In their recent article in this Journal, Alvarez and Nagler model strategic voting in the 1987 British general election using multinomial probit.R. Michael Alvarez and Jonathan Nagler, ‘A New Approach to Modelling Strategic Voting in Multiparty Elections, British Journal of Political Science, 30 (2000), 57–75. They claim to find lower levels of such voting than many other approaches. They also demonstrate purportedly that estimates of strategic voting obtained using self-reported motivations are biased by the time elapsed between the election and the point at which the self-report is obtained. In this Comment I show that these assertions are incorrect.