Expert Commentary

2008 U.S. presidential primaries through the lens of prediction markets

2011 study by Stanford University and the California Institute of Technology on how the presidential primary system shapes certain outcomes.

Political scientists continue to explore how the current U.S. presidential primary system shapes the nomination process and produces certain kinds of outcomes. Critics note that candidates in primaries often must court party voters with more extreme positions, and early primary states such as Iowa and New Hampshire are not demographically or politically representative of the country as a whole.

A 2011 study by Stanford University and the California Institute of Technology, “The 2008 Presidential Primary through the Lens of Prediction Markets,” analyzes the influence of 2008 presidential primary elections results on subsequent primary outcomes by using predictions market data. (Such markets are venues where “shares” relating to the outcomes of events can be bought or sold by the public.) Researchers measured whether or not the probability of a given candidate winning the general election, as decided by the predictions market, changed before and after each primary contest.

The study finding’s include:

  • The length of the primary season does not impact a candidate’s ability to win the general election.
  • “Certain states have a disproportionate influence on the nominating process. However, the states that have the largest impact are not necessarily New Hampshire and Iowa.”
  • Primary contests can demonstrate a candidate’s strength with particular demographic groups and influence future outcomes.  For example, Democratic candidate Hillary Clinton’s greatest prediction market gains came after her May 2008 primary victories in the states of Indiana, North Carolina, Kentucky and Oregon.
  • While an expected primary win did not necessarily influence a candidate’s value in the predictions market, a surprise win did. Barack Obama’s string of unexpected victories in the Super Tuesday primary states of Louisiana, Nebraska and Washington, for instance, helped to affirm him as the Democratic nominee frontrunner.
  • The electoral prospects of Republic nominee John McCain’s closest rivals, Mike Huckabee and Mitt Romney, rose and fell despite McCain’s domination in the primary contests; researchers consider these fluctuations to be signs of Republican displeasure with McCain’s candidacy.
  • While Super Tuesday’s outcomes influenced the number of state delegates each candidate won, their overall impact on subsequent primaries were limited. The researchers suggest that, in 2008, “it was clearly advantageous for states to seek out unique spots on the calendar, even if that meant going later than [other states].”

The researchers concede that their data are based on the predictions market from a single presidential election and may not be generalizable, noting that “commentators (e.g. Pew 2008) have been quick to point out the difference between 2008 and a typical election year.”

About The Author