Right, but if anything that would overrepresent Republican performance among exit polls, no? Unless there were methodology issues with their random exit samplin

Right, but if anything that would overrepresent Republican performance among exit polls, no? Unless there were methodology issues with their random exit sampling or voters did not accurately disclose their votes. I guess all this depends on the disparate sorts of polling data that was available and how it was all weighted. Gelman et al published their model so it's worth digging into.