Media Predict Blog

Donald Trump will not win the Republican nomination July 27, 2015

According to a recent telephone poll conducted by CNN/ORC, Donald Trump leads the current field of Republican candidates, clocking in at 18%. Former Florida governor Jeb Bush trails just behind at 15%, within the poll's margin of error.

An interesting piece of news. One that is fueling a discussion, you might say. A discussion about something that just isn't going to happen.

Yes, Trump might be enjoying a moment, but when you ask a large group of people to place a bet on whether he will actually win, as we do regularly here at Media Predict, almost no one thinks he will.

In fact, odds are currently hovering at a miserable 5.6% – meaning that shares purchased at 5.6 cents would be worth $1 if Trump were to win. You could multiply your money 18 times over on that bet, if he indeed got the nomination. And yet even at these odds, participants in our marketplace are still betting against Trump by a ratio of 2.3 to 1.

To be fair, all of these recent headlines did give him a bit of a boost – he’s up from June, when he was at 1.2%.

Ultimately Trump’s recent 18% poll number will be as reliable as the same CNN poll that had the 2012 Presidential race “too close to call” on the eve of the election. That less-than-accurate result came after a nomination process in which a gallery of eccentric candidates, from Bachmann to Perry to Cain, all realized ephemeral fame. Bad polling data fueled that circus too.

The crisis for polling is existential. It’s not a matter of statistical significance or sampling error. Rather, today, people know what is happening when they participate in a national poll: that is, nothing is real. Instead of naming the candidate they will actually vote for, a savvy public uses polls to articulate discontent, to send a message – one that may have no connection to future behavior. Polls have become a form of social media, a Twitter-like platform for the temporary expression of mood.

Should we be interested in why, in such a forum, people would express a fitful willingness to vote for Trump, and frustration with the other Republican candidates? Sure, analyze away.

But Donald Trump will not win the Republican nomination. The degree to which prediction markets guide our discussion of him is the degree to which we can talk about more useful things.

For the first time, the market swings to the Democrats June 23, 2015

For the first time, Democrats have pulled ahead of Republicans in our 2016 Presidential Election market.

Prediction markets are well-documented as a better long-range indicator than polls. So in January we figured we'd get ahead of the 2016 election with some early results. Thus far it's been a bumpy ride for the incumbents, with Republicans at time showing a probability of 60%, and Democrats' chances sagging in the 40% range over the past five months.

That changed this week, with the Democrats pulling even over the weekend and nosing ahead for the first time. Traders favored Democrats by a ratio of 1.2:1 last Friday. Here is trading over the past 24 hours.

All major demographics bet more heavily on Democrats than Republicans in the past five days -- but young people showed the most conviction, with men 18-34 and women 18-34 betting Democratic by huge ratios of 3.5:1 and 2.5:1. However those who say they are undecided voters are still betting Republican, 1.6:1.

A small number of traders saw a perception problem in relation to recent events in South Carolina ("Republicans are too far to the right to appeal to the majority of people"). However far more traders cited the strength of Hilary Clinton's early campaign ("so far no one of the others could beat Clinton").

Stay tuned for more!