As annoying as polls are perceived to be, they do provide a relatively neutral narrative as to how a race is progressing. Additionally, politics is like sports in that people want to know who’s up or down at a given point in time. Therefore, it’s important to understand what polls are and how they work, since the reality is, polls can shape the trajectory of a race.

Polling – the basics

A poll is, quite simply, a representative sample of voters at a given point in time. What leads to different results is the fact that getting that “representative sample” is a very subjective process. 

In fact, it can be argued that a significant driver of the ultimate results is the content of that “representative sample.” The easiest way to sample (and the least accurate) is a list of phone numbers. However, this method immediately brings data quality issues to the forefront. Even assuming that the person at the other end of the line is a registered voter, can we be certain that the voter is eligible to vote in the election? For instance, elected officials not elected statewide are elected from a geographical area that varies in size from a handful of precincts to regions of the state. Without understanding the boundaries of the applicable election district, there is a risk that voters will be asked to participate in a poll that is essentially irrelevant to them. Plus, political campaigns have been embarrassed when they’ve sent mail to voters not living in a district.

Another way of sampling voters is use of a voter file to call ONLY those voters living in a district or state. While this method of polling eliminates nonvoters (or irrelevant voters) from the equation, there are disadvantages to this method as well: (1) if the voter list hasn’t been updated, voters who have recently registered (or who moved into a fast growing area) will not be sampled, (2) depending on the type of election (Presidential, statewide, special election, tax election), a pollster may be calling people who have no intention of voting in that particular election.

Finally, pollsters can limit their sample to “likely” voters. Even though this is a subjective term (for example, you have people who vote in every election as opposed to those who only vote in Presidential contests), the idea is to evaluate a voter’s history before deciding whether that voter should be included in a poll sample. Alternatively, pollsters can use a “prescreening” question to ask the respondent how likely he/she is to vote in the upcoming election. Those who are not sufficiently interested are not included in the results.

Therefore, when evaluating a poll release, it’s important to determine if the poll was of registered or likely voters, because those who are merely “registered” are less likely to show up at the polls and vote. In fact, in the state of Louisiana, 17% of those on the voter rolls have never voted, while another 11% last voted before the 2008 Presidential election. From the author’s standpoint, it would not make sense to poll these voters.

It’s also worth mentioning, too, that regardless of the method used to sample voters, there are data quality issues on the “back end” as well: not all demographics participate in a poll equally, and this reality can and does skew the results. To illustrate: the Louisiana electorate as of May 1, 2012 was approximately 30% black. If a poll sample were done where the respondents were 20% black, you have a poll that, depending on the election, may be improperly weighted towards a Republican or conservative candidate, and should not be considered a credible poll.

Ways to evaluate a poll

We have already mentioned above that the type of sample employed by a pollster is something that should be considered when evaluating a poll. There are additional factors which should be considered when critiquing a poll:

(1)    Who released the poll: In other words, was it a candidate/interest group (business, labor, etc.) who released numbers favorable to his/her campaign, or was it a (presumably) more objective media outlet (newspaper, TV station, magazine) paying for the poll ? And if the media outlet paid for the poll, which polling firm conducted the poll?

(2)    When the poll was conducted: A poll conducted two days ago is a very good assessment of a candidate’s strength/weakness. On the other hand, a poll conducted in January would pretty much be useless, because (a) of the “staleness” of the data, and (b) the further along you are in campaign season, the tempo of a campaign picks up, and voters can and do change their minds;

(3)    Trends are important: If the same pollster is polling the race multiple times, and there is clear movement towards a candidate, that trend is important to note. Similarly, if multiple pollsters polling at the same time are showing movement towards a candidate, that “collective movement” is also worth noting;

Our method of analyzing polls

We believe that a collection of polls on a given race tells a more complete story than an individual poll. Therefore, as the election season progresses, we periodically review the polling taken on Presidential and Congressional races, and we load this data into a database. We then take an average of all polls for that race over a period of time. Right now, we examine the last 28 days of polling, but closer to Election Day, that “window” shrinks to 14, or even 7 days of polling, since public opinion becomes more fluid the closer you get to Election Day.

Upcoming

In the next installment of this article, we will discuss the performance of pollsters in several selected races in the 2010 election cycle, and we will begin reporting on polling that is already underway for the 2012 election cycle. 

Current Generic Congressional Vote (14 day rolling average): 44-43% Republican

Prior Generic Congressional Vote (14 day rolling average): 42-40% Republican

The anti-incumbent wave which in 2010 cost 63 Congressmen (7 in the primary and 56 on Election Day) their jobs showed no signs of abating after last night’s primaries in Indiana, North Carolina, and West Virginia. The latest casualty was 36 year incumbent Republican Senator Richard Lugar (he last had a competitive race in 1982). He could only receive the support of 39% of Indiana Republicans, and his poor showing was due to a combination of things: his lengthy Senate tenure, his lack of an Indiana presence (he has not lived in Indiana in decades), and a perception that he was too accommodating to President Obama specifically and Democrats in general.

This anti-incumbent wave has consistently been more pronounced with Republican primary voters – even though the six primary casualties have come equally from the Democratic and the Republican ranks, all but one of the Democratic defeats have been due to reapportionment. Furthermore, in addition to the three Republicans who have been defeated so far, seven more Republican incumbents have been renominated with less than 60% of the vote. And 41 states haven’t held their primaries yet.

Are these six primary losses an especially large number? It seems that way, although we are still in the early stages of “primary season”. In 2010, seven incumbents (three Democrats and four Republicans) were defeated in their primaries, while the anti-incumbent year of 1992 produced 15 primary defeats (12 Democrats and 3 Republicans). 

