In a hypothetical 2024 presidential election between President Biden and former President Trump, 47% would vote for Biden and 40% for Trump.
Once again, this screams nonresponse bias.
This is a Biden +26 sample, so I'm not so sure.
If anything, it suggests the undecided vote (in the Presidential match-up at least) leans Dem, as they often do in blue states.
IDK, something still seems very odd - You have 15% of Biden 2020 voters who would vote for... Trump in the 2024 rematch?
Their Democrat sample also has 21% of Dems voting for Trump in 2024 and nearly the same voting for Zeldin this year.
And maybe I'm reading it wrong, but only 17 people in their sample are "did not vote" in 2020? Does that mean there is barely any either non-2020 or new voters in this sample?
Even the “Does the FBI search make you more or less likely to support Trump in 2024” and 24% of Democrats say “more likely” (!!), and in the overall sample, in New York of all places, there is a higher total for “More likely” (35%) than “Less Likely” (34%). In a Biden +26 sample?!
IDK, maybe I'm totally off here. But that all just seems very odd.
Buddy there have been several polls of this race and others and you seem to keep trying to dive into the cross tabs to find something to discredit them/ confirm your priors. Why are you even in the polling board? It's not changing your opinion on any of the races? You have a theory on this midterm and are sticking with it. Save your self the deep dive on poll crosstabs.
As for some of your specific problems this time. With the large caveat to mostly ignore poll subsamples.
1. First time voters.- A very small percentage of the American population is first time eligible. Historically (ie in literally every election since I have been alive) more people turn out in presidential elections than in midterms. I understand you are going to counter with
Dobbs but a large majority of people motivated by that decision
hated Donald Trump and already turned out in both 2018 and 2020. Opposition to Lee Zeldin and a smattering of house races don't inspire as much as opposition to Trump.
New York's contested gov primary occurred after Dobbs leak and had less turnout than 4 years prior (2018) and the 2020 presidential primary that occurred after Sanders dropped out.
2. Non response bias- the last several cycles it there has been a GOP non response bias. You think this has switched. Polls won't ever show it by their nature. This does not always have to favor the GOP but I personally don't see how Democrats have become harder to reach in the past two years. You may be right and if so you can say you called it.
3. Given Biden's approval the fact he would lose some 2020 Biden voters in a hypothetical rematch with Trump is unsurprising. This forum does not have many swing voters as the super informed election nerds tend to be hyper partisan. This does not mean he can't get those voters back in 2024.
These voters often vote against the party in power. Some to "punish" the presidents party, which has disappointed them, others because their presidential vote was more "against the other guy" than for the president's priorities and thus they naturally prefer a divided gov.
