't believe e.g. Biden's strength with seniors (which was a complete polling artifact in 2020) then Trump is being underestimated in the toplines by an entire percentage point.
The reason I believe in the direction of the shifts is because this is remarkably consistent across polls, none of these shifts have been present in polling before, and they fit with actual election results that we have seen. Dems did very poorly with young voters in the CA primaries and quite well with older voters.