For months, the Republican Presidential contest has overshadowed these Congressional primaries, although with the Republican field effectively cleared out (since the last article, both Rick Santorum and Newt Gingrich are no longer contesting the race), we can now turn our focus to these races. And almost by coincidence, this will be a busy month for statewide primaries: there are six more primaries to be held before May 31, with three (Idaho, Nebraska, and Oregon) coming up next Tuesday.

Similarly, Congressional filing has been relatively quiet recently, although 32 states have concluded their qualifying. The next filing deadlines will be on May 15 in Michigan and May 18 in Washington State.  Thus far, 48 House members (28 Democrats and 20 Republicans) and 11 Senators (7 Democrats and 4 Republicans) are not returning this fall. This group of 59 Congressmen is actually a larger group than the 52 who retired in 2010, but is still less than the 64 who retired in 1996 (the 1994 GOP landslide triggered a wave of retirements from (mostly) Democrats who did not want to be in the minority). Similarly, it’s worth noting that 18 more states have not concluded their filing deadlines, so there may very well be additional retirements.

Presidential

Current President Obama Job Approval (14 day rolling average): 48-49%

Prior President Obama Job Approval (14 day rolling average): 47-48%

Equally as newsworthy as the Congressional results were the Presidential results. Republicans have quickly coalesced behind Mitt Romney: now that only Ron Paul is in the race, Romney’s winning percentages have finally spiked – his weakest showing last night was the 65% of the vote he received in Indiana.

What was newsworthy were the results on the Democratic side. We had noticed for some time that in the handful of states that have bothered to hold a Democratic primary, President Obama has not been the unanimous choice of his party’s voters, and his relative strength depends on whether he has a named opponent or not. In the first table (shown below), we have ranked Obama’s percentages in the states where he ran against “Uncommitted.” In general, his numbers were respectable, but not unanimous, and last night’s results in North Carolina (where he received 79%) should give the Obama campaign cause for concern, since this was a state he unexpectedly carried in 2008, and even then, it was by a narrow margin. Having 20% of Democratic primary voters against you in that context means that this state is almost certain to vote for Romney in November.

Where Obama’s problems really come into focus is when he has had a named opponent (or opponents), which we have detailed in the table below – the West Virginia results were particularly eye opening:

President Obama should be concerned about the West Virginia numbers for two reasons: (1) he was running against a convicted felon who is in jail in Texas, (2) President Obama’s level of support even in West Virginia was uneven: he lost 10 rural counties, and if you look at the counties not containing government employees or universities (where 82% of the vote was cast), Obama only received 57% of the Democratic primary vote.

Granted, President Obama was not expected to carry West Virginia (this was one of his weakest primary and general election states in 2008, and his popularity hasn’t improved since then), but the Democratic voting demographic in that state (rural, blue collar, unionized) is also present in portions of Ohio and Pennsylvania – states Obama needs to carry this fall.

In coming weeks, we will look at the general election map, since it is important to follow the Presidential race from where it really counts: the individual states.

(1144 delegates to win) Romney 658 (58%), Santorum 281 (25%), Gingrich 135, Ron Paul 51, Huntsman 2 (Source: WSJ)

Presidential Contest 4/3 (Romney = blue, Santorum = pink, Gingrich = red, upcoming = white)

Primary results/looking ahead

The Republican Presidential nomination contest has become increasingly predictable, as the outcome of a given state can now be predetermined by a state’s geographic location and/or its demographics. Two states and the District of Columbia held their primaries this past Tuesday, and Mitt Romney was the victor in all three states (he defeated Santorum 49-29% in Maryland, 44-37% in Wisconsin, and received 70% of the vote in Washington DC). While the Romney victories on Tuesday were not a surprise, the patterns of support for Romney in Wisconsin were eerily similar to Illinois, and unless those patterns can be reversed, Santorum will be unable to compete effectively with Romney.

Statewide, Romney’s Wisconsin margin was 7 points, but his entire margin of victory came from Milwaukee and its suburbs – his 58-31% margin over Santorum was very similar to the 55-29% lead Romney had over Santorum in the Chicago suburbs in last month’s Illinois primary. Similarly, “outstate” Wisconsin (i.e., that part of the state outside of Milwaukee) actually preferred Santorum by a 1,600 vote margin, while in “outstate” Illinois, Santorum eked out a 43-39% victory over Romney.

In simple political terms, this means that Santorum has a solid base in the Deep South, the Great Plains states, and the rural areas in the industrial Midwest. However, his rural strength there is not strong enough (in terms of raw votes and percentages) to overtake Romney’s dominance in the urban areas. This electoral landscape will be forbidding for Santorum in the near future, as the next primaries are on April 24 in more liberal northeastern states (Rhode Island, Connecticut, New York, Pennsylvania, and Delaware). Santorum has a shot in Pennsylvania (his home state), but a composite of recent polls shows Santorum with only a 40-38% lead there.

So by April 25 (the day after the Northeastern primaries), Santorum faces a dilemma. There will likely be pressure for him to get out of the race, given that Romney would have had a month’s worth of favorable primary results, even though the reality is, the contests were on Romney friendly turf. Equally as complicating for Santorum is the fact that the eight primaries to be held in May will be on (for Santorum) friendly turf where he has a chance to dominate in at least seven of those states.

Turnout and President Obama – While the media narrative is that the Republican primary electorate is dispirited, the truth is more complicated. There has consistently been an energized GOP turnout in states where there were competitive contests. In fact, in Wisconsin, turnout was 94% higher than it was in 2008, while in Maryland (which was not considered a competitive primary state), Republican turnout was 26% lower than it was four years ago.

