Oh, now that’s an interesting exit poll excuse.
The King of Zembla has found a new and improved (if by improved, you mean threadbare) excuse for the exit polls for the 2004 "election" in the United States not predicting the same winner as the one that the (supplied by the Evil Party) tabulation machines and the state officials (who, by a happy coincidence, turned out to be members of the Evil Party) produced.
The exit polls contacted more supporters of Kerry than of Bush because of "the failure of interviewers to follow the selection rate," said Warren Mitofsky, who conducted the exit polls along with Joe Lenski of Edison Media Research.
The King of Zembla makes some unfair comments implying that changing the selection rate won't make any difference (it won't if you're using one of those boring reality-based definitions of "selection rate", but that's not important), but he has to realize that the proper selection rate is to only ask people who are wearing buttons indicating their support for the Coward in Chief (or, failing that, to just not ask anyone and wait for instructions from the local (member of the Evil Party reelection committee) Secretary of State on what the poll results should be.) Remember, exit polls are not to double-check the actual poll results for honesty (you only do that in barbarian lands where you need to make certain that the Empire's chosen vassal wins the "election" there, but you've not been able to properly export the American electoral system and have your boy win no matter who the rabble might want to vote for), but to apply a reputable veneer to the election you've already stolen. The exit polls in the United States failed at this, and thus will have to be hidden away until they can be scrubbed to match the official results.
Remember, those 10,000 lucky coincidences took a lot of time, money, and blackmail to get into place. It would be unpatriotic to report things that might make hoi polloi realize that their country was just stolen out from under their noses.