Is the big U.S. Senate race in Arkansas tied? Or is Tom Cotton well out front in a head-to-head match among “definite” voters, whatever those are?
But is Mark Pryor ahead if all potential voters and the full field, including peripheral candidacies that include a Libertarian and a Green and even an independent candidate not actually known to be running, are polled?
And does a generic Republican, meaning a nonexistent one, perhaps a piece of cardboard appearing on the ballot beside an “R,” poll better against Pryor than the really existing—if not effervescently existing—Cotton?
Allow me to answer all those questions this way: Yes, according to a recently released poll.
It depends not so much on how you look at it as how you choose to slice and dice it, whether for tactical purposes or self-pleasing ones or both.
A score in the process of a sports contest is not subject to interpretation. The score is what it is—Alabama 38, Arkansas 13, let’s say.
But, in politics, there is no real score until actual votes are cast and counted. So the polls taken along the way amount to whatever you can spin from them. And who is to argue? No one has actually voted yet and been counted. So you can’t be proven empirically wrong.
In hotly contested, high-stakes political races, of which we confront a couple in Arkansas, the competing partisans crave a lead in polls.
It buoys them. It makes them think they will win. It enables them to discredit the other guy’s strength. It opens the spigot for campaign donations. It offers potential value as a self-fulfilling prophecy. It gives a campaign a triumphant-seeming narrative.
So whenever a new poll comes out, the guys on top pounce to brag on themselves and disparage the other guys, and the guys on bottom pounce to dismiss, dispute and discredit the survey.
Leaders in polls say, “See? Told you so.” Laggers in polls say, “Yeah, but …”
Bogus poll, they’ll say. Bad methodology, they’ll say. Skewed sample, they’ll say. Partisan pollster, they’ll say. Idiot pollster, they’ll say.
It’s mostly a matter of adapting and reshaping and restating basic facts to accentuate the positive and mitigate the negative—or, in some cases, accentuate whatever remote positivity can be imaginatively woven.
Never, however, have I encountered a poll so divergently touted or contradictorily spun as the one publicized Monday in the big Pryor-Cotton race.
The best way to tell the story would be chronologically.
I heard of a poll showing the race 46-46, which would be good news for Pryor, the incumbent laden with Obamacare who is generally believed to be slightly behind in a state that deplores Obamacare.
Politico, the eminent national political website, reported the poll Monday morning. It indeed showed a 46-46 tie, based on 400 calls in the state in mid-February.
I got a copy of the survey and found that it was part of a multistate poll conducted by the commission of an energy group to assess attitudes toward the Keystone Pipeline and natural gas fracking. The purpose was to determine how those attitudes might affect Senate races in battleground states.
Republicans initially dismissed the poll, noting that the firm conducting it, Hickman Analytics Inc., is run by Harrison Hickman, who was the regular pollster years ago for Mark Pryor’s dad, David.
But then Republicans further read the complete questionnaire and full findings and decided the poll was fine.
Indeed, Hickman does not work for Mark Pryor. He surely would be loath to dummy up numbers buoying the son of a former client in a survey paid for by energy interests seeking worthwhile data in a context including Arkansas, but also far transcending it.
What Republicans found they liked about the poll was that the 46-46 deadlock was produced by all potential voters. When screened for “definite voters,” the respondents favored Cotton 51-42.
U.S. News ran with that angle and declared on its blog that Cotton led Pryor by 51-42 in Arkansas … unless Pryor and the Democrats could produce a big turnout by which the 46-46 finding might become relevant.
Pryor partisans countered as follows: Yeah, but how about that horse-race finding when the respondents were confronted with the full field of candidates, including the peripheral ones—Green, Libertarian, independent? In that context, Pryor led Cotton 40-37.
But there is no known independent candidacy.
So, by the way, asked Pryor partisans: Did you see that Pryor was tied with Cotton 46-46, but well behind—47-39—when polled opposite a “generic” Republican?
They asked: What does that tell you?
The first thing it tells me is that it’s a tad pitiable to point to a finding that your guy would lose by eight points to a sheet of cardboard labeled “Republican.”
But the second thing it tells me is surely more to the Pryor partisans’ intended point: Cotton polls worse than a sheet of Republican cardboard because the definition of him as an eerie extremist—fully accurate—seems to be penetrating the consciousness of the Arkansas electorate.
So there you have it: The poll is good for Pryor, but bad for Pryor. It’s good for Cotton, but bad for Cotton.
What, then, is the real takeaway, the real story?
It’s that the race remains so close as to be effectively spun either way.
It’s that the race offers Cotton the general advantage owing to the rolling Republican tide in Arkansas. It’s that Cotton’s advantage appears to be eroding slightly as Democrats pound Cotton for votes against Arkansas farmers and student loans and disaster aid and keeping the government open.
Since we’re in March, the month of madness, let’s conclude by saying that, in the fine tradition of NCAA tournament basketball, this contest will be fun to watch throughout, but maybe not decided until the last shot.
John Brummett’s column appears regularly in the Arkansas Democrat-Gazette. Email him at jbrummett@arkansasonline.com. Read his blog at brummett.arkansasonline.com, or his @johnbrummett Twitter feed.