We have also noticed that when President Obama is on the Democratic primary ballot and his opponent is “uncommitted”, he typically wins near unanimous percentages. With named opponents, his percentage is somewhat lower. Since President Obama ran against “uncommitted” in Wisconsin and Maryland, his percentages were naturally near unanimous (88% in Maryland and 98% in Wisconsin), but it’s worth remembering that he got less than 60% in Oklahoma against named opponents, while in Louisiana, that percentage was 76% (36% outside of Orleans and East Baton Rouge parishes).

Delegates

In the delegate count compiled by the Wall Street Journal, Romney has a 58-25% lead with 49% of the delegates chosen. This means that Romney only needs 42% of the remaining delegates to get the nomination, and Santorum has to get 75% of the remaining delegates. If Newt Gingrich’s delegates were all to go to Santorum, the 75% threshold is lowered a bit – to 64%

(1144 delegates to win) Romney 568 (55%), Santorum 273 (27%), Gingrich 135, Ron Paul 50, Huntsman 2 (Source: WSJ)

Presidential Contest 3/24 (Romney = blue, Santorum = pink, Gingrich = red, upcoming = white)

Louisiana Primary – the Republican contest

The South is now Santorum country. This was not apparent when Newt Gingrich was sweeping the South Carolina primary back in January, but starting with the Alabama and Mississippi contests, enough of the 60-65% of the Republican vote that has consistently been “anti Romney” had begun to move to Santorum to deny Gingrich wins in those states. This movement continued before the Louisiana primary: 65% of Louisiana Republicans again voted “anti Romney”, and Rick Santorum racked up an impressive 49-27% vote over Mitt Romney (the remaining 16% of the anti-Romney vote went to Newt Gingrich). This was a broad based victory: Santorum carried 63 out of Louisiana ‘s 64 parishes (Romney carried Orleans Parish), and if you look at the elections through the prism of the urban/suburban/rural split that is prevalent in these contests, you begin to appreciate how broad Santorum’s Southern support is now.

pink = Santorum <50%, red = Santorum 50-60%, purple = Santorum 60% or more

The “urban cores” of Orleans and East Baton Rouge Parish not only have large black populations, but have a growing “garden district liberal” constituency. Accordingly, this is where Mitt Romney ran the strongest, although Santorum still defeated Romney 41-34% here. The suburban parishes of Baton Rouge and New Orleans were even more solidly in Santorum’s camp: he was preferred over Romney 47-28%. In the smaller cities and rural areas, Santorum dominated: he led Romney 52-24%.

Santorum’s dominance in the South directly impacts Newt Gingrich. He only received 16% of the vote statewide, and his performance was nearly identical in the urban areas, suburbs, and smaller cities/rural areas. This is about half of what he received in Alabama and Mississippi, and about a third of his performance in South Carolina and Georgia.

Turnout and President Obama – We have consistently noted a strong correlation between the perceived level of competition in each state and the strength of Republican turnout relative to 2008. Even though it was widely believed that Santorum would carry Louisiana, the major candidates invested some time here, and voters responded accordingly: the 186K turnout on the Republican side was 15% higher than it was in 2008. About 12% of the electorate voted early.

(REVISED 3/25 PM) We have also noticed another pattern: when President Obama is on the Democratic primary ballot and his opponent is “uncommitted”, he typically wins near unanimous percentages. With named opponents, his percentage is somewhat lower (in fact, he received less than 60% of the vote in Oklahoma against minor primary challengers). In Louisiana, the 76% he received is mildly impressive, until you realize that his support was uneven. While his coalition of white liberals and blacks enabled him to get 93% of the vote in the “urban cores” of Baton Rouge and New Orleans (included in that total was 75% of the white vote in those parishes), suburban and rural white Democrats only gave President Obama 36% of their votes. This primary vote (which, it needs to be remembered, is coming from party activists who would actually vote in a party primary) shows the extent to which President Obama is broadly unpopular among the white electorate in this state outside of Baton Rouge and New Orleans.

tan = Obama <50%, light blue = Obama 50-60%, medium blue = Obama 60-70%, dark blue = Obama 70% or more

Looking ahead/the “delegate race”

Republican primaries have become predictable affairs. For the trajectory of the race to change, Santorum has to surprise Romney in at least one of the states thought to be in the Romney column. On paper, the April 3/24 contests appear unfavorable to Santorum (6 of the 7 contests held will be in the Northeast/New England, and the remaining contest will be in Wisconsin). On April 3, Maryland, Wisconsin, and the District of Columbia vote. Santorum needs to win Wisconsin to break the string of Romney victories in the Upper Midwest.

What about the delegate math? So far, Romney has a 55-27% lead over Santorum in the delegate count (as compiled by the Wall Street Journal) with 45% of the delegates chosen. For Romney to reach the magic number of 1,144, he needs to get 46% of the remaining delegates. Santorum has to get 69% of the remaining delegates to reach 1,144 by himself, although if his/Gingrich’s delegates were combined, that number would fall to 58% of remaining delegates.

• (1144 delegates to win) Romney 563 (56%), Santorum 263 (26%), Gingrich 135, Ron Paul 50, Huntsman 2 (Source: WSJ)
• President Obama Job Approval 47-48% (14 day rolling average)

Presidential Contest 3/21 (Romney = blue, Santorum = pink, Gingrich = red, upcoming = tan)

Illinois Primary

Now that the Presidential race has essentially become a two man race between Mitt Romney and Rick Santorum, it is becoming obvious that each candidate has a reliable voting base: Romney is the candidate of moderate/liberal Republicans (particularly in the urban cores) who are more inclined towards economic issues. These Republicans tend to dominate Republican electorates along the Pacific/Atlantic coasts, as well as in the industrial Midwest. Rick Santorum’s strength is with social issue conservatives, as well as voters in smaller towns/rural areas in the vast interior of the country. Originally, the Deep South was solidly in Newt Gingrich’s corner, but Santorum’s twin victories last week in Mississippi and Alabama have made the Deep South “Santorum Country” now.

