It has become an article of faith that in 1982, Tom Bradley, the African-American mayor of Los Angeles, was far ahead of his Republican challenger, George Deukmejian, in the polls for the California gubernatorial election.  The surprise caused by Bradley's loss is attributed to white voters who lied to pollsters about how they intended vote, thus explaining what is believed to be the wide discrepancy between the pre-election polls and what actually happened.  This is now the accepted and unchallenged wisdom by political pundits.  It has become known as "The Bradley Effect."

Some Trump supporters are citing "The Bradley Effect" as evidence for the proposition that voters are not being honest with pollsters when stating that they won't vote for him.  These Trump supporters believe that on Election Day, the polls will be proven wrong, and a groundswell of hidden Trump supporters will turn out.

The problem is that there never was a "Bradley Effect" in the 1982 California governor's race.  In a report datelined October 29, four days before the election, the Washington Post reported the results of the most recent poll, which showed that in the past three weeks, Bradley's lead had been cut from 14 to 3 points.  Hence, the most recent poll taken before the election was well within the margin of error.  There had been a very strong trend in the weeks leading up to the election away from Bradley and toward his opponent, and voters were openly telling this to pollsters.  Deukmejian won the election by 93,000 votes out of the 7.6 million votes cast.

Why did Bradley drop so quickly in the polls?  Such shifts are not without precedent.  Gerald Ford trailed Jimmy Carter by 30 points in 1976 and lost the popular vote by two points.  George H. Bush trailed Michael Dukakis in 1988 by double digits and ended up winning by 8 points.  In the last days of the 1982 California governor's race, anti-gun control advocates spent a great deal of money tying Bradley to unpopular proposed gun control legislation that lost by 2 to 1.  Following the election, the Washington Post, in a report datelined November 5, cited political analysts who attributed Bradley's loss to a "massive conservative turnout to defeat a state gun control measure."

There was also another reason. I lived in Los Angeles and had worked for Reagan's election in 1980, but I was one of the many Republicans who initially planned to vote for Bradley.  I considered him to have been an excellent mayor of Los Angeles.  However, part of the negative advertising run by the Deukmejian campaign was to tie Bradley to then-Democratic governor Jerry Brown, who was running for the Senate.  This is what led me to vote for Deukmejian.  Brown had become increasingly unpopular, and this was reflected in his loss by 500,000 votes to his Republican challenger.  The Deukmejian campaign had successfully tied Bradley to Brown in the minds of many voters.  Brown's loss is the overlooked aspect of what happened in California in 1982.  Brown was white, yet he lost by more than five times as many votes as Bradley.  This shows quite conclusively that white voters were not voting along racial lines, much less lying to pollsters.  In fact, Brown's unpopularity probably cost Bradley the governorship.

In the final analysis, the facts of California's 1982 gubernatorial election are much less important than popular misconceptions and myths.  The "Bradley Effect" has now become an urban legend in American politics and it is probably here to stay.