Joseph Sohm /

Forget the 2016 Polls: Nobody Knows Anything Yet

The campaigns and the political media are fixated on the presidential race—but actual voters aren’t.

Ima­gine in­ter­view­ing for a top-level job where the hir­ing com­mit­tee hasn’t got­ten around to read­ing your re­sume, but HR keeps polling its mem­bers any­way to see if you should be brought back for the next round of in­ter­views—and all the while, you’re sink­ing to­ward bank­ruptcy.

Even worse, ima­gine that the hir­ing com­mit­tee is in­stead lean­ing to­ward a TV game show host and a mo­tiv­a­tion­al speak­er, neither of whom has any ob­vi­ous rel­ev­ant ex­per­i­ence for the job.

A dozen Re­pub­lic­an pres­id­en­tial can­did­ates don’t need to ima­gine any of this. They’ve been liv­ing it most of this year—the po­ten­tial vic­tims of sur­veys that shouldn’t ac­tu­ally mat­ter yet.

“The polls are not pre­dict­ive of where we’re go­ing to be in three months,” said Lee Mirin­goff, dir­ect­or of the Mar­ist In­sti­tute for Pub­lic Opin­ion at Mar­ist Col­lege. “These voters right now on the Re­pub­lic­an side are not baked in.”

This type of ana­lys­is has offered little com­fort to can­did­ates like one-time front-run­ner Jeb Bush, whose cam­paign staged a donor sum­mit last month to calm con­trib­ut­ors alarmed at the state of the con­test. “Race will re­main flu­id for some time be­cause … voters have A.D.D.,” read one slide in the cam­paign’s present­a­tion.

Three cur­rent and former gov­ernors, in­clud­ing Bush, lan­guish in single di­gits. Three oth­ers, once con­sidered ser­i­ous con­tenders, have already dropped out. And all of this has come weeks and months be­fore the first bal­lot is cast, and be­fore tens of mil­lions of dol­lars in ad­vert­ising—much of it TV at­tack ads—is un­leashed that will al­most cer­tainly re­shape the field.

Atop the polls, mean­while, sit de­veloper-turned-real­ity-TV-star Don­ald Trump and re­tired neurosur­geon-turned-in­spir­a­tion­al-au­thor Ben Car­son, both of whom have rid­den celebrity and haven’t bothered with ser­i­ous cam­paign struc­tures.

Mirin­goff said the size of Trump’s crowds has more to do with people’s de­sire to see an en­ter­tain­ing TV star in per­son. “It’s a hap­pen­ing event,” Mirin­goff said. “That and go­ing to caucus are dif­fer­ent levels of com­mit­ment.”

At the same point in the 2012 race, just over two months be­fore the Iowa caucuses, pizza-chain ex­ec­ut­ive Her­man Cain had a clear lead in Iowa, while even­tu­al win­ner Rick San­tor­um was at 4 per­cent.

“His­tory sug­gests that these can­did­ates don’t usu­ally win,” said Mi­chael Dimock, pres­id­ent of Pew Re­search Cen­ter, adding that the ap­par­ent suc­cess of “out­siders” at this point in the pro­cess is pre­cisely be­cause voters really don’t have that much hard in­form­a­tion about the can­did­ates yet.

“There’s just way too much noise in the sys­tem right now,” he said.

A lot of that noise is in­her­ent to polling in primary elec­tions. While the can­did­ates and the polit­ic­al con­sult­ing world pay close at­ten­tion to news cov­er­age about the pres­id­en­tial race, the av­er­age Amer­ic­an does not—even in early-vot­ing states such as Iowa and New Hamp­shire.

This is re­flec­ted, poll­sters say, in large per­cent­ages of re­spond­ents who say they still have not settled on a can­did­ate. Dimock said most ac­tu­al voters right now have per­haps a passing in­terest in the cam­paigns: “That’s a little bit dif­fer­ent from start­ing to think about your choice.”

Layered onto this fun­da­ment­al lack of deep voter in­terest are the lo­gist­ic­al dif­fi­culties in mod­ern polit­ic­al polling. More and more Amer­ic­ans do not have home land­lines any­more, only cell phones. And those num­bers, by law, must be manu­ally dialed, driv­ing up costs. The ma­jor­ity of Amer­ic­ans, re­gard­less of what type of phone they have, do not an­swer in­com­ing num­bers they don’t re­cog­nize. These factors pro­duce a re­sponse rate in sur­veys of 8 per­cent, com­pared to 80 per­cent or so a few dec­ades ago.

And then there are the sample sizes, of­ten so small that the mar­gins of er­ror are lar­ger than the spreads among a host of can­did­ates. An ABC News/Wash­ing­ton Post poll re­leased this week­end had Trump lead­ing na­tion­ally with 32 per­cent, Car­son in second at 22 per­cent, and then 10 can­did­ates ran­ging from Sen. Marco Ru­bio at 11 per­cent down to Sen. Lind­sey Gra­ham and former Sen. Rick San­tor­um at 1 per­cent.

But be­cause the sample size of 423 Re­pub­lic­an re­spond­ents pro­duces a 5.5-point mar­gin of er­ror, those 10 can­did­ates from Ru­bio to San­tor­um were stat­ist­ic­ally tied.

John Dick, founder of the polling and re­search firm Civic Sci­ence, said such de­pend­ence on ob­vi­ously im­pre­cise sur­veys is ac­tu­ally do­ing voters a dis­ser­vice. “It is cat­egor­ic­ally ir­re­spons­ible, in my opin­ion,” Dick said.

Now add in the even-big­ger chal­lenge of de­term­in­ing how closely the sample of those sur­veyed will re­flect the re­l­at­ively small uni­verse of those people who ac­tu­ally vote in primar­ies and caucuses. Poll­sters de­term­ine who is likely to vote in these con­tests primary by ask­ing them—which brings up the re­lated prob­lem of people who an­swer their phones and who agree to take a sur­vey and then tell the poll­ster they will vote when his­tory sug­gests they will not.

In the ABC/Wash­ing­ton Post sur­vey, for ex­ample, 63 per­cent of the re­gistered voters tak­ing the poll said they are “cer­tain” to vote in their state’s primary or caucus, while an­oth­er 14 per­cent said they would “prob­ably” vote. A Fox News pollre­leased on Sunday sim­il­arly had 79 per­cent of re­spond­ents say­ing they are likely to vote.

If 77 or 79 per­cent of re­gistered voters truly wind up vot­ing in their primar­ies, it would shat­ter turnout re­cords across the coun­try. In 2012, only 11 per­cent of those eli­gible to vote na­tion­ally voted in a primary or caucus. Among Re­pub­lic­ans, less than a third of those who turned out for nom­in­ee Mitt Rom­ney in the Novem­ber gen­er­al elec­tion had bothered show­ing up to an earli­er caucus or primary.

“There’s not a good match from likely voters in preelec­tion polls to those folks who are ac­tu­ally go­ing to turn out,” said Cliff Zukin, a Rut­gers Uni­versity polit­ic­al-sci­ence pro­fess­or and a past pres­id­ent of the Amer­ic­an As­so­ci­ation for Pub­lic Opin­ion Re­search. “The lower the turnout, the more er­ror. … So you have to take it with a grain of salt.”

(This phe­nomen­on of telling poll­sters the “so­cially de­sir­able” thing is not unique to vot­ing. Churches would be far more crowded each Sunday and char­it­ies would be sig­ni­fic­antly bet­ter fun­ded if those re­spond­ing to sur­veys were be­ing com­pletely hon­est. “A lot of it is hu­man nature,” said Pew’s Dimock. “People want to think well of them­selves.”)

None of these prob­lems would mat­ter that much if early polls were merely en­ter­tain­ment, fod­der for polit­ic­al news and the junkies who fol­low it. But un­for­tu­nately for many of the GOP can­did­ates, there is one oth­er group that closely fol­lows the im­pre­cise polling and its at­tend­ant news cov­er­age: the donors whose checks keep the cam­paigns go­ing.

“There is a de­sire from all of us, but par­tic­u­larly donors, to have a level of pre­ci­sion in polling that’s really im­possible,” Dimock said, adding that many donors see their con­tri­bu­tions as in­vest­ments. “You want to have an im­pact with it, which means not put­ting it on a los­ing bet.”

Donor angst, in fact, spurred the pro-Bush Right to Rise su­per PAC to start air­ing $1 mil­lion in na­tion­al cable TV ads on Fox News last month—ex­actly one day after the group’s strategist, Mike Murphy, was quoted in a Bloomberg art­icle ri­dicul­ing the idea of spend­ing money for na­tion­al TV ads this early in the primary sea­son.


Yet even calm­ing down donors skit­tish about weak polling is part of a nor­mal pres­id­en­tial cam­paign sea­son. What’s made this cam­paign dif­fer­ent is that im­pre­cise polling has also been used to gen­er­ate thresholds that have re­leg­ated can­did­ates to far-less-viewed sec­ond­ary de­bates.

The Mar­ist Poll’s Mirin­goff said it was that de­cision by the Re­pub­lic­an Na­tion­al Com­mit­tee and the net­works tele­vis­ing the de­bates that led him and his me­dia part­ner, Mc­Clatchy, to stop con­duct­ing horse-race polling for the sum­mer. “It was just a bad use of polls by the me­dia at a time when people didn’t know these can­did­ates at all,” Mirin­goff said. “We were happy not to be part of that.”

Of the sev­en can­did­ates in the first “kids’ table” de­bate in Au­gust, two have left the race en­tirely. An­oth­er three—Gra­ham, former New York Gov. George Pa­taki and former Vir­gin­ia Gov. Jim Gilmore—have over sub­sequent de­bates been dropped from even the lower-tier event.

