Tlaloc argues quite convincingly that our current method of electing a President is flawed in at least two major respects: first, it values voters disproportionately, and second, it essentially disenfranchises voters who vote for the losing candidate within their state. I think that there are two reasonable counters to this argument:

First, Voters control the value of their own vote because they can move between states and there is no waiting period for "state citizenship" or anything like that, and because they can vote for the presumptive winning candidate if it bothers them that their vote for the losing candidate wouldn't "count" towards the result. If someone in CA feels like he is getting screwed by the EC, he can always move to Wyoming and vote for the Republican candidate. This isn't always a practical solution to the issues Tlaloc raises but it's available, so I thought it ought to at least be mentioned.

Second, the disproportionate problem is mitigated (if not outright resolved) by the on-the-ground realities of how votes within states determine the winner. If you measure the value of a vote in terms of what is paid to get it by politicians, in advertising dollars or campaign expenses or legislative promises, then it is not necessarily true that a voter in CA counts less than one in WY; a candidate might find it worthwhile to spend more per voter in CA to try to get all those electoral votes, and basically write off WY. Somewhat oddly, this also addresses the disenfrachisement problem because a voter backing a candidate in a losing state does have influence through his vote, or even just through expressing his opinion in a poll -- his support can help convince candidates to devote more time and money to that state, which will make his vote more influential as described above.

There are a few arguments in favor of the EC that I think are less convincing than those advanced by Tlaloc; I want to just briefly mention them too: (1) The EC is traditional. Well, so what? After 200+ years of trial and error shouldn't we be able to improve the system? (2) The EC is federalist Why? The Senate already protects the interests of smaller states, it's not clear to me that the Presidential election should try to do so as well. (3) The EC forces politicians to garner widespread support. In practice, the support within just a handful of key "swing states" is all that matters -- the rest of the country is polarized enough along partisan lines that there is no real need for a Republican to campaign in a traditionally red state or a Democrat in a strongly blue state. So we get a national obsession with Ohio and Florida, and the issues that matter to those particular states.

There are three good reasons that I see to keep the EC:
** Some disproportionate influence is good. If candidates only had to sway the largest demographic in the country then that's what they'd do. By breaking it down by states campaigns are forced to court sub-groups of voters to try to get an edge in that state, and thus they respond to the particular demographics and issues of concern within each region. This means (in theory) that a President will be concerned with many different groups of Americans rather than just the largest group nationally.
** The EC encourages voting. In a national election with polls showing a 2% lead for the other candidate, you know your guy is going to lose by about 6 million votes and yours is useless. Within a given state, that same 2% lead is a much smaller number of votes. Tlaloc mentioned that if the EC concept is so great why don't we use it within states? Well, maybe we should -- a 2% lead within a county is smaller still. This compromise would simultaneously make the election more representative of the popular vote and make each individual vote more meaningful.
** The lopsided EC victories convey legitimacy. Races that are close nationally often result in large EC margins of victory due to the all-or-nothing nature of the vote within each state. I think this is actually useful in smoothing over partisan rancor post-election, as voters are more easily able to accept the results as definitive. I have to admit that when this goes wrong -- when the EC hangs on a few thousand votes within a particular state -- there is certainly lingering bitterness from the losing side.

Something that deserves its own diary: the way the EC treats third parties, and whether this is good or bad. Anyone with a particular interest in that topic should toss something up for discussion.