1 Does the Propensity to Strategically Defect From One’s Preferred Party Vary Across Elections, Parties and Voters?

The paper examines the propensity to strategically defect from one’s preferred party. The study uses CSES module 2 and covers 25 democratic lower house elections. We find that non partisans and supporters of weak parties are more prone to cast a strategic vote and that the less informed are as likely to strategically defect as the most informed. The electoral system, the degree of disproportionality and polarization, and the number of parties have no direct effect. The only significant contextual effect is a conditional one. While strategic desertion is almost exclusively at the expense of weak parties in the most disproportional systems, the bias is much more muted in the most proportional systems. The paper examines how the propensity to strategically defect from one’s preferred party depends on characteristics of voters, parties, and elections. Looking at the amount and sources of strategic defection is important for at least two reasons. First, electoral outcomes are often interpreted as mandates (see Fowler and Smirnov 2007), which implies that voters are expressing their sincere preferences when they decide which party/candidate to support. That interpretation has to be revisited if vote choice i