Wednesday, July 15, 2009

Revisiting Democratic Delegate Allocation (1976-2008)

A few weeks ago when the Democratic Change Commission was holding its first meeting, FHQ posted a series of graphs the DNC produced to show the frontloading of delegate allocation over time. As I said then, "That, folks, is the impact of frontloading in a nutshell." To see the shift from 1976 to 2008 is somewhat staggering.

...well, if only the graphs were a little better. I wanted to see what those figures would look like if a line was added to account for the cumulative percentage of delegates throughout any given year's primary calendar. Ideally, you'd see a nearly even distribution of contests across primary season and a relatively straight line from 0 to 100% delegates allocated.

And that is essentially what is demonstrated in the recreation of the 1976 figure from the DNC (see below). There is some undulation, but basically there is a fairly even (linear) growth to that cumulative line. The same is true of the 1980 figure (not shown). The blue area, then is the weekly percentage of delegates allocated, while the red area is all the weeks to a given point stacked on top of each other.

[Click to Enlarge]

But in 1984, what starts is what I'll call the "volcano with a wind out of the west" phenomenon. [That's a long way of saying frontloading.] What popped up in 1984 was a burst of delegate allocation activity toward the beginning of the process. And over time that "volcano" has grown from a small hill to the towering mountain of Super Tuesday in 2008 (seen below).

[Click to Enlarge]

These are handy visuals that would fit right in (with a GOP version as well) on the monthly frontloading maps that adorn the left sidebar. Assembled, they basically tell the tale of frontloading since the McGovern-Fraser reforms took effect.

*I mentioned the 1984 figure, and should add that I'll be putting the entire series up at some point (probably in individual posts as time allows). In saying that I should also say that there is a problem with the 1988 and 1992 figures. If you look closely at the originals in the link at this post's outset, you'll see that those two years' patterns (and underlying data) mirror each other exactly. If however, you look at the calendars from 1988 and 1992, you can clearly see that they are similar, but not the same, calendars. At some point I'll have to fix that (I suspect the 1988 figure is the one that is off. Pay close attention to the hump on the left side of the Super Tuesday peak. That's Georgia and Massachusetts and the other states that jumped to the first week in March when the Democratic window expanded to include that week in 1992. There was no similar group of contests -- not in terms of numbers of delegates -- in 1988. Most of those southern contests were on the second Tuesday in March.)

Recent Posts:
On the Polling Horizon: Louisiana 2012?

State of the Race: Virginia (7/15/09)

North Carolina in 2012: Obama - 49, Palin - 42

No comments: