How Crucial Is It To Win Primaries In Swing States???

With Mitt Romney escaping Michigan bruised, but intact, his attention now turns to a perhaps more crucial contest—Ohio.  Among the Super Tuesday contests, Ohio is important not just for the number of delegates it will award but because the Buckeye State will garner intense interest in the fall.  Whereas President Obama’s polling numbers in Michigan have been quite strong, and he won the state by more than 16 points in 2008, Ohio is less friendly territory.

One question we might ask is whether there is a correlation between a candidate’s performance in a state’s primary and how they will fare in November. The ability to pivot from the primary to the general is a skill that all winning candidates must develop.  On one hand, there’s reason to doubt a clear connection between a state’s primary and general contest.  Primaries, we know, bring a much more ideological electorate to the polls.  A losing primary candidate may have been “too moderate” for the party faithful—but consequently more competitive in the more moderate fall electorate. 

On the other hand, primaries give candidates the opportunity to build an organization and campaign infrastructure that can be put to work in November.  Those candidates who can win primaries and caucuses are those who demonstrate the ability to build the massive organization that will be crucial to winning the general.  Much of this organization will be directed toward the larger general election audience once the nomination is secured.  If they fail at this during the primaries, they may fail at it during the fall.

Should a connection between the primary and general exist, it is of most importance in “toss-up” or “swing” states.    Mitt Romney’s primary loss in South Carolina will not—absent complete collapse—matter in the fall.  Clearly more important was what transpired the next week in Florida.  It’s almost impossible to conceive of a Romney (or Santorum for that matter) win in November that doesn’t include winning the Sunshine State. 

So what does history tell us???

Most recent nomination contests—with the exception of the Democrats in 2008—have wrapped up quite quickly.  With the winner rolling through state after state there haven’t been a large number of states that allow us to explore the question of whether candidates can bounce back from primary losses.  However if, as it now seems, the GOP contest is going to go on for a while, we should have the opportunity to dig into this phenomenon some more.  Despite the relative lack of test cases, there are some examples that jump out.

Looking at those competitive or “toss-up” states, we find relatively few instances in recent cycles where a candidate lost his party’s primary or caucus there, and then recovered to win the state in November.  The one exception to this is Barack Obama.  In 2008, Obama lost spring contests in New Hampshire, Nevada, Florida, Ohio, Pennsylvania, Michigan, and New Mexico.  Despite this, he won all of them in the general.  In this sense, the long nomination fight that forced Obama to build statewide organizations may have paid off in the fall.

John McCain, who wrapped up the GOP contests more swiftly and with fewer losses, failed to win any swing states that he also lost during the primaries.  In 2004, John Kerry swept to the nomination in even faster order with only a small handful of primary losses.  None were in states seen as competitive at the time.   In 2000, George W. Bush won one swing state he lost in the primary—New Hampshire—while losing another--Michigan.  Bob Dole, in 1996, lost Missouri during the nomination contest but bounced back to win it in the fall.  In 1992 Bill Clinton managed to lose 3 spring state contests that he put in his column in November—New Hampshire, Colorado, and Nevada.  Finally, in 1988, Michael Dukakis managed one of his few fall wins in Iowa, whose caucuses he lost.  On the flip side, whereas George H.W. Bush only lost nine states in the general, three came in normally competitive states that he lost during the primary season—Iowa, Minnesota, and Washington.

So, despite the small number of cases that fit our definition—primary losses in swing states--there seems to be pretty good reason for Mitt Romney to worry about next week’s vote in Ohio.   Like Florida, Ohio is a state that GOP badly needs in November.  Should he fail to defeat Rick Santorum there next week, the loss may prove to be more lethal than even Michigan would have been.

A Short Take on Florida, Including Ominous Turnout Numbers

Mitt Romney's victory yesterday in Florida is obviously a shot in the arm to his campaign, especially after the shock of South Carolina.  Above, I've posted a map of the primary results, courtesy of Dave Leip's U.S. Election Atlas.  The counties highlighted in Green were won by Romney; Blue were Gingrich victories.  To make some sense of the map, I'd refer back to a post I wrote in the weeks leading up to the 2008 election.

Florida's political geography is extremely fascinating.  The northern part of the state, including the panhandle, more closely resembles the neighboring states of Georgia and Alabama than it does the rest of the state.  More rural and with a large military presence, these counties have a stronger "Deep South" flavor--more Evangelicals and social conservatives.  Though less populous than other regions in the state, it favored Gingrich.  Exit polling from yesterday's vote confirms Gingrich's (and Santorum's) appeal to these voters and serves to confirm that Romney has still not sold this important GOP bloc on his candidacy.

On the turnout front, some more troubling news for the GOP.  As I wrote recently, turnout in New Hampshire, while up compared to 2008, did not increase at the rate we might expect for a party energized and positioned to recapture the White House.  I did some quick calculations on the most recent contests.  South Carolina saw an impressive 36% increase in Republican primary turnout over 2008 (603,856 votes vs. 445,677).  In Florida, however--a much more important state in November--turnout was actually down 14% compared to four years ago (1,669,585 votes vs. 1,949,498).

