Newswise — The role of non-competitive states in presidential elections has been underestimated, an analysis by a pair of game theorists shows. In addition, their findings, which draw from election results over the previous four presidential contests, indicate that 2016 could be a repeat of the 2008 presidential election in which the winning candidate could obtain an Electoral College majority while losing all of the battleground states.

“Contrary to conventional wisdom, the non-competitive states in a U.S. presidential election do count, but in a way that is different from the competitive states,” write the study’s co-authors, New York University Professor Steven Brams and Marc Kilgour, a mathematics professor at Canada’s Wildrid Laurier University.

Their work, which appears online in the journal Public Choice, centers on the set-up power of non-competitive, or deeply blue and deeply red, states. By structuring the contest between the major-party candidates, they influence—if not determine—the outcome of presidential elections.

The researchers’ model considers how noncompetitive states affect, or set up, presidential campaigns in competitive states. Specifically, they define three measures of set-up power: winningness (the proportion of splits of competitive states that make a candidate the winner), vulnerability (the proportion of a candidate’s winning coalitions in which a single competitive state can change the outcome), and fragility (the expected number of competitive states in a winning coalition that can disrupt victory). The set-up power of the non-competitive states increases with winningness and decreases with vulnerability and fragility.

Brams and Kilgour then show how these factors spotlight the advantages of the candidate who leads in electoral votes of noncompetitive states. Their model pointed to the eventual winner in the past four presidential elections. # # #