Romney vs. Perry: How the Numbers (and the Calendar) Stack Up

By Sean Trende - August 16, 2011

‹‹Previous Page |1 | 2 | 3 | Next Page››

Question 2: What Does the Final GOP Primary Calendar Look Like?

In 2008, the primary schedule played a larger role in determining the Democratic nominee than many like to remember or admit. Florida and Michigan advanced their primary dates into January, were denied delegates by the Democratic Party, and cost Hillary Clinton badly needed pledged delegates. Likewise, had Super Tuesday been followed by voting in pro-Clinton states like West Virginia, Pennsylvania and Ohio, rather than such pro-Obama states as Louisiana, Maryland and Virginia, the momentum Clinton would have gained might have been too much for the Obama campaign to withstand.

A similar situation could play out for the GOP in 2012. Now, let’s first acknowledge that if Perry wins New Hampshire, or Romney takes South Carolina or Iowa, or Bachmann wins either of those two states plus Iowa, it could effectively end this whole thing before we even get to Super Tuesday. Republicans have historically coalesced behind a candidate fairly early on. If a candidate ends up unifying the Republican Party, or the other major candidates implode before Super Tuesday, then much of what follows is academic.

But let’s assume, for purposes of discussion, that Bachmann, Perry and Romney end up splitting the GOP much as Clinton and Obama did. In such a drawn-out contest, understanding the calendar will be key to understanding the nomination battle. This is all a bit dodgy at the moment, because states are still toying with where to place their primaries. But for now, let’s break the voting into three phases: January/February, March, and April/May/June, using data from The Green Papers and from

The following chart shows the primaries and caucuses presently scheduled in January and February. The states are sorted by number of delegates. The four right-hand columns show how the Republican primary electorates in these states broke down in 2008, according to exit polls. It shows the percentages that identified as evangelical conservatives, as “very conservative,” “somewhat conservative,” with a combined column for “moderate,” “somewhat liberal,” or “very liberal.” In a few states, older exit-poll data was used to fill in blanks, although this doesn’t become an issue until the later primaries. For some states, no data at all could be located (these states aren’t included in the weighted average); their lines are blank. (The numbers add up to more than 100 percent because religious affiliation is a separate question from ideology in the exit polls; the “very conservative/somewhat conservative/moderate” columns add up to approximately 100 percent.) 

Most of these data are presented simply for reference. There are two things you really need to pay attention to here. First, note the states with asterisks. The RNC has decided to strip half of the delegates from any state that holds a primary or caucus before March 1, other than Iowa, New Hampshire, South Carolina, or Nevada. Some states are considering pushing their primaries back, although these also tend to be the more moderate states, like Wisconsin and New Jersey. The more conservative states seem to be hanging tough, for now. In other words, you could end up with some of the more conservative states in the GOP electorate losing clout at the convention.

Second, and more importantly, take a look at the bottom row. It shows the weighted average of the delegates for each group of states. To put it differently, if all the moderates and liberals vote for the same candidate in January/February, all the “somewhat conservative” voters vote for the same candidate, and all of the “very conservative” voters cast ballots for the same candidate, the “very conservative” candidate would get roughly 30 percent of the delegates, while the “moderate/liberal” candidate would get 35% of the delegates. This would never happen, of course, but this just offers a rough guesstimate as to what the primary electorate looks like here, for reasons that will become apparent in a minute.

In March, it is more of the same: 

The electorate becomes a bit more conservative, and a lot more evangelical. If we had data for Alaska, Hawaii and other states, we might expect the evangelical number to be diluted a bit. On the other hand, Pat Robertson won Alaska’s caucuses in 1988, so their composition is hard to predict with certainty. Now, let’s turn to April: 

‹‹Previous Page |1 | 2 | 3 | Next Page››

Sean Trende is senior elections analyst for RealClearPolitics. He is a co-author of the 2014 Almanac of American Politics and author of The Lost Majority. He can be reached at Follow him on Twitter @SeanTrende.

Mitt Romney for Mayor
Carl M. Cannon · November 16, 2014
Perry in N.H.: Preparing WH Bid, Will Decide in '15
Kathleen Ronayne · November 11, 2014

Latest On Twitter