Here's some Florida turnout analysis (including an interesting graph of county data) from Michael McDonald, one of the foremost scholars of voter participation.

Does President Obama Have A "Cushion" In 2012???

As we get more and more polling data about how President Obama stacks up against his potential Republican rivals, it's important to remember that in many ways the national percentages are irrelevant.  Presidential contests are really state by state races.  The ultimate goal is to compile the 270 electoral votes necessary to win.  Because of the vast differences across states and regions--something that this site aspires to capture--neither Obama or Romney/Gingrich will have the same level of support everywhere.

Thus, as we begin to look to November, it's useful to look back at recent elections, especially 2008, and see how the candidates varied across the states.  This will give us the opportunity to see how likely it is that either candidate will be able to bring new states into their coalition. 

Another way to state this is: how much ground must the Republican nominee make up based upon what happened in 2008?  How much of a "cushion" does Obama have?  Does he have any realistic opportunity to build upon his 2008 margin?

To help answer this, I decided to look at how many states were actually closely decided in 2008.  These would be the main targets for both candidates, especially the Republican nominee who needs to improve dramatically on John McCain's 173 electoral votes.  I produced the following table that lists each state based upon the winning candidate's margin of victory.

What we see is that, beyond the 7% national spread between Obama and McCain, the state by state results are even more impressive for the Presdident.  If we use a spread of 5% as an arbitrary definition of a "close" outcome, we see (highlighted in yellow) that only six states were decided by such a margin in 2008.  Of these, Barack Obama won 4 (NC, FL, IN, OH) while McCain won 2 (MO, MT).  If we wanted to be a bit more generous in our definition of "close" to include states decided by 10% or less, we get an additional nine states, 4 won by Obama (VA, CO, IA, NH) and 5 won by McCain (GA, SD, AZ, ND, SC). 

In the final column of the table, I've listed the number of electoral votes that will be awarded by these states in 2012.  Here is where we can get a real sense of the magnitude of the task for the Republican nominee.  If we assume that states in 2012 will vote roughly as they did four years ago, the GOP nominee must win every state they won in 2008, plus North Carolina, Florida, Indiana, Ohio, Virginia, Colorado, AND Iowa in order to caputre the White House.

When we look at previous presidential elections, we see that there were many more "close" states than we saw in 2008.  For example, in 2004 we saw twelve states decided by 5% or less and twenty one with a 10% or less margin:

In 2000 there was a similar bunching of states, also with twelve decided by 5% or less.  Twenty two had a 10% or less margin:

When we look at the data on a state by state basis, the magnitude of each party's win over these past three cycles becomes magnified.   This is especially true, it seems, for 2008.  While Obama's 52.9% of the popular vote was the highest of any Democratic nominee since Lyndon Johnson in 1964, it is also true that he managed to win a lot of states by a large margin.  One might miss this if they were concerned only with the national numbers.







Is There An Enthusiasm Gap Among Republicans???

Things have been dark here for many months.  Now that the campaign is heating up, I'm going to try getting some things up on a more regular basis.  I've got a couple of posts in the works but let's start with a short little data exploration.

With two Republican votes already in the books and another taking place in South Carolina this weekend, one question that has gotten a bit of attention is whether Republican voters are enthusiastic about their choices, especially now that the field is winnowing.  For any party hoping to win the presidency--or any other election for that matter--turning out your voters is of primary importance.  The assumption going into 2012 for Republicans was that given the degree of opposition on the right to the Obama presidency, and coming on the tails of their success in the 2010 midterms, there would be tremendous energy and activism mobilized to propel whoever won the nomination into the White House.

While we've only had a few contests so far, there is reason to wonder whether this assumption is in fact true. 

If we look at the results from New Hampshire, a total of 248,447 votes were cast in the Republican primary across all candidates, more than in any recent Granite State GOP primary.  When compared to 2008, this year's vote was an increase of 3.6%.

To get a sense of whether this increase is significant or tells us anything about the state of the GOP electorate, though, we need some baseline of comparison.  I decided to look at recent New Hampshire primaries in which one party was trying to take over the White House from the other--a scenario that would seem to be ripe for increased turnout and mobilization.  When we look at these contests, 2012 doesn't stack up well.

For example, in 2008 the Democrats saw a 31% increase in turnout over 2004 (287,556 vs. 219,787 votes).  Also on the Democratic side--and also a successful party flip of the White House--1992 saw the Democrats increase their turnout by an even more impressive 36% over 1988 (167,664 vs. 122,912 votes).  Looking at Republicans, in 2000 the GOP turnout was 16% higher than it was in 1996 (238,206 vs. 205,856 votes).

Unlike in later contests where the eventual nominee becomes established and most candidates have dropped out, New Hampshire primaries have full fields and permissive voting procedures--it is an "open" primary.  Thus, we would expect the voting there to be a relatively good barometer of the party's enthusiasm.  If the turnout results in New Hampshire continue into the later contests, there's reason for GOP leaders--and the eventual nominee--to worry about the fall.