Friday, October 28, 2016

Evan McMullin's odds are improving

Evan McMullin’s star is rising. The insurgent candidate has only been on the campaign trail for two and a half months, an unbelievably short time for a presidential campaign, but he is already poised to win a state, a feat not accomplished by the Libertarian Party in its 44-year history (although the party has received one electoral vote from a faithless elector).

McMullin received some welcome publicity earlier in October when the FiveThirtyEight political blog rated his chances of winning the White House as high as three percent. Since then, McMullin surged to the lead in one Utah poll and a statistical tie in two others.

Due to extreme interest in McMullin’s campaign, FiveThirtyEight recently revisited the few Utah polls available. The analysis found that McMullin actually has a better chance of winning Utah than the average of polls indicates.

The problem with the polls is that several pollsters treat third party candidates differently than the main party candidates. Two polls that heavily favor Trump had issues that might have hurt the outcome for McMullin. A YouGov poll that gave Trump a 17-point lead only allowed respondents to select McMullin if they first selected “someone else.” Another poll by Monmouth gave Trump a 14-point lead over McMullin and six over Clinton. This poll listed third party candidates as an initial choice, but not as a secondary choice for undecided voters.

Another poll may have given McMullin an unfair boost. A poll by Rasmussen that showed McMullin one point behind Trump listed McMullin’s political affiliation as “independent conservative” when he will actually be listed on the Utah ballot as “unaffiliated.” FiveThirtyEight noted that the description as a conservative may have helped McMullin in the deeply conservative state.

The analysts at FiveThirtyEight make predictions on two models, one using polls and the “nowcast,” an estimate of what would happen if the election were held today. Using all available polls, McMullin has an estimated 14 percent chance from the polls-only forecast and 22 percent from the nowcast. When polls that didn’t treat McMullin the same as the major party candidates were excluded, his chances rose to 23 and 38 percent respectively.

The bettors on the political betting site, Predict It, seem to agree with the FiveThirtyEight assessment. Bets that McMullin will win Utah are currently trading at 36 cents. Bets that he loses are 67 cents. This is very close to the FiveThirtyEight estimate of McMullin’s chances of success.

Winning Utah is only part of McMullin’s overall strategy to deny the presidency to Trump and Hillary though. To win the election, McMullin must throw the Electoral College into a tie by denying both Trump and Hillary 270 electoral votes. FiveThirtyEight rates that possibility at less than one percent.

The blog doesn’t provide an estimate of McMullin’s chances of going all the way to the White House. They say that what would happen if the election were forced into the House of Representatives by an Electoral College deadlock “is entirely speculative.” Nevertheless, they do say that pollsters should “treat him like a candidate who has a chance to win. Because he does.”

McMullin has even earned a spot on Predict It’s market for who will win the presidential election. A bet on McMullin can currently be placed for one cent compared to 81 cents for Hillary Clinton and 21 cents for Donald Trump. In a year in which rules seem made to be broken, a one cent bet that pays a dollar if McMullin becomes president might be a good investment.

Originally published on The Resurgent

No comments: