Showing posts with label end game calculations. Show all posts
Showing posts with label end game calculations. Show all posts

Tuesday, April 10, 2012

Santorum Suspends: A Nomination Race in Context

Well, FHQ will add its two cents. And not surprisingly, we'll look at Rick Santorum's decision to suspend his presidential campaign through the lens of the delegate count.

All too often delegate counts don't matter in the grand scheme of things in most presidential nomination races. To the extent that they do, it is fleeting. Counting up delegates is only consequential and/or necessary in a couple of instances:
  1. After the race has progressed far enough that one candidate has effectively taken all/most of the momentum -- or continued riding it from the invisible primary portion of the campaign -- and thus stretched out to a large enough lead to make a comeback unlikely if not impossible.
  2. If the race has progressed to a point where two or more candidates are trading primary and caucus wins and staying within range of each other in terms of the delegate count. 
These are the extremes and throughout much of the post-reform era the process has moved ever closer to the former rather than the latter. A constantly frontloading calendar gave, for much of that period, frontrunning candidates a greater and better opportunity to effectively wrap up the nomination early if they had established themselves as the clear frontrunner heading into the contest portion of the race. The nomination races in both parties in 2000 and for the Democrats in 2004 are good examples. But if 2008 demonstrated anything it was that if the invisible primary (fundraising, poll position and endorsement) has proven inconclusive, then true delegate counting may ensue. Certainly, this was more the case on the Democratic side in 2008 than among the Republican candidates.

One easy way of describing the 2012 Republican nomination race is to say that despite all the rules changes and all the calendar movement, it still played out pretty much like 2008. Super Tuesday came and went with one candidate well ahead of the others in the delegate count and a month later it was over. Of course, John McCain was way out in front of his rivals in 2008 after the February 5 Super Tuesday series of contests, but a month later -- after wins in Ohio, Rhode Island, Texas and Vermont -- the Arizona senator wrapped up the nomination. And that's just as it was in 2012. Romney emerged from Super Tuesday on March 6 with a sizable enough delegate lead and eliminated his final viable opponent a month later after wins in Maryland, Washington, DC and Wisconsin.

Now, the explanation is more complex than that. After all McCain surpassed the 1191 delegate mark to officially clinch the nomination a month after Super Tuesday, whereas Romney will continue to march toward 1144 in a semi-contested to uncontested way for the rest of the calendar. The point here is not to minimize that distinction. Rather, the intent is to point out that while delegate counting is fun -- more so for some of us than others -- often these contests for a party's nomination are more a process of elimination. Presumptive nominees don't often have to concern themselves with the sorts of gain-deficit ratios and other delegate calculations Barbara Norrander (2000) so eloquently describes in discussing the end game of nomination contests. No, more often than not, it is simply a matter of a frontrunner eliminating his or her final viable opponent (Norrander 1996).

We counted delegates for a while in 2012, but this one ended like so many other presidential nomination races of the post-reform era ended: with the runner-up withdrawing. In this case, Mitt Romney had established enough of a delegate lead that a Santorum comeback was unlikely if not impossible.


Recent Posts:
Cart Before the Horse: Pennsylvania/Colorado Edition

Maine Legislature Exploring Presidential Primary Option for 2016

More on Santorum Delegate Math and Some Thoughts on Texas as Winner-Take-All


Are you following FHQ on TwitterGoogle+ and Facebook? Click on the links to join in.

Friday, May 9, 2008

The Delegate Race: Is Obama There Yet?

Editor's Note: The following comes to us from University of Georgia political science professor, Paul Gurian.

In her 2000 article
"The End Game in Post-Reform Presidential Nominations", Barbara Norrander evaluated several indicators of when a nomination campaign is effectively over, specifically, when the frontrunner's last remaining opponent would drop out of the race. Two of these indicators are especially relevant at this point in time: the "gain-deficit ratio" (Collat, Kelley and Rogowski, 1981) and the "bandwagon curve" (Straffin, 1977).
These formulas are not precise predictors of when a candidate will drop out -- that is a decision made by the candidate and her staff. However they do indicate when conditions are such that the frontrunner's nomination seems inevitable -- conditions that figure into the opponent's decision-making.

Before the Indiana and North Carolina primaries, Obama needed 273 delegates to achieve a majority. To exceed the gain-deficit ratio threshold of .36, he needed to win 98.3 delegates. On Tuesday he won approximately 97 or 98 pledged delegates (estimates vary from 95-100). However, he also won about 10 super-delegates the next day. Whether the super-delegates are counted as part of the same "event" or not, Obama is very close to, or just above the threshold. (These numbers change if one assumes that the Michigan and Florida delegates will be seated.)

The Straffin formula suggests that Obama is still just shy of the "bandwagon" threshold. Considering the delegates won by each candidate, the current delegate margin (Obama's delegate total divided by Clinton's) is 1.09, not quite enough to exceed the current threshold of 1.14 (see below).

Although it is still possible for Clinton to win the nomination, there is an emerging consensus among reporters and politicians that the race is over. This perception coincides with the two mathematical indicators: it's not over yet, but it's very close.


Calculations:
BO 1854; HRC 1696 (as of 5/9)
total BO+HRC = 3550
total all dels = 4049
BO+HRC/totalDels = 0.876759693752
4.596 - 7.28 + 3.824 = 1.14
BO/HRC = 1.09

--posted by Paul-Henri Gurian (Associate Professor of Political Science, University of Georgia)


Related: (from fivethirtyeight.com...)

Everything you always wanted to know about Obama's pledged delegate clinching scenarios


Recent Posts:
ABC News: Obama Now Leads in Superdelegates

Obama's Slide Revisited

Unpledged Add-On Delegates