Last night’s contest was in Illinois, and Illinois is one of several states (like New York) where conservative Republicans have never had much strength, even in the primaries. This is largely due to Illinois’ having an urban dominated electorate: the vast majority of the vote is cast in “Chicagoland” (Chicago and adjacent suburbs). When economic issues were front and center, this area was competitive for Republicans, and in fact Republicans carried Illinois in every Presidential election between 1968 and 1988. Since then, social issues have been more predominant, and Illinois (thanks to large Democratic margins in Chicagoland) has become solidly Democratic in every Presidential election since 1988.

This is context under which Illinois’ Republicans voted. Statewide, Romney defeated Santorum 47-35% by amassing a 55-29% lead in Chicagoland, where 53% of the vote was cast. However, it didn’t hurt that Romney essentially split the Downstate vote with Santorum (Santorum could only eke out a narrow 43-39% win outside of Chicagoland). Curiously, the southern part of the state (roughly everything south of the state capitol of Springfield) is similar in attitude to the Deep South, and accordingly, Santorum’s strongest percentages were in this area.

Voter Turnout

We have noted that a state’s voter turnout (compared to 2008) in Presidential  contests has become fairly predictable: there is strong correlation between the perceived level of competition in each state and the strength of Republican turnout relative to 2008. In this case, the Illinois vote was 1% higher than it was in 2008, as both of the major candidates heavily contested the state.

Louisiana primary

This Saturday (March 24), Louisiana has its primary, and given the demographic terrain (Deep South, substantial rural population, and an active religious conservative voter bloc that is firmly behind Santorum), we expect Santorum to win fairly easily, although it will be interesting to see what kind of residual strength Newt Gingrich has – we have noticed that in Deep South primary contests: the “anti Romney” vote (in other words, the combination of the Santorum and the Gingrich vote) in those states was 60-65%, while Romney has consistently been getting 25-30% of the GOP primary vote.

The signs seem to point to a Santorum victory here: Louisiana has concluded its in person early voting (mail in absentees can still be accepted up until Friday), and as of last night, 20,337 Republicans have already cast their ballots. Thus far, the early voting has been strongest in parishes that favored Mike Huckabee in 2008 (he carried the state 43-42% over John McCain in 2008), and it’s reasonable to assume that the Huckabee vote is transferable to Santorum.

What does this early voting volume mean in terms of expected turnout ? While the early voting/absentee turnout is more than double the 9,214 who voted early in the 2008 Presidential primary, we do not believe that the eventual voter turnout will be twice the 2008 figure of 161K. In person early voting is a relatively recent policy, and it is gradually becoming more popular as time goes on (16% of those voting in the 2011 statewide elections voted early, as opposed to 6% in the 2008 primary). We think that voter turnout will be more like 180K, which suggests that about 13% of the final vote will have been cast early.

Looking ahead and the “delegate race”

While Romney’s Illinois primary win puts him in the driver’s seat, it is important to note that “game changers” can occur. From a surface glance, the 7 primary contests in April that follow the Louisiana contest favor Romney, since 6 of the 7 contests held will be in the Northeast/New England (the remaining contest will be in Wisconsin). While we assume that Romney would win at least 5 of those contests, it’s easy to forget now that Santorum emerged as the “anti Romney” by carrying Minnesota, Colorado, and Missouri on the same night, when the presumption at the time was that those states were in Romney’s corner.

What about the delegate math? So far, Romney has a 56-26% lead over Santorum in the delegate count (as compiled by the Wall Street Journal) with 44% of the delegates chosen. For Romney to reach the magic number of 1,144, he needs to get 46% of the remaining delegates. Santorum has to get 69% of the remaining delegates to reach 1,144 by himself, although if his/Gingrich’s delegates were combined, that number would fall to 59% of remaining delegates.

Congressional/Statewide –

• Generic Congressional vote: 44-42 Republican/Democrat (14 day rolling average)

Illinois became the fourth state to hold its Congressional primaries last night. There were no real surprises, although 20 year Republican incumbent Don Manzullo lost in the Republican primary against freshman Republican Adam Kinzinger – their districts were combined during redistricting. Two defeated Democratic incumbents tried to regain their old seats, and one of those two (Bill Foster) was successful, at least among Democratic primary voters.

The Congressional primary calendar has been relatively quiet so far, and this lack of activity will continue into April: only two states (Maryland and Pennsylvania) hold primaries next month.

On the Congressional qualifying front, this has been a very busy month. Maine, Utah, Nevada, Iowa, and New Mexico concluded their candidate qualifying. There were no real surprises there, although we now know from more complete information from California and Texas that a significant number of districts escaped partisan composition. In Texas, four Republican and one Democratic district escaped partisan competition: in California, (despite an independent commission’s drawing the district lines), 10 Democratic and two Republican districts are guaranteed to stay in their respective partisan columns. This means that 26 districts (14 Democratic and 12 Republican) are “safe” for the respective party. The Congressional playing field will become clearer by the end of this month, when a total of 26 states will see their candidate qualifying finish – Missouri, South Carolina, South Dakota, and Virginia will see their Congressional filing conclude next week.

There were some recent additional Congressional retirements in New York and Washington State, which brings the retirement count to 44 House members (26 Democrats and 18 Republicans) and 10 Senators (7 Democrats and 3 Republicans).

