3) Casey somehow losing Latinos by double digits in the LV screen (goes from Casey+5 RV to +2 overall), continues to suggest big assumptions about the type of people pollsters think show up in Nov.
Which, coincidentally, aligns with least likely voters being pro-Trump.
(2/3)
FWIW, today’s NYT/Siena PA poll rests on some key metrics.
1) Biden wins Philly 54-30, again suggesting a generational collapse in the black vote.
2) Suburbs shift right from ‘20 despite trending left in every cycle since then, as recently as a few months ago.
(1/3)
I’m grateful to Siena for doing this poll and am happy with them weighing those regional subgroups, which is precisely why some of those numbers smell fishy to me.
It’s still a long way to Election Day. Let’s see if this continues in September (I suspect not).
(3/3)