One Last Thing: Who can go the distance?

Barack Obama's electoral flaws manifested themselves this week as he lost primaries in Texas, Ohio and Rhode Island. If you've been reading this column and paying close attention to the voting coalitions of Obama and Hillary Rodham Clinton, this came as no surprise. By now, others have told you how "unstoppable" Obama is enough times that you should know better.

Obama had a monthlong string of victories in states tailor-made for his campaign. He had a month of the most fawning and deferential media coverage imaginable. He had a month of presumed inevitability that saw many otherwise serious people calling for Clinton to leave the race. He had a month in which he raised $55 million, enabling him to outspend Clinton 2-1 in Ohio and Texas.

And yet he lost the two most relevant states since Super Tuesday.

Obama still leads in pledged delegates and the popular vote, as long as you exclude the results in Michigan and Florida (more of which anon). Democratic primaries in 10 states, Puerto Rico and Guam remain - and the demographic makeup of those states is largely favorable for Clinton. To this point, demography has been destiny - and there's no reason to think that will change. Obama should win Mississippi, North Carolina, and a couple of smaller states. Clinton should win Pennsylvania, Indiana, Kentucky and West Virginia.

If that is what happens, neither candidate will have enough pledged delegates to secure the nomination before the Democratic convention in August. Obama would arrive in Denver with a lead of perhaps two dozen pledged delegates.

What happens then? Each candidate will make a case to the Democratic Party that he or she has the more legitimate claim. The superdelegates will then decide the nomination.

Obama's argument will be as follows: He garnered more pledged delegates and he does better in prospective polling matchups against John McCain.

Clinton's case is more complicated - but I think it has a better chance of working. She can argue that Obama's lead stands up only if you exclude the results of Michigan and Florida. These two states were stripped of their delegates by the national party after naughty state party officials decided to move the primaries up on the calendar. Clinton and Obama both promised not to campaign in either state. Clinton won both primaries handily.

The Obama camp insists that rules are rules and Florida and Michigan shouldn't count. But is that really a winning argument within the Democratic Party? About 2.3 million Democrats voted in those primaries. Remember that in 2000, Democrats fervently argued that Florida election law was less important than making sure "every vote counted." The party then claimed that holding to the strict letter of the law was tantamount to disenfranchisement.

If "voter intent" was more important than the law in 2000, shouldn't it also trump mere party rules in 2008? Obama's argument about the paramount importance of rules might work with a Republican audience, but it runs counter to the ideological framework of the Democratic Party.

The Obama camp complains that its candidate wasn't even on the ballot in Michigan. Fair enough. But Obama was on the ballot in Florida, and he even ran TV ads there, in violation of his pledge not to campaign in the Sunshine State. Clinton won Florida going away, 50 percent to 33 percent.

Come the convention, the inclusion of Florida alone will likely give Clinton the lead in pledged delegates and the popular vote. The inclusion of Michigan and Florida would almost certainly give her the lead in both.

In any case, Obama has a lead in pledged delegates only because he has been strong in caucus states. Obama can reasonably claim that caucuses are just as valid as primaries. But in truth, caucuses are not a particularly good indicator of electoral strength. Consider Texas: Obama won the Texas caucuses even as he lost the primary by a slim but hardly trivial margin (about 100,000 votes).

Finally, while Obama might lead McCain in theoretical matchups, Clinton can make the case that her voting coalition will be more formidable in the general election.

Obama's support comes in large part from reliably Republican states such as Idaho, Utah, Georgia and South Carolina. Democrats have no chance in those states come November. Meanwhile, Clinton will have won at least eight of the 11 largest states, including must-win battleground states such as Florida and Ohio (and Pennsylvania).

Remember, too, that Obama's coalition is composed of more reliably Democratic base voters: African Americans, voters making over $100,000, and young voters. These are groups that Democratic candidates carry most easily. If Clinton is the nominee, she can take these groups for granted.

By contrast, Clinton's coalition - women, older voters, whites making less than $50,000, Catholics, Hispanics - would be McCain swing voters in a race against Obama. Obama hasn't been successful in wooing those voters yet, so it's unclear why anyone would believe he will finally carry them (and then defend them from a very appealing McCain) in November.

In other words, if you look at the underlying fundamentals of the race, and not just the theoretical polls, Clinton can make a strong case that she is the candidate better suited to challenging McCain and winning the White House.

Don't be fooled by authoritative-sounding delegate math: If neither candidate has the required number of delegates, then holding a slim delegate lead matters much less than being able to articulate a convincing argument to the superdelegates, who are free to vote as they please.

Clinton has at least even odds at being the nominee. And she would probably be the stronger candidate for the Democrats.