This may be the year Electoral College ranks No. 1

If it comes to it, Framers' machinery ready to install the less popular man

By Scot Lehigh, Globe Staff, 11/5/2000

It was only a remote possibility until now. But the sort of disarray that rocked the United States twice in the second half of the 19th century - an Electoral College win for the popular-vote loser - now looms larger because of a candidate's 24-year-old drunken-driving arrest.

Specialists say a popular-vote-loss-but-electoral-vote-win scenario is improbable unless the candidates are within 2.5 points of each other. Generally, polls have shown Texas Governor George W. Bush ahead of Vice President Al Gore by more than that. But with controversy over Bush's previously unacknowledged arrest roiling the campaign - and, by some estimations, tightening the race even more - the candidates could be within that margin by Tuesday.

How could a candidate lose by winning or vice versa?

Because the battle for Electoral College votes is a state-by-state affair. And whether one wins a state's popular vote by five or 500,000 makes no difference; the candidate who carries the state wins all the electoral votes. Everywhere but Maine and Nebraska, that is, which allocate some electoral votes based on who wins the congressional districts.

Thus, because Bush has had large leads in many Southern and Western states, while the race has been razor-close in big battleground states, the Republican could run up a narrow popular-vote victory but see enough electoral-vote mother lodes - states like, say, Michigan, Pennsylvania, or Florida - go narrowly to Gore to give the Democrat the victory.

Robert Erikson and Karl Sigman, professors at Columbia University, recently analyzed the chances of such an occurrence this year.

According to Erikson, if the popular vote were tied, Gore would have a 99.4 percent chance of winning the Electoral College. Even if Bush were to win by 2 percent in the popular vote, Gore would have a 50-50 chance of prevailing in the Electoral College, Erikson says.

''But if it's 3 points in the popular vote, Gore would only have a slim chance, and if it's 4 or 5 points, then it looks like an Electoral College landslide'' for Bush, said Erikson, a political scientist.

It is unusual for a candidate to lose the popular vote but win the electoral vote - but it's not unheard of. In the last century, similar election imbalances occurred twice in 12 years: In 1876, Democrat Samuel Tilden won the popular vote, 51 to 48 percent, over Republican Rutherford B. Hayes.

But amidst allegations of skullduggery, the appointment of a special commission, and months of finagling, Hayes was eventually awarded the Electoral College victory, 185 to 184.

In 1888, incumbent President Grover Cleveland prevailed in the popular vote, winning by a slender 48.6 percent to 47.8, only to see his Republican opponent, Benjamin Harrison, win the Electoral College, 233 to 168.

If a similar disparity occurs this year, there would almost certainly be attempts to pressure electors to honor the popular-vote outcome. Although not legally bound to their preferred candidate, normally electors cast their votes all but automatically for the candidate who carries their state.

Remaining remote, even with a tie in the popular vote, is the prospect of an Electoral College tie, with each man getting 269 votes. The chance of that is about one in 100. In that case, the new House of Representatives would pick the president.