We are now half way through in person early voting – it started this past Saturday and will continue until next Saturday. After that, mail in absentee ballots will still be accepted up until the day before Election Day. From examining early voting statistics provided to us by the Secretary of State, there are two things we noticed:

Low turnout numerically, but better than 2008

Yesterday’s cumulative turnout was 19,640, of whom 8,929 were Republicans. What does this early voting volume mean? Now that we are half way though, we expect that volume to pick up on Friday and/or Saturday, so that by Election Day, there will be about 22K early votes – the drop-off in early voting between Saturday and Monday was greater than we had forecast, which is why we adjusted our initial estimate of 25K. Still, when you compare this number against the 9,200 who early voted in the 2008 Republican Presidential primary, it appears that  voter turnout would be a little more than double what it was in 2008.

However, it is unlikely that Republican turnout will be more than double the 2008 count, because in person early voting is a relatively recent concept in Louisiana (before the law was changed, you could only vote absentee before Election Day under certain circumstances). In the 2008 primary, 6% of the Republican primary vote was cast early. Since early voting has become more popular over time (16% of those voting in the 2011 statewide elections voted early), we think at the present time that voter turnout will be between 14 and 28%, or a turnout of between 109-218K Republicans (for comparison’s sake, 161K voted for the Republican candidate in the 2008 primary)

Santorum boost?

As part of our analysis of the early vote, we looked at where (i.e., which parishes) the early votes were coming from. In 2008, Mike Huckabee narrowly defeated John McCain 43-42%. His victory was achieved by running up large margins in the more rural areas outside of the Baton Rouge, New Orleans, and Lafayette media markets. In fact, the parishes he carried only represented 45% of the total vote cast and 47% of the early vote cast.  The early voting so far shows that 55% of the early vote cast came from the “Huckabee parishes.”  While this is slightly down from the 57% figure we reported after Saturday’s voting, the indications are still that the evangelical voters who supported Huckabee in 2008 are more likely to vote for Santorum than Gingrich (or even Romney) – especially since Santorum’s dual victories in Mississippi and Alabama give him credibility with Southern voters now, and the strong early vote in the “Huckabee parishes” benefits Santorum’s candidacy. Below is a graphical depiction of the 2008 vote by parish in the Republican primary. We will re analyze this data once in person early voting has concluded.

Red = Huckabee, Blue = McCain

Conclusion

Why do we make a big deal about early voting? When the Legislature essentially established “no fault” early voting several years ago, you now have a noticeable constituency of people who prefer the convenience of early voting, and this constituency has thus far ranged from 6 to 16% – a politician would be foolish to ignore this many voters, especially in a closely contested race. Also, too, early voting numbers are the first ones that are typically reported after polls have closed at 8 PM.

(1144 delegates to win) Romney 495 (53%), Santorum 252 (27%), Gingrich 131, Ron Paul 48, Huntsman 2 (Source: WSJ)

President Obama Job Approval 47-49% (14 day rolling average)

Presidential Contest 3/14 (Romney = blue, Santorum = pink, Gingrich = red)

Alabama, Mississippi, and Hawaii

The tempo of the Presidential campaign changed last night. We had noted in our previous analysis that  “…Frontrunner Mitt Romney’s coalition includes moderate/liberal Republicans on either coast, affluent urban neighborhoods, Western voters, and Mormons. Rick Santorum now has a national base of religious voters, middle income suburbs, and small towns, with one important exception: portions of the Deep South with a large percentage of native Southerners – that voter bloc is still solidly behind Newt Gingrich….”  This pattern has mostly been set in stone with each contest, and was further reinforced by Santorum’s defeating Romney 51-21% in the Kansas caucuses on Saturday, while in last night’s Hawaii caucus, Romney returned the favor by defeating Santorum 45-25%. However, our original presumption that Newt was unshakable in the Deep South (especially in areas with few “outsiders”) was rebutted last night with twin Santorum victories in Alabama and Mississippi.

With the contests we have seen in the Deep South thus far, we have noticed that the “anti Romney” vote there has consistently been in the 60-65% range, while Mitt Romney has remained in the 25-30% range. Before Rick Santorum was considered a viable candidate, that “anti Romney” vote was dominated by Newt Gingrich, which is how he carried Georgia and South Carolina. What changed in Alabama and Mississippi last night was that enough of the “anti Romney” vote (which was, coincidentally, 64% in both states) migrated into the Santorum camp to enable him to win both states. Assuming this movement of the “anti Romneys” towards Santorum continues/accelerates, Santorum will be in an excellent position to win Louisiana when it holds its primary next Saturday, March 24.

How was Santorum able to win both states ? In each state, the small towns/rural areas outside the major metropolitan areas (Mobile/Birmingham/Montgomery in Alabama and Jackson/Gulf Coast/Memphis suburbs in Mississippi) cast about 60% of the vote, and in those counties, Santorum dominated, garnering 37% in Alabama (Gingrich got 30% and Romney 27%) and 34% in Mississippi (Gingrich got 33% and Romney 29%). The urban areas in both states voted for Romney, but with only 34% of the primary vote (Santorum received 32% and Gingrich 29%). This coalition of rural areas and suburbs is where Santorum assembled his victories, although it’s worth noticing that with local establishment support, Romney was able to garner pluralities in most counties in the Mississippi Delta.

However, it would be a mistake to count Romney out, although it’s clear he is not, from a popular vote perspective, the dominant candidate. Illinois has its primary next Tuesday March 20, and the Republican electorate there (especially in Chicagoland) is more moderate, and this ideological dynamic heavily favors Mitt Romney. Furthermore, the April contests are tilted towards Romney, since 6 of the 7 contests held will be in the Northeast/New England (the remaining contest will be in Wisconsin). For the equilibrium/dynamics of the contest to change, Santorum needs to win, at a minimum, Pennsylvania and Wisconsin, as well as at least one of the remaining primary states. With his electoral base in the rural South diluted, Newt Gingrich has largely become a non factor in the Presidential contest now.

