The Edwards Primary

I saw this concept on MrSuper.org and it caught my attention.

I have commented before that major gains in net delegates from any contest are rare given the proportional distribution formula for delegates in the Democratic primary. The only way to gain a large net of delegates in a primary contest, excluding a blowout victory in a gigantic state, is to:

(1) Win a healthy margin of victory in a very large state.

When Senator Clinton beat Senator Obama in California by 10%, she was able to win a net pick-up of 38 delegates (204-166). That same margin of victory (10%) in New Jersey netted her 9 delegates, while in Ohio it only netted her 7 delegates. Part of this differential is due to the way delegates are allocated by congressional or state assembly districts, or allocated based on prior Democratic turnout in presidential elections.

or:

(2) Win by a large margin in a medium-sized state.

Virginia saw under a million voters come to the polls, but the 29% advantage for Obama netted him 25 delegates. Georgia also had 1.3 million voters come to the polls, and Obama's 35% margin of victory netted him 33 delegates.

In context, those 58 delegates represented a large net gain for Obama than did the results of the Ohio (7 net delegates), Pennsylvania (10), New Jersey (11), Rhode Island (5), Massachussetts (17) and Arizona (6).

or:

(3) Simply dominate a small state

Obama picked up 9 net delegates with a 75% - 24% win in the District of Columbia and 12 net delegates with an 80% - 17% victory in Idaho. A 25% win in Colorado netted him plus-15 delegates. Clinton picked up 19 net delegates with a 44% win in Arkansas.



The Remaining Contests

The remaining contests after North Carolina do not represent significant opportunities to net a large margin of delegates for Senator Clinton. Her best chances of gaining a net swing of 10 delegates are a 32% win in delegate-poor West Virginia, a 20% win in Kentucky, or a 20% win in Puerto Rico.

Obama can pick up a net of 10 delegates in Oregon with a 20% win or a 20% win in Puerto Rico.

Montana and South Dakota will only swing between 1 and 3 delegates maximum if the margin of victory is under 18%.


The Edwards Primary

So, what's the point here?

MrSuper.org points out that Edwards now claims 19 delegates in his column that could be swung one way or the other. Since he did not end his campaign (which would have released those delegates won), but rather suspended it, Edwards still maintains some leverage in how those delegates ultimately break.

In the big scheme of things, those 19 delegates are not going to decide who wins the most pledged delegates, and most likely will not factor into who wins the nomination.

However, if Edwards were to endorse a candidate, end his campaign, and those 19 delegates shifted to one or the other candidate, the end result would be the same as an astronomical victory in one of the remaining states (like a 36% win in Puerto Rico or 38% win in Kentucky) or the same as a healthy margin of victory in three contests for either candidate:

A 16% win in Oregon [+8], an 18% win in Montana [+4], and a 12% win in
Puerto-Rico [+7] for Obama would also equal +19 delegates.

For Clinton, a 24% win in West Virginia [+6], a 16% win in
Kentucky [+7] and a 12% win in Puerto Rico [+7] would equal +20 delegates.

MrSuper's calling this effect the "Edwards Primary" is an interesting way of looking at the impact of such a move, but upon further review, I think this analysis shows how even with huge victories in states she should win, Senator Clinton will not close the gap on Senator Obama by any appreciable measure, nor will Senator Obama increase his lead on Senator Clinton by any large number, either.

This is why I have been projecting that the final contests, especially those after May 6th, are essentially a deadlock and are essentially non-factors in the pledged delegate race. The only really significant moment might be on May 20th, when Obama is projected to claim victory in the pledged delegate race by surpassing the magic number of 1627 in pledged delegates.

Any other impact is simply politics, not mathematical in nature.