Predicting Turnout in an Unpredictable Race
It’s clear that Hillary Clinton is winning. Beyond that, things get tricky.
by Charlie Cook
Aug. 22, 2016, 8:01 p.m.
This presÂidÂenÂtial race seems to have staÂbilÂized at a point where the probÂabÂilÂity of HilÂlary ClinÂton beatÂing DonÂald Trump is very high, but the marÂgin is tough to preÂdict—as is the rate of voter turnout.
Sure, there could be an exoÂgenÂous world event, terÂrorÂist atÂtack, or ecoÂnomÂic calamÂity that could change the dyÂnamÂics of this race, or there could be deÂvelÂopÂments inÂvolving the canÂdidÂates themÂselves that could change the traÂjectÂory of this elecÂtion. But for every chalÂlenge ClinÂton faces, there is at least an equal one for Trump; think of ClinÂton’s emails versus Trump’s inÂcome-tax reÂturns.
First, look at the naÂtionÂal horse-race numÂbers: UsÂing the RealÂClearÂPolitÂics.com poll avÂerÂages, ClinÂton leads by 5.5 points in the two-way triÂal heat, 47.0 perÂcent to Trump’s 41.5. On the four-way, inÂcludÂing LiberÂtariÂan Gary JohnÂson and Green Party nomÂinÂee Jill Stein, ClinÂton is at 41.6 perÂcent to Trump’s 37.1, a lead of 4.5 points. Leads this long after the two conÂvenÂtion bounces have settled genÂerÂally hold up.
Then look at the swing states, again usÂing the RCP avÂerÂages. ClinÂton leads in every single state that PresÂidÂent Obama carÂried in 2012 and runs baÂsicÂally even in AriÂzona and GeorÂgia. If a DemoÂcrat can win AriÂzona or GeorÂgia, he or she doesn’t need to; it means the elecÂtion is already won, as othÂer states would have come across the finÂish line earliÂer. A ClinÂton win in either state would just be icing on the cake. And if a DemoÂcrat is feelÂing comÂfortÂable enough to scale back efÂforts in ColÂorÂado and VirÂginÂia, that’s quite something, givÂen both states’ reÂcent hisÂtory of close conÂtests.
Third, look at how the canÂdidÂates are perÂceived, again usÂing the RCP avÂerÂages of all of the maÂjor naÂtionÂal polls. ClinÂton has 43 perÂcent viewÂing her faÂvorÂably, 53.5 perÂcent unÂfaÂvorÂably, for a net-minus 10.5 points—deadly in any othÂer race, but not against Trump’s 33 perÂcent faÂvorÂable, 62.7 perÂcent unÂfaÂvorÂable, for a toxÂic net-minus 29.7 points. It’s awÂfully hard for someone with negÂatÂives that high to turn a race around.
FiÂnally, think about the camÂpaigns: One side acÂtuÂally has one; the othÂer has outÂsourced its esÂsenÂtial funcÂtions to the ReÂpubÂlicÂan NaÂtionÂal ComÂmitÂtee. If the race were to get close, havÂing a full-fledged field orÂganÂizÂaÂtion could well make the difÂferÂence of a point or 2 in critÂicÂal states.
The fact that withÂin two weeks of Labor Day, GOP canÂdidÂates are deÂbatÂing the etiquette of exÂactly when it would be apÂproÂpriÂate to jump ship makes it pretty clear where the elecÂtion is headed unÂless something cataÂclysÂmic hapÂpens.
Much has been said about ClinÂton and Trump havÂing the highest unÂfaÂvorÂable ratÂings of any presÂidÂenÂtial nomÂinÂees in hisÂtory, but few seem to have conÂsidered what that really means. More AmerÂicÂans will be votÂing for a canÂdidÂate they really don’t like or trust than ever beÂfore. There will be voters who are not fans on Trump who will vote for him anyÂway beÂcause they can’t bring themÂselves to vote for ClinÂton. Polls sugÂgest there are many more who don’t parÂticÂuÂlarly like ClinÂton but can’t posÂsibly vote for Trump. We could have an elecÂtion with alÂmost as many people votÂing against someone as for someone.
With such a large pool of amÂbiÂvalÂent or unÂenthuÂsiÂastÂic voters, gauging turnout will be parÂticÂuÂlarly probÂlemÂatÂic. Many are sugÂgestÂing an unÂusuÂally or hisÂtorÂicÂally low turnout, but my own theÂory is that we are more likely to have an avÂerÂage to high level of voter parÂtiÂcipÂaÂtion. The inÂtensÂity of the hatred for one or the othÂer is so high among so many that the opÂtion of reÂfrainÂing from votÂing at all seems abÂhorÂrent to many of these voters. Some can’t see themÂselves sleepÂing at night if Trump were to win and they hadn’t voted against him; othÂers feel the same way about ClinÂton. GivÂen that elecÂtion ofÂfiÂcials do not sepÂarÂate the balÂlots of those votÂing against someone from those votÂing for someone, a vote is a vote, no matÂter the motive.
The polling is clear that AmerÂicÂans are watchÂing this elecÂtion closely, and that while many are unÂenthuÂsiÂastÂic about their choices, there is a strong senÂtiÂment that the outÂcome is imÂportÂant, that who wins matÂters. That points to pretty good turnout, even if voters don’t have a real spring in their step walkÂing up to the polling place or mark their balÂlot with a flourÂish.
In conÂverÂsaÂtions with pollÂsters and camÂpaign opÂerÂatÂives, it’s inÂterÂestÂing to deÂvelÂop proÂfiles of what kind of voters are or are not likely to be Trump supÂportÂers. Trump backÂers are far more likely to be white than nonÂwhite, they are more likely to be male than feÂmale, they are more likely to have less than a colÂlege deÂgree, and are more likely to be 45 years of age or older. But less disÂcussed is that they are more likely to live in small-town or rurÂal AmerÂicÂan than in or anyÂwhere near a city.
I realÂize that someone is about to point out to me a young, AfricÂan-AmerÂicÂan or Latino woÂman with a PhD livÂing in a three-story brownÂstone walk-up who loves Trump, but those are the exÂcepÂtions, not the rule. InÂdeed, among voter groups that comÂbine sevÂerÂal atÂtribÂutes—white, male, no colÂlege deÂgree, over-45, and livÂing outÂside urbÂan areas—Trump will likely over-perÂform Mitt RomÂney, while greatly unÂderÂperÂformÂing RomÂney among colÂlege-eduÂcated whites, parÂticÂuÂlarly those who are woÂmen.
DemoÂgraphÂicÂally, Trump has painted himÂself inÂto a corner. That doesn’t mean that there is no way for him to esÂcape that corner, but it will reÂquire conÂsidÂerÂable athÂletÂic abilÂity and agilÂity to pull it off—more than he now apÂpears to have.
https://www.nationaljournal.com/s/640792?unlock=XBKX9FQVTAWR4EKW&mref=homepage-free
| 