Turnout and President Obama – We had noted in our analysis last week that there was a strong correlation between the perceived level of competition in each state and the strength of Republican turnout relative to 2008. This pattern continued in last night’s contests: because the three major candidates (Romney, Santorum, and Gingrich) were perceived to be competitive, and in both states, turnout was up over 2008. In Alabama, turnout was 10% higher than 2008, while in Mississippi, turnout was double what it was in 2008.

We also noticed that a pattern has begun to emerge in states where President Obama is on the ballot: when he runs against “uncommitted”, he typically wins near unanimous percentages.  This again is what happened in Alabama, when “uncommitted” was President Obama’s only opponent – he received 81% of the primary vote (he received less than 60% of the vote in Oklahoma against minor primary challengers).

In the next installment of this article, we will discuss how early voting is proceeding in Louisiana, since we are essentially are at the half way mark.

Congressional/Statewide contests

Generic Congressional vote: 45-42 Republican/Democrat (14 day rolling average)

Alabama and Mississippi also became the second and third states to hold Congressional primaries last night. While there were no surprises, we noticed a continuation of two occurrences that we first saw in Ohio: (1) defeated incumbents were unable to make a comeback, (2) current incumbents (especially on the Republican side) are facing competitive primary races. In Alabama, party switcher Parker Griffith was humiliated with a 33% showing in the 2010 Republican primary. He sought a rematch, but lost additional ground: he could only garner 29% against Republican incumbent Mo Brooks.

Also newsworthy was that three Republican incumbents faced substantial primary challenges, and each of the three won his primary with less than 60% of the vote. While these incumbents represent Republican districts and are probably safe this fall, it does indicate that there is lingering dissatisfaction out there with primary voters that Republican incumbents would be wise to take note of.

Next week, Illinois is holding its Congressional primaries at the same time as the Presidential race. After that, the Congressional primary schedule is quiet in April: only two states (Maryland and Pennsylvania) hold primaries.

Meanwhile, Congressional qualifying has concluded in four more states (California, Idaho, Montana, and Texas). There were no surprises in Idaho and Montana, and we do not yet have the official list in Texas and California. Filing also concludes this week in Maine, Utah, Iowa, and Nevada.

At the present time, the “retirement count” is now 42 House members (25 Democrats and 17 Republicans) and 10 Senators (7 Democrats and 3 Republicans).  The Congressional playing field will become clearer by the end of the month: by then, a total of 26 states will see their candidate qualifying finish.

Early voting for this primary started yesterday and will continue until next Saturday (mail in ballots will still be accepted up until the day before Election Day). From examining early voting statistics provided to us by the Secretary of State, there are two things we noticed:

Low turnout numerically, but better than 2008

Yesterday’s turnout was 10,578, of whom 4,389 were Republicans. What does this early voting volume mean? Given the fact that the first day of early voting is typically the busiest day, we project at the present time that by Election Day, there will be about 25K early votes. When you compare this number against the 9,200 who early voted in the 2008 Republican Presidential primary, it appears that  voter turnout would be nearly three times what it was in 2008.

However, it is unlikely that Republican turnout will be triple the 2008 count, because in person early voting is a relatively recent concept in Louisiana (before the law was changed, you could only vote absentee before Election Day under certain circumstances). In the 2008 primary, 6% of the Republican primary vote was cast early. Since early voting has become more popular over time (16% of those voting in the 2011 statewide elections voted early), we think at the present time that voter turnout will be between 16 and 32%, or a turnout of between 125-251K Republicans (for comparison’s sake, 161K voted for the Republican candidate in the 2008 primary)

Santorum boost?

As part of our analysis of the early vote, we looked at where (i.e., which parishes) the early votes were coming from. In 2008, Mike Huckabee narrowly defeated John McCain 43-42%. His victory was achieved by running up large margins in the more rural areas outside of the Baton Rouge, New Orleans, and Lafayette media markets. In fact, the parishes he carried only represented 45% of the total vote cast and 47% of the early vote cast. Yesterday’s early voting numbers showed that 57% of the early vote cast came from the “Huckabee parishes.” Since the evangelical voters who supported Huckabee in 2008 are more likely to vote for Santorum than Gingrich (or even Romney), we see the distribution of early votes as a signal that Santorum may run strongly here on March 24. Below is a graphical depiction of the 2008 vote by parish in the Republican primary.

Red = Huckabee, Blue = McCain

Looking ahead

We will be recalibrating our turnout predictions as the week goes on, because we know from historical data that early voting for the next six days will be lower than yesterday’s figures, but the extent to which there will be a “plunge” is not yet known.

Why do we make a big deal about early voting? When the Legislature essentially established “no fault” early voting several years ago, you now have a noticeable constituency of people who prefer the convenience of early voting, and this constituency has thus far ranged from 6 to 16% – a politician would be foolish to ignore this many voters, especially in a closely contested race. Also, too, early voting numbers are the first ones that are typically reported after polls have closed at 8 PM.

(1144 delegates to win) Romney 415 (56%), Santorum 176 (24%), Gingrich 105, Ron Paul 47, Huntsman 2 (WSJ)

President Obama Job Approval 47-49% (14 day rolling average)

Presidential Contest 3/7 (Romney = blue, Santorum = pink, Gingrich = red)

Last night, 10 states held their Presidential contests, and the results seem to indicate two things: (1) the Republican primary contest is very much still in play, and (2) the demographic battle lines are solidifying in terms of bases of support for each candidate. Frontrunner Mitt Romney’s coalition includes moderate/liberal Republicans on either coast, affluent urban neighborhoods, Western voters, and Mormons. Rick Santorum now has a national base of religious voters, middle income suburbs, and small towns, with one important exception: portions of the Deep South with a large percentage of native Southerners – that voter bloc is still solidly behind Newt Gingrich.

