FacebookTwitterRedditEmail

“Electability” is Real (Unless Married with the Junk Science of Ideological Spectrum Analysis)

Drawing by Nathaniel St. Clair

If electability were dependent on how moderate a candidate is, or is perceived to be, Jimmy Carter would likely have won a second term. We also likely would have had Presidents Mondale, Kerry, McCain, Romney and President Hillary Clinton. Maybe also President Gore.

Rolling Stone’s Matt Taibbi is disgusted with the way “electability” was used to shout down Bernie Sanders in 2016 and has already been deployed against Elizabeth Warren ahead of 2020. Taibbi wants to deem the whole enterprise “alchemy and a crock.”

But this is the wrong take. Far more effective are the four simple words that drive PUMA people over the edge: Bernie would have won. (Party Unity My Ass or PUMA supporters of Hillary Clinton in 2008 refused to support Barack Obama in the general election for November.) Not only is it true that Bernie Sanders would have won, as even the most grudging pro-Clinton data people had to eventually admit, but it is also true that a proper look at the data showed that Sanders would likely win, where Clinton would struggle against Trump, way back in the spring of 2016, well before the conventions.

For those who have not previously followed my work, I have been writing on polling, elections data, and politics for the last three years here at CounterPunch, using my #10at10 model to analyze and forecast elections. I made the case that Sanders was more electable in May 2016 in an article entitled “Electoral Votes Matter: Hillary Clinton or Bernie Sanders vs. Donald Trump.” Pennsylvania, Wisconsin, and Michigan as well as Minnesota and several other states were vulnerable to a Clinton loss to Trump at that time while Sanders was comfortably ahead by 10% or more in the average of polling in those four states, as well as in Arizona. He was ahead by 5% or more in enough states to go well over 300 electoral votes. Princeton’s Sam Wang, extending Christopher Wlezien and Robert Erikson’s work, ran the math showing that polling is fairly predictive in the spring of an election year, becomes more wobbly in the summer, and returns to form after Labour Day. (Wang had to issue a correction after wrongly calculating what that truth meant for Clinton’s electability at the time.)

The problem with data arguments about electability in the U.S. is that a whole school of thought, from establishments in parties to famous data people who should very well know better, has insisted that a candidate closer to the center is more likely to win. People with this view often present elaborate charts and measurement tools to make the case as to who is more electable. This “ideological spectrum analysis” is, at least when married to electability arguments, a junk science and is rooted in the flawed assumption that the electorate is basically polarized along party lines and that candidates compete for centrists who identify as independents. This view of independents as centrists to be wooed has been debunked over and over and over, but persists anyway. Again, Presidents McCain, Romney, and Hillary Clinton roundly approve of this confusion!

So how do we evaluate electability on the basis of real data if ideology is not critical?

There are three key factors that can help project how an election will go many months in advance – not in rigid detail, things can always change – but in fairly predictive broad strokes. Will one candidate or party mostl likely trounce the other or others? Or will it be a close race? The three factors, fairly consistent as measured in multiple nation states, are (1) the average of early polling where present, (2) favorability or approval ratings, particularly of the candidate or leader of a party, and (3) economic confidence. The third factor here is especially attached to the incumbent party or candidate.

Last week, my column noted that going back to 1976, candidates polling first or second in the first half of a year prior to election have won eight of the last nine GOP primaries and six of nine Democratic primaries with a seventh (Dukakis) winning from the third place position at this point in the cycle. This week’s update sees Kamala Harris steadily continuing to close the gap, but remaining solidly in third place behind Joe Biden and Sanders.

Upcoming weekly columns will look at polling versus Trump and favorability in more detail, but here is a quick peek at where things stand on favorability for the top six candidates as compared to where Hillary Clinton’s and Bernie Sanders’ net favorability travelled across the 2015 landscape. I will continue updating this chart over the remainder of the year, leaving the dotted lines showing Clinton’s and Sanders’ average net rating at the same point in the cycle. Net favorability, it should be noted, is fairly independent of name recognition; Sanders’ name recognition remained below or near only 50% well into the Fall of 2015, changing with the Democratic debates.

Our quick look here shows why it is not surprising that Biden and Sanders remain #1 and #2 with likely Democratic primary voters who have consistently identified the ability to beat Trump as the most important quality in a candidate heading into 2020.

Doug Johnson Hatlem writes on polling, elections data, and politics. For questions, comments, or to inquire about syndicating this weekly column for the 2020 cycle in your outlet, he can be contacted on Twitter @djjohnso (DMs open) or at djjohnso@yahoo.com (subject line #10at10 Election Column).

FacebookTwitterRedditEmail