The October 2008 issue of PS published a symposiumof presidential and congressional forecasts made in the summerleading up to the election. This article is an assessment of theaccuracy of their models.

The Incumbency and National Conditions Model predicted that JohnMcCain would receive 44.3% of the two-party presidential vote; heended up with 46.6% (at this writing, November 13, 2008), yielding aprediction error of 2.4 points. This error is slightly smaller thanthe mean absolute value of the out-of-sample forecasts from 1952 to2004 (2.68), so the 2008 election result is not anomalous in thecontext of this model. The addition to the model of the 2008 outcomedoes not generate many important changes to the coefficients, exceptthat the significance of incumbency interaction went from 0.08 to0.02 and the standard error and adjusted R2 of the modelboth indicate a slight improvement in fit.