This electoral mosaic has now established Rick Santorum as the candidate who can compete against Romney at a national level, although the fact that both Newt Gingrich and Ron Paul are remaining in the race means that Romney can/will keep winning primaries/caucuses with less than 50% of the vote.

With the sheer number of races last night, we will examine the results through the prism of what they mean for each of the three major contenders.

Newt Gingrich – His overwhelming win in Georgia (47-26% over Romney) kept him in the race, and was a wholehearted endorsement from a state he represented in Congress for two decades. The contours of his support in Georgia were quite similar to South Carolina – he carried all but the urban cores of Atlanta and Savannah. This means that he is in a strong position for next week’s primaries in Alabama and Mississippi, as well as the March 24 primary in Louisiana (early voting begins this Saturday). Beyond that, it is unclear what else he will win before the May primaries in North Carolina, Kentucky, and Arkansas – outside of Georgia, he was only competitive in Tennessee and Oklahoma, with a close third place finish in both states.

Rick Santorum – His wins in North Dakota, Tennessee, and Oklahoma have established him as a national contender and not just a regional candidate, and his strength is in rural and/or less affluent areas. He has a major hurdle to overcome; however, he has thus far been unable to defeat Romney in any of the larger states. And these states will have a substantial presence in the race starting with Illinois’ March 20 primary. Plus, the fact that conservative opposition to Romney is divided also puts Santorum at a competitive disadvantage.

Mitt Romney – The fact that he carried six states (Virginia, Massachusetts, Vermont, Idaho, Alaska, and crucial Ohio) would in other election years make him the prohibitive front runner. Except that in the states he won, his performance reiterates his inability to “close the deal.” Even with Ron Paul as his sole opponent in Virginia (Gingrich and Santorum failed to qualify), he only received 60% of the vote there – outside of the Washington DC suburbs in northern Virginia, his percentage was only 58%. In Vermont, a state thought to be overwhelmingly his only gave him 40% of the vote.

Nevertheless, his Ohio win over Santorum (combined with last week’s hard fought victories in Michigan, Wyoming, and the state of Washington) shows that he knows what it takes to get to the finish line. And his continuing strength in urban areas does theoretically give him the ability to compete against President Obama in swing states like Colorado, Virginia, Florida, Ohio, and New Hampshire.

From a tactical perspective, he is in a strong position to become the nominee as long as Newt Gingrich remains in the race, since he can continue to win with pluralities. Furthermore, he is well positioned to sweep Illinois’ primary on March 20, as well as at least 5 of the 7 states holding contests in April.

Upcoming contests – We expect that this month’s upcoming contests will do little to settle the race. This weekend, Kansas is having its caucuses. Given his level of support in smaller towns/the Midwest, Santorum should do well here, although Newt Gingrich will attempt to compete. Next Tuesday, three states are having contests: Alabama, Mississippi, and Hawaii. We expect Romney to win in Hawaii and Gingrich to sweep Alabama and Mississippi. However, the fact that Santorum carried Tennessee means that he can compete in those states. From there, the Illinois primary is on March 20, and we expect a strong Romney vote from that state. That leaves us with the March 24 Louisiana primary, and while Newt Gingrich is in a strong position to win, Santorum will be competitive with Christian conservative voters.

Turnout and President Obama – These miscellaneous topics have only received passing mention, yet they do have a bearing on the 2012 election. We have begun to notice a pattern with Republican turnout compared to 2008: in general, the more competitive a state’s contest is, the stronger that turnout has been. For example, the Ohio contest was very competitive (as was Michigan last week), and in both cases, increases in turnout were in double digits compared to the 2008 Republican primary in those states (Michigan’s turnout increased 12%, while Ohio showed an 18% increase). On the flip side, the fact that the Virginia race was a two man race resulted in a 46% plunge in turnout compared to 2008, while in Massachusetts,  turnout was down 28%.

Finally, we have been assessing President Obama’s electoral position by following the percentage of the vote he has been getting in states holding Democratic presidential primaries. One interesting pattern seems to have developed: when he runs against “uncommitted”, he typically wins near unanimous percentages. This was the case in Massachusetts (89%), Michigan, (89%), and Tennessee (90%).  In states where he actually had opposition on the primary ballot, his percentages have been lower. For example, he received 82% of the New Hampshire primary vote, while in Oklahoma, he received an embarrassingly low 57% of the Democratic primary vote. If these kinds of percentages were to continue, it would be a warning sign for the President in places like the Midwest, where the rural Democratic vote is a critical part of building a winning coalition in states like Minnesota, Michigan, and Wisconsin.

Congressional/Statewide contests

Generic Congressional vote: 44-42 Republican/Democrat (14 day rolling average)

Last night was important in another way as well: Ohio was the first state to have its primary for statewide/Congressional races. Curiously, primary voters were not particularly friendly to incumbents in either party. Republican Jean Schmidt has had competitive races both in the primary and general election ever since her narrow win in 2005, and was upset 49-43% by Iraq war veteran Brad Wenstrup. Democrat Dennis Kucinich also was defeated 56-40% in the primary, although in his case, it was because redistricting paired him with another Democratic incumbent. Democratic primary voters were also unkind to defeated incumbents seeking a comeback: Democrat Mary Jo Kilroy was elected in 2008 on President Obama’s coattails in a Columbus based district, but her support of the Democratic agenda cost her the seat in 2010. Even though redistricting was kind to her – the Republican Legislature created a safely Democratic seat in inner city Columbus – she still came up short in the Democratic primary.