At a re­cent can­did­ates for­um sponsored by the Re­pub­lic­an Party of Flor­ida, Gilmore called the sys­tem “ri­dicu­lous” be­cause low stand­ings were keep­ing the can­did­ates from par­ti­cip­at­ing in the main de­bates—which in turn re­in­forced the poor stand­ings in the polls. Gilmore even asked those at­tend­ing his Sun­shine Sum­mit speech to sup­port him where it really coun­ted: in pub­lic opin­ion sur­veys. “Start to tell the poll­sters, be­cause it seems to be all about polling, that you’re for Jim Gilmore,” he said.

He blamed both the tele­vi­sion net­works for giv­ing what he called un­lim­ited air­time to Trump for the pur­pose of boost­ing their rat­ings, which Gilmore said fur­ther lif­ted Trump’s stand­ing in the polls, as well as the RNC for al­low­ing it to hap­pen.

(The net­works and the RNC have de­fen­ded their formats as driv­en by the large field of can­did­ates and the im­possib­il­ity of in­clud­ing all of them in a single de­bate.)

“You have this cir­cu­lar thing go­ing on here where the polling na­tion­ally is de­cid­ing who gets in the de­bates,” Gilmore said at a news con­fer­ence fol­low­ing his speech. “And the polling na­tion­ally is de­cided by who the me­dia choose to pro­mote. And then they get on the stage, and they get more at­ten­tion and more poll num­bers.”


Re­gard­less of how things stand in late Novem­ber, though, both poll­sters and cam­paign strategists agree things will likely look very dif­fer­ent on the eve of the open­ing con­test in Iowa on Feb. 1.

At cam­paign events in Iowa and New Hamp­shire, a num­ber of those sup­port­ing Trump said they had not voted pre­vi­ously in caucuses and primar­ies. An op­er­at­ive from a rival cam­paign said Trump can­not win Iowa un­less he brings in vast num­bers of new voters to the caucuses, much the way Pres­id­ent Obama did in 2008. But un­like Obama, Trump has not in­ves­ted in the sort of or­gan­iz­a­tion needed to do that.

As for Car­son, he has in­ves­ted in hardly any cam­paign struc­ture at all. In his most re­cent Fed­er­al Elec­tion Com­mis­sion fil­ing, he re­por­ted spend­ing $169,000 for staff. In con­trast, Sen. Ted Cruz and Sen. Marco Ru­bio re­por­ted $563,000 and $592,000, re­spect­ively.

“Car­son’s got just noth­ing on the ground, ex­cept that he’s got evan­gel­ic­al sup­port that’s self-or­gan­iz­ing,” said Fer­gus Cul­len, a former chair­man of the New Hamp­shire Re­pub­lic­an Party.

Front-run­ner Trump, mean­while, is likely to see mil­lions of dol­lars, if not tens of mil­lions of dol­lars, of TV ads at­tack­ing him—an as­sault that will al­most cer­tainly drag down his num­bers. In­deed, the only time Trump de­clined in Iowa polls co­in­cided ex­actly with the peri­od the con­ser­vat­ive group Club for Growth ran $1 mil­lion worth of anti-Trump ads in Iowa this au­tumn. The su­per PAC sup­port­ing Ohio Gov. John Kasich has already said it will spend $2.5 mil­lion to at­tack Trump, and oth­er can­did­ates and out­side groups are likely to fol­low.

Un­less Trump de­cides to spend ser­i­ously from his own bank ac­counts—and so far he has not; his cam­paign has largely been fun­ded by pur­chasers of his hats and T-shirts—he could eas­ily lose both of the first two states, giv­ing fresh mo­mentum to those can­did­ates who do well there.

“His whole cam­paign is built on lead­ing in the polls,” Cul­len said. “So when he loses, I don’t know how he comes back.”

And if any can­did­ate has reas­on to hope for a Trump de­cline and a re­turn to nor­mal GOP pres­id­en­tial polit­ics, it is Bush, who has in­ves­ted heav­ily on the idea that turn­ing out sup­port­ers us­ing voter lists in the early-primary states is the way to win the nom­in­a­tion—and who has the most to lose if the race is in­stead based on celebrity and per­form­ance art.

“Nobody who leads the polls at this point is likely to be­come pres­id­ent or even their party’s nom­in­ee,” Bush’s top strategist Sally Brad­shaw said. “The one un­known is just how large the field re­mains after Feb­ru­ary. Can­did­ates will need re­sources to go the dis­tance and play in March. Jeb will have those re­sources, and we are pre­pared to play the long game.”

Pew’s Dimock is not ready to guess which can­did­ate or can­did­ates will emerge from the cur­rent field. At the same time, he is also not per­suaded that today’s state of af­fairs rep­res­ents any sort of fun­da­ment­al shift in pres­id­en­tial polit­ics. “There are things that are dif­fer­ent this cycle, but there are many things that feel the same,” he said.

(Image via Joseph Sohm / )