Here we have a new Quinnipiac poll, one of two that came out yesterday, and it’s a bit of a puzzle.
In Texas and Ohio, two states where President Trump won easily in 2016, the president holds a slight lead in Texas and it’s too close to call in Ohio, according to a Quinnipiac (KWIN-uh-pea-ack) University poll of likely voters in both states. These are the first surveys from the Quinnipiac University Poll in both Texas and Ohio to use likely voters and results cannot be compared to prior surveys of registered voters.
“With six weeks to go until Election Day and most minds made up, Ohio could hinge on a sliver of likely voters who signal they may have a change of heart and the four percent who say they are unsure right now who they’ll back. At this point, it’s a toss-up,” said Quinnipiac University Polling Analyst Mary Snow.
“It is close but leaning toward Trump in Texas. There are still a slim number of likely voters who are undecided or on the fence about their choice, which could leave just enough wiggle room for either candidate to take Texas’ many electoral votes,” said Quinnipiac University Polling Analyst Tim Malloy.
MIND MADE UP
In Ohio, 97 percent of likely voters who selected a candidate in the presidential match up say their minds are made up, with 3 percent saying they might change their minds. In Texas, 94 percent say their minds are made up, with 5 percent saying their minds might change.
VOTING IN 2020
In Ohio, 46 percent of likely voters plan on voting in person on Election Day. Thirty-five percent plan on voting by mail/absentee ballot, and 16 percent plan on voting at an early voting location.
In Texas, 47 percent of likely voters plan on voting at an early voting location, 38 percent plan on voting in person, and 13 percent plan on voting by mail/absentee ballots.
TEXAS: BIDEN VS. TRUMP
Likely voters in Texas give President Trump a mixed favorability rating, with 49 percent saying they have a favorable opinion of him and 47 percent saying they have an unfavorable opinion.
Former Vice President Biden has a negative favorability rating among likely voters in Texas, 41 – 52 percent.
Trump has clear leads in three of five categories among likely voters when asked who would do a better job handling issues:
On handling the economy: Trump 58 percent, Biden 39 percent;
On handling the military: Trump 52 percent, Biden 45 percent;
On keeping your family safe: Trump 52 percent, Biden 44 percent;
On handling the response to the coronavirus: Trump 49 percent, Biden 47 percent;
On handling racial inequality: Biden 50 percent, Trump 45 percent.
TEXAS: TRUMP APPROVALS
Likely voters are divided on the way Trump is handling his job as president, 50 – 48 percent, and are similarly split on his handling of the response to the coronavirus, 49 – 49 percent.
TEXAS: SENATE RACE
In the race for the U.S. Senate where incumbent Republican Senator John Cornyn is seeking a fourth term, Cornyn leads Democratic challenger MJ Hegar 50 – 42 percent. Eighty-four percent of voters say their minds are made up, while 15 percent say they may change their minds.
Thirty-nine percent of likely voters have a favorable opinion of Cornyn, 30 percent say unfavorable, and 30 percent say they haven’t heard enough about him. Twenty-nine percent of likely voters have a favorable opinion of Hegar, 19 percent say unfavorable, and 50 percent say they haven’t heard enough about her.
The Texas crosstabs are here. This is the best poll Trump has had in awhile, the first I can recall where he’s reached fifty percent, and a six-point improvement for him over the July Quinnipiac poll, in which he trailed Biden 45-44.
All of that is straightforward and somewhat ominous for Biden, but a peek under the hood raises some questions about what these numbers mean. To illustrate, let me compare some of the subgroups from this poll to those same groups from that July poll, for which that data is here.
July July Sept Sept Group Biden Trump Biden Trump ========================================== Men 39% 48% 41% 55% Women 49% 40% 50% 46% GOP 6% 89% 6% 91% Dem 94% 3% 95% 4% Ind 51% 32% 51% 43% 18-34 46% 32% 56% 42% 35-49 48% 40% 40% 56% 50-64 43% 52% 44% 51% 65+ 42% 53% 46% 50% White men 28% 61% 30% 67% White women 31% 62% 43% 53% Black 89% 6% 79% 19% Latino 53% 29% 51% 43%
Let’s just say, there are some mighty big swings, in both directions. I’m not exactly sure how one could coherently account for all of them. I feel quite confident saying that Donald Trump will not get nearly 20% of the Black vote – every other poll tops him out at nine or ten percent, which I think is a tad high but plausible – and I have no idea how the 35 to 49 contingent could go from being a decent Biden plurality to a significant Trump majority. By the same token, Biden cutting a thirty-one point deficit among white women to ten points seems like a stretch. The most likely explanation in all this is some small sample size weirdness, and as such it’s not worth putting too much energy into trying to figure it all out. It is what it is, and if we’re lucky Quinnipiac will do an October poll, which will either see things revert back to what we have mostly seen before, or present us with more of a puzzle. I don’t know what else to say.