The folks at the Denver Post must have some incredible sources. On Saturday night, November 1, a reader visiting the paper's election-data site, DenverElections.com, discovered that partial results in a slew of races had already been published -- and they remained in place the next day. (As of this hour, they no longer appear.) In all likelihood, the figures represented a test that the average web surfer wasn't supposed to see. But if not, John McCain should be very happy. Click "More" to see page captures of assorted Post's "predictions."
Apparently, the Bradley Effect -- the notion (named for former Los Angeles Mayor Tom Bradley) that white voters tell pollsters they favor an African-American candidate but wind up casting their ballot for a fellow Caucasian -- remains viable. The Post says that with 2217 of 3215 precincts reporting, McCain, who's been trailing in state polls for weeks, is leading Barack Obama 41 percent to 36 percent. Even more surprising: Thomas Stevens, the Objectivist Party candidate, is registering 2 percent, kicking the crap out of better-known minor-party candidates such as Libertarian Bob Barr and the astonishingly persistent Ralph Nader. Granted, the numbers allotted to each of these folks seem off percentage-wise, but never mind. Don't rain on the Objectivists' parade.
Things aren't looking good for amendments 48, 49 and 52, as seen here; all of them are losing with, yes, 68 percent of precincts reporting. But other amendments listed below these three are faring better: amendments 47, 50 and 54 all seem bound for glory.
Want more evidence that polls don't mean anything? In the Post's prognostication, Republican Bob Schaffer, who's spent the entire election season trailing badly in the U.S. Senate contest, is thumping Democrat Mark Udall 44 percent to 39 percent. Either those so-called experts were badly mistaken or the media doesn't suffer from a liberal bias after all. Thanks, Post statisticians, for clearing that last one up. -- Michael Roberts
