|Photo courtesy of Club For Growth TV ad.
A new poll in Arkansas from the Democratic leaning pollster Public Policy Polling puts incumbent Senator Mark Pryor ahead of his Republican opponent Congressman Tim Cotton by just one point, or 43-42%, with 16% undecided. This result might be interpreted by some as impressive for a Democratic Senator in a deep-red state, especially considering the conventional wisdom around the race for much of the last year as one of the Republican Party's likeliest Senate pick-up opportunities.
But not so fast. Tom Jensen tweeted out this crucial bit of information following the official poll release:
Obama's overall approval in Arkansas is 33/61. Among voters undecided for Senate it's 13/74: http://t.co/6dS4ewv60o
— PublicPolicyPolling (@ppppolls) April 30, 2014
In a race where both major party candidates are polling in the lower 40s, the undecided vote could dramatically alter the outcome. And there are several data points that would assist in providing clues about where those undecided voters will ultimately end up. Are they Democrats or Republicans? Young or Old? Liberal or Conservative? Among PPP respondents who said they voted in the 2012 election, it's the Romney voters that remain more undecided than Obama voters (17% to 8%). That's an obvious plus for Tom Cotton (R). Conservatives are more undecided than liberals (16% to 12%), another plus for Cotton. A full eighteen percent of Independent voters remain undecided (more than Republicans or Democrats), and they support Cotton over Pryor 50-31%. Yet at the same time, women, who support Pryor by greater margins than men, are more undecided than men. African Americans, as well as younger voters, both of which being groups more likely to support Pryor, are more undecided than whites or older voters.
But perhaps more important than any of these stats regarding undecideds is what Tom Jensen tweeted above - the fact that the President's approval rating is upside down with these voters by 61 points!
Unfortunately, PPP didn't provide a crosstab of how poll respondents that approved or disapproved of Obama's job performance said they would vote in a Pryor vs. Cotton match-up. But pollsters that have provided such info in the past would note that there's a very close correlation between a voters feelings towards the President's job performance, and how he or she may cast a vote in a federal partisan contest.
Suppose those sixteen percent of Arkansas voters that said they were undecided between Mark Pryor and Tom Cotton supported the two candidates by the same proportion they approved or disapproved of President Obama? Or in other words, suppose 13% of those undecided voters end up supporting Mark Pryor, while 74% end up supporting Cotton. How would PPP's final result have looked?
Senator Mark Pryor's one point lead over Tom Cotton evaporates into an eight point DEFICIT in the event undecided voters break for the candidates in a proportion identical to their approval of Obama's job performance. What was a 43-42% Democratic lead becomes a 53-45% Republican lead.
Obviously, this is purely speculative. But the big news from this PPP poll shouldn't be the fact that Pryor's clinging to a lead. It should be that if the crosstabs are to be believed, Pryor's lead is fleeting.