Like polls, prediction markets failed to see Trump's victory coming, economist says

November 11, 2016, University of Kansas

As network anchors and pundits appeared stunned at Donald Trump's Electoral College victory Tuesday night, it was obvious much political polling data had missed the mark.

Also most —often touted as a better indicator for election outcomes—had widely predicted a Hillary Clinton victory, including one that had put her probability of victory at 83 percent.

"Prediction markets now have a lot of egg on their faces with the presidential election and the Brexit vote earlier this year," said Koleman Strumpf, a University of Kansas professor of business economics.

Prediction markets, or information markets, are exchange-traded markets that allow trading on the outcome of events, and the prices of the exchange can be an indicator of what a crowd thinks a probability of an outcome will be. Strumpf, who is associate editor of the Journal of Prediction Markets, said the markets were useful in elections when they cropped up about 100 years ago, especially in the absence of public polling.

After Tuesday's , economists and data scientists will likely examine specifically what went wrong with most prediction markets and whether they have reached their limits at being useful, he said.

As far as the U.S. , it's possible polling data too heavily influenced decisions in the prediction markets.

"If apparently the information was bad, then the markets weren't able to figure that out," Strumpf said. "However, the markets' job is to see through that typically. For lack of a better expression, it seemed everything became an echo chamber. People got convinced of the inevitability of Clinton winning, and they were unwilling to fully weight a set of circumstances that could lead to what actually did happen, Trump winning the Electoral College majority."

Explore further: Get better election predictions by combining diverse forecasts

Related Stories

Algorithm has the word on U.S. election

November 4, 2016

A University of Queensland computer scientist who harvests social media to accurately predict election results says he can confidently call next week's United States election.

Recommended for you

Coffee-based colloids for direct solar absorption

March 22, 2019

Solar energy is one of the most promising resources to help reduce fossil fuel consumption and mitigate greenhouse gas emissions to power a sustainable future. Devices presently in use to convert solar energy into thermal ...

EPA adviser is promoting harmful ideas, scientists say

March 22, 2019

The Trump administration's reliance on industry-funded environmental specialists is again coming under fire, this time by researchers who say that Louis Anthony "Tony" Cox Jr., who leads a key Environmental Protection Agency ...

3 comments

Adjust slider to filter visible comments by rank

Display comments: newest first

gkam
Nov 16, 2016
This comment has been removed by a moderator.
axemaster
5 / 5 (2) Nov 19, 2016
It's sort of interesting to see how the polls and prediction markets are being criticized. Is this really fair? Remember: they are only predicting probabilities. If they said that Trump had a 20% chance to win, that's not the same as saying "Trump will lose".

It's also interesting to note that these predictions weren't even wrong by much - Clinton won the popular vote by a fairly wide margin (>1.4 million votes). So as an outsider looking in, it isn't surprising to me that the polls predicted a Clinton victory.

Before the trolls attack, let me just say that I hated both of these candidates with a passion, and I'm not plugging for Clinton here. I'm merely reciting some of the facts.
snoosebaum
not rated yet Nov 19, 2016
more establishment humor , pretending that they WEREN'T LYING!!

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.