Democrats : popular vote may now be key

Hillary Clinton's victories in Ohio and Texas last Tuesday may have done little to reduce Barack Obama's lead in the pledged delegate count (she enjoyed a net gain of just 10 delegates on the night according to the latest figures from Realclearpolitics) but the margin of her victory in the popular vote, which she won by 330,000 votes across the two states , raises an intriguing possibility : that come the end of the primary season in June, Democrats could face a situation where Barack Obama has more pledged delegates yet Hillary Clinton has an overall lead in the popular vote.

According to the rules, the Democratic nomination is awarded to the candidate who amasses most delegates - that is, a sum of both pledged delegates (awarded according to the results of primaries and caucuses) and super delegates (senior party officials who are free to vote whichever way they choose). However so far this primary campaign, the Obama camp, aware from the start that Senator Clinton would likely hold an edge among super delegates, has been making the argument that the winner of the pledged delegate count should receive the nomination, the reasoning being that pledged delegates represent the will of Democrat voters across the country.

So far they seem to have been winning this debate : senior Democrats like House Speaker Nancy Pelosi and New Mexico Governor Bill Richardson (both neutral in the Clinton-Obama race) have publicly stated that super delegates should not overturn the will of the people by voting against the winner of the most pledged delegates. One Democrat, former Virginia Governor Doug Wilder, even warned of riots at the party convention in August if Senator Clinton were to win the nomination by relying on support from super delegates after winning less pledged delegates than Obama.

The argument that the pledged delegate count represents the will of the voters would of course lose much of its validity if Mrs Clinton were to win the overall popular vote but win less pledged delegates than Mr Obama, creating a real quandary for the Democrats, a party still bitter about the manner of their loss in the 2000 presidential election when their candidate Al Gore won the popular vote but lost to George W. Bush because of an inferior number of electoral college votes.

With 13 primaries remaining, Obama's lead of 153 in the pledged delegate count represents a formidable margin due to the proportional representation system used by the Democrats, which make it very difficult for a candidate to win anything other than a narrow majority of delegates from any state). His lead of just under 600,000 in the popular vote is far less secure however, and a continuation of Clinton's strong performance in Texas and Ohio across the remaining contests could well see the New York senator achieve a narrow victory in the total popular vote, especially if, as seems increasingly likely, the large states of Florida and Michigan hold do-over primaries.

An analysis of likely results and turnout numbers in the remaning states underlines this fact. To calculate an approximation of expected turnout, we can analyse several factors - the number of Democrat voters in each sate, turnout in the 2004 primary, whether the primary is open to all voters or closed to registered Democrats only, and turnout trends in those states that have already held their primaries and caucuses this year.

To calculate expected results we can look at demographic factors in each state, opinion polling and market expectations (based on bookmaker odds, Intrade percentages and Rasmussen Markets). The figures we arrive at are of course merely rough estimates, yet nonetheless they are useful in determining the broad parameters of the results we are likely to see, highlighting what outcomes may be considered possible (such as Senator Clinton winning the popular vote) and which outcomes may be considered unrealistic or improbable (such as Clinton winning the pledged delegate total).

Below is the list of possible turnouts and results that we can use as guide to likely outcomes in the remaining states. I have assumed Florida and Michigan will hold regular do-overs (and that their 54 super delegates will be reinstated) and in all cases I have been generous to Hillary Clinton, on the basis that only a continuation of her strong polling of last Tuesday will give her any chance whatsoever of securing the nomination. I have excluded the primary in Guam due to its small size (only 1,049 voted in the 2000 primary and only 4 delegates are at stake) and lack of polling data. It is still unclear who John Edwards' 26 pledged delegates will vote for at the convention, and as such I have excluded them from the calculations at this stage.

Date, state, projected turnout and projected result :

11-Mar	Mississippi	370,000	Obama wins by 20%
22-Apr	Pennsylvania	1,675,000	Clinton wins by 14%
06-May	Indiana	785,000	Clinton wins by 4%
06-May	North Carolina	1,236,000	Obama wins by 10%
13-May	West Virginia	300,000	Clinton wins by 14%
20-May	Kentucky	410,000	Clinton wins by 20%
20-May	Oregon	540,000	Obama wins by 10%
01-Jun	Puerto Rico	400,000	Clinton wins by 20%
03-Jun	Montana	140,000	Obama wins by 10%
03-Jun	South Dakota	85,000	Obama wins by 10%
tbd	Florida	1,900,000	Clinton wins by 16%
tbd	Michigan	1,500,000	Clinton wins by 5%

These results would see Obama winning the final pledged delegate count by 103 and Clinton narrowing the gap in the popular vote to the tiny margin of 15,000 (less than 0.05% of the final total, well within margins of error). With Hillary currently leading by 35 amongst super delegates this would leave her needing to successfully court 58% of the 399 remaining uncommitted super delegates to make up this deficit in the overall total, a task which would be made much easier if she were able to present herself as the people's choice due to even a narrow victory in the popular vote.

It should be pointed out at this stage that despite the fact that popular vote totals are being routinely reported and discussed by both politicians and major media outlets, these figures do not include numbers from the caucuses held in Texas, Washington, Iowa, Nevada or Maine, all but one of which (Nevada) were won by wide margins by Obama and none of which have released official popular vote totals. Of these by far the most significant omissions are those of Texas and Washington due to the size of their populations (21 million and 6 million respectively).

If voter turnout in Texas and Washington matched the patterns seen in other open caucuses this season, Obama's margins of victory, 12% (provisional figure) and 37% respectively, would represent a net win in the popular vote in those two contests of approximately 130,000, a potentially decisive number in such a close race as this. Expect to hear a lot more talk from the Obama campaign about the omission of these two caucuses if a situation does unfold where Clinton gains a lead in the popular vote.

If Senator Clinton manages to narrow the gap, it will be the last major primary in Puerto Rico on June 1st that ultimately decides the overall popular vote, the result of which is so critical to both candidates attempts to woo super delegates. The idea that Puerto Rico - which is not a US state and whose residents cannot even vote in the presidential election in November - could effectively decide the Democratic noimination will be controversial to say the least, especially if the Texas and Washington caucuses continue to be ommitted from the popular vote total.

In a campaign where winning the popular vote may now prove crucial, where hundreds of millions of dollars have been spent and which has been followed by the eyes of the world for months, is it possible that the outcome could finally come down to something as prosaic as whether party officials in Texas and Washington made provisions to record the number of attendees (rather than simply voting percentages) at their caucuses ?