Cool it with those `trial heats'

SPARE us - for now - another political story about ``who's on first.'' It is common fare these days. The relatively early emergence of Vice-President George Bush and Massachusetts Gov. Michael Dukakis as the likely nominees of their respective parties has prompted pollsters to pair the two off in ``if-the-election-were-held-today'' match-ups.

This has given the candidates' advisers and political pundits tea leaves to sift through for divinations about the relative strengths and weaknesses in each campaign camp. But trial-heat results should be taken with a large measure of skepticism. If taken too seriously at this early date, the spin put on them could become self-fulfilling prophecy.

Trial heats this early are essentially meaningless. A significant portion of the electorate doesn't begin to get serious about candidates until after the conventions. As much as 60 percent of voters may have made up their minds already, based largely on party affiliation. That leaves 40 percent - independents and voters only loosely tied to their declared party - to swing with the wind of current events. This last group tends to overwhelm surveys taken this early.

Moreover, nationwide trial heats fail to shed light on how the vote might go ``if the election were held today'' in the Electoral College, which ultimately decides who the next occupant of the White House will be. Those results would be just as shaky as any other trial heat, but at least they would focus attention on the body making the final choice. It's worth remembering that in 1960, Richard Nixon received a plurality of the popular vote (after discounting Democrats elected to the college who vowed not to vote for John Kennedy) but lost the White House in the Electoral College.

Trial heats as campaign tools don't really take on any value until after the conventions. These days, analysts would perform a greater public service by emphasizing what polls are saying about national concerns and issues rather than horse-race scenarios. That would help focus attention on the substance of the election, and, it is hoped, make any trial-heat numbers more meaningful come fall.