VOTE VALUES VARY—The Constitution allocates electoral votes to each state based on that state's representation in Congress. Each state receives two electoral votes corresponding to its number of U.S. senators. When states with unequal populations receive similar numbers of electoral votes, states with smaller populations gain a mathematical advantage. Thus, every voter's ballot does not carry the same weight. The typical citizen of Wyoming has on average four times as much influence in determining an electoral vote for president as the typical citizen of California and twice as much influence as the typical citizen of Texas.

In addition, under the Electoral College third-party candidates like Ralph Nader in 2000 may actually determine the outcome of the election. There is little question that Nader cost Gore the election in 2000. Most Nader voters would have voted for Gore in the absence of a Nader candidacy. Gore lost Florida by 537 votes, while Nader received 97,488 votes in that state. Pat Buchanan and the Libertarian candidate Harry Browne received a total of only 33,899, which were more likely to have otherwise gone to Bush.

Similarly, Gore lost New Hampshire by 7,211 votes while Nader received 22,198 votes. Buchanan and Browne together received 5,372 votes. Gore would have been elected if he had won either state.

THIS HOUSE IS NOT YOUR HOUSE—If no candidate receives a majority of the Electoral College vote, the House of Representatives selects the president. In this selection, every state has one vote. It would be possible for the seven members of the House representing the seven smallest states, with a combined population of about 4.9 million Americans, to have more say in the selection of the president than the 177 members representing the six largest states, with a combined population of 119 million people. The selection of the president by the House is the most egregious violation of democratic principles in American politics.

One net result of these distorting factors is that the candidate who receives the most votes in the country may not win the election, as happened in 1824, 1876, 1888, 1960 (as I demonstrate in my book Why the Electoral College Is Bad for America), and 2000.

The Electoral College distorts the preferences of Americans and thus the results of presidential elections. It violates political equality, favoring some citizens over others, depending solely upon the state in which people cast their votes for president. Can we reconcile such a selection system with the ideal of political equality, one of the most deeply ingrained of democratic principles? We cannot.

CONSENSUS WINNERS?—According to its supporters, one of the primary virtues of the Electoral College is that candidates must obtain concurrent majorities from around the country in order to win. In other words, these advocates argue, by guaranteeing a specific number of electoral votes to each state, the Electoral College forces the winner to pay attention to all regions of the country and to build broad coalitions by winning a wide geographic distribution of states that helps his coalition mirror the nation.

If states did not employ the unit rule by allocating all their electoral votes as a bloc, the argument goes, candidates might appeal to clusters of voters whose votes could be aggregated across states and regions. This could be potentially divisive and lead to discord because the coalition behind a candidate might represent only one part of society.

Electoral College advocates build their case about its virtues on a set of faulty premises. Do candidates try to build broad national coalitions by appealing to voters throughout the nation? Except in a superficial fashion, the candidates of both parties virtually ignore large sections of the country.

For example, in the presidential election of 2000, one of the most competitive elections in history, the Electoral College distorted the political system by providing incentives for candidates to campaign actively in only 17 "battleground" states, and largely to ignore the other 33 states and Washington, D.C. The candidates made few visits and did little or no advertising in the Great Plains, the Rockies, the Deep South, and the Southwest—except for New Mexico. The candidates displayed a similar pattern in 2004.

With a few exceptions, small states were not among the "battleground" states. Indeed, it is difficult to imagine how presidential candidates could be less attentive to small states. Under direct election of the president, candidates would be much more attentive to small states. They would have incentives to appeal to all voters, not just those strategically located in swing states.

Do candidates actually win support across regions? Anyone examining the red and blue states on an election-night map knows that candidates tend to win with regional support. John Kerry and Al Gore won the West Coast, the Northeast and the Upper Midwest, while George W. Bush won the South, the Mountain States and the rural Midwest.

Do winning candidates receive majority support across social strata? In 2000, George W. Bush did not win a larger percentage than Al Gore of the votes cast by women, blacks, Hispanics, Asian-Americans, the elderly, the poor, members of labor unions, those with a high-school education or less, those with postgraduate education, Catholics, Jews, liberals, moderates, urbanites, those with less than $50,000 of household income, voters aged 18 to 29 and 60 and older—in addition to those living in the East and West. It is no criticism of the winner, but his vote simply did not represent winning concurrent majorities across the major strata of American society.