Geege Schuman stashed this in Politics 2012
Post-convention polls can be volatile, but suggestions to "wait for more data" misses the point of political forecasting The idea is to make predictions "when the information is murkier — either by (correctly) discerning a trend a few days before other methods do, or (just as importantly) by avoiding attaching too much significance to one based on flimsy evidence."
It does make me wonder what anyone who hasn't made up their minds yet is waiting for.
I think they've mind up their minds, but they might be less (or more!) inclined to vote if their candidate appears to be a lost cause. Projections similarly could de-motivate (or motivate!) contributors.
Wow, in that sense polls become self-fulfilling prophecies.
And therefore polls offer incentives to game or to lie.
Big whopper data.