For those who look for patterns in the tea leaves, the most salient data point from the recent JMC Analytics Senate election poll is not the reversion to Alabama’s partisan “mean”. Moore has reclaimed the solid lead he held when JMC Analytics first polled the state at the end of September.
In both surveys, Moore pulls 48% of the electorate. However, additional data points should give pause to Roy Moore’s assumption that Alabama’s broad and deep Republican base will unconditionally stand by their man.
Tale of the Undecideds
Moore is standing still. Since the late-September poll, the percentage of undecided voters has fallen from 11% to 5% of those surveyed. While Moore’s steadies himself at 48% of the vote, he has picked up none of this new set of voters who have moved from the ranks of the undecided to decided.
Doug Jones has gained half of this 6% who count as newly decided, with the other 3% going to other marginal candidates. Among the 5% who remain undecided, 49% favor Moore and 44% favor Jones, but given the small sample set (about 30 respondents), that tally has no significance.
Tale of the Unqualifieds
When filtered without the partisan lens, 48% of those polled say Doug Jones is qualified to serve as Alabama’s U.S. Senator based on the campaign he has run so far, with 40% sayiing he is unqualified. In late September, 50% of those surveyed believed Roy Moore was qualified, with only 35% labeling him as unqualified. In the most recent survey, 49% of likely voters believe Moore is qualified, with 46% now deeming him unqualified.
As with the tale of the Undecideds, Moore’s inability to rise above 50% of the electorate when queried about his qualifications indicates a creeping dry rot to his support. His unqualified ratings are presently 6% higher than Doug Jones. Indeed, the gap between Moore’s qualified and unqualified percentages has essentially remained unchanged since the sexual allegations surfaced, even as the percentage of voters who say they will vote for him has recovered lost ground.
Tale of the Tabs
JMC crosstabs for each question are also illuminating, indicating that the election pivots most sharply on assumptions made about the strength of the evangelical vote in the state. In the first JMC poll, 53% of those surveyed said they were evangelical Christians. In the most recent poll, 59% assigned themselves evangelical status.
This data is also interesting, and not necessarily favorable to Moore, as the correlation between those identifiying as evangelical and those who support Moore has shifted dramatically south, even accounting for his recovery in absolute support numbers.
Consider the table below.
We can create ratios of “evangelical intensity” percentages in relation to support, qualified, and unqualified percentages for Roy Moore across the 3 JMC polls. The proximity of these numbers to each other indicates an “evangelical intensity” percentage – how closely, in other words, support for Moore maps to the self-identified evangelical cohort in the state.
The numbers are revealing. Between Poll 1 and Poll 3, spanning the pre-predator” and “post-predator” phases of the election campaign, the percentage of self-identified evangelicals increases from 53% to 59% At the same time, Moore’s “support” and “qualified” percentages remain stable, while his “unqualified” percentages increase dramatically, from 35% to 46%.
In relation to evangelical sentiment, the evangelical intensity of Moore’s “support” slips from 91% to 81%. The evangelical intensity of his “qeasure increases from 66% to 78%. None of these numbers indicate a stable electoral base for Moore heading into the last 10 days prior to the election.
Tale of the Turnout
Steve Bannon, styling himself as Robespierre, plans to return to Alabama next week to rally the base on behalf of Roy Moore. We have seen this picture before, and can only hope, as Karl Marx wrote about the 18th Brumaire, that while the 1st instance of Bannon’s intervention plays out as tragedy, the 2nd instance next will play out as farce.
But clearly turnout will be key, and it is notable that the demographic groups who vote less reliably – younger voters, in particular – are underweighted in this recent landline survey. Mobilizing young voters and African-American voters and women will be the pivot upon which this election turns.