Unlike the Presidential contest, Congressional/statewide contests are moving at a much slower pace: only three states are having their primaries this month. Alabama and Mississippi are holding primaries in conjunction with the Presidential primary next Tuesday, while the Illinois primary is also being held in conjunction with the state’s March 20 Presidential primary. There will be competitive contests worth watching in each of those three states.

Meanwhile, Congressional qualifying has concluded in four more states (Arkansas, Nebraska, North Carolina, and Oregon), which means that the Congressional playing field is set in 13 states. While there were no surprise retirements or unopposed candidates, the Senate race in Nebraska has gotten more competitive, as former Democratic senator Bob Kerrey changed his mind and decided to seek his old seat after all in an open seat contest.

If you count the incumbents who lost in Ohio, the “retirement count” is now 42 House members (25 Democrats and 17 Republicans) and 10 Senators (7 Democrats and 3 Republicans).  The Congressional playing field will become clearer by the end of the month: 13 more states will conclude candidate qualifying by March 31. In fact, by the end of this week, filing will conclude in California, Idaho, and Texas (the courts have implemented interim Congressional maps and have decreed a May 29 primary)

(1144 delegates to win) Romney 163, Santorum 83, Gingrich 32, Ron Paul 19, Huntsman 2 (WSJ)

President Obama Job Approval 47-49% (14 day rolling average)

Presidential Contest 2/29 (Romney = blue, Santorum = pink, Gingrich = red)

The ever changing Republican Presidential contest has now swung back to Mitt Romney, although not as decisively as he may have hoped. Despite the apparent inconsistency of the Republican contest, however, battle lines within the Republican Party have begun to form in terms of patterns of support each candidate has, which we think will be apparent in each remaining contest. Romney has become the candidate of more affluent white collar suburbs and moderate/liberal Republicans, while Santorum is getting his support from rural and/or blue collar areas, and Newt Gingrich’s appeal is the most potent in areas with a significant number of native Southerners. Let’s recap last night’s two contests through this prism:

Michigan: On the surface, it should be impressive that Romney carried the state 41-38% over Santorum (Ron Paul received 12% and Newt 7%), until you realize this is a state where he grew up and his father was governor back in the 1960s. Not to mention the fact that he defeated John McCain by a wider 39-30% margin in 2008 with more candidates in the race.

The support each candidate received paralleled the developing fault lines within the party: in a five county area encompassing Detroit in southeast Michigan, you have a heavily black and union/ethnic Democratic area that also contains a significant number of moderate suburban voters whose mild Democratic preferences have enabled Democratic Presidential candidates to carry the state in each election between 1992 and 2008. 41% of the statewide vote was cast in this area last night, and Romney racked up an impressive 45-32% lead over Santorum. In the remaining 78 counties “Outstate”, Santorum had a 42-38% margin over Romney, with strongest support in more rural and/or conservative areas in western and northern Michigan. This geographic split in support is why the delegates awarded in Michigan split down the middle between Romney and Santorum.

It’s also worth noting that Republican turnout was 976,000 – a figure 12% higher than in 2008. This (turnout) is becoming the untold story of the 2012 primary season, because it’s become apparent so far that each time there is perceived to be a competitive race in that state, Republican turnout increases. This has now happened in Iowa, New Hampshire, South Carolina, Maine, and Michigan.

Finally, it’s worth noting that there was a Democratic primary as well. Though President Obama was unopposed, he received 89% of the Democratic primary vote. While this is impressive, the vote for Obama against “uncommitted” was uneven across the state. He scored over 91% in Detroit, while in rural blue collar areas like the Upper Peninsula; he received 84% of the vote. This is a sign that he still has weaknesses with traditional Democrats – this is an area that gave him only 52% of the vote in 2008, and the Upper Peninsula voted in a Republican for Congress in the 2010 Republican landslide.

Arizona: This state was never in doubt, and Romney coasted to a 47-27% win over Rick Santorum (Newt Gingrich received 16% and Ron Paul 8%), carrying every county. While the fact that there is a Mormon presence here certainly didn’t hurt, Romney did the best with the more affluent white collar voters in Phoenix (where a majority of votes are cast) – he had a wide 50-25% lead over Santorum. Because this was a “winner take all” state, Romney won all of the state’s delegates.

We had noted in the Michigan write up that turnout (compared to 2008) seems to be dependent on the level of perceived competition, and this is what happened in Arizona. Everyone expected Romney to win there, and accordingly, turnout was down 16% from 2008.

What lies ahead

The state of Washington is holding a nonbinding caucus this Saturday. While the mildly liberal orientation of this state would seem to suggest a Romney win, caucuses (more so than primaries) tend to reflect the views of its party activists, which helps Santorum, who in a February 21 poll conducted by Public Policy Polling had a 38-27% lead over Romney. If this lead held up, it would be shown that Santorum has become the conservative alternative to Romney, and can go toe to toe with him in next week’s contests.

Going into Super Tuesday (now a week away), Santorum seems to have the upper hand in Ohio, Oklahoma, and Tennessee. Romney should dominate in Massachusetts, Vermont, and Virginia (only he and Ron Paul qualified on the primary ballot). Newt has a respectable 36-27% lead in Georgia over Santorum, and is expected to carry the state, given that he was a Georgia Congressman for 20 years.

That leaves Alaska, Idaho, and North Dakota. Given Santorum’s strength in more rural Midwestern enclaves, he should win North Dakota, while the heavily Mormon population in Idaho (as well as the migration of white collar Californians to Boise) should enable Romney to win there. Alaska is a tossup, given that you have a white collar urban area (Anchorage) surrounded by rural territory.

We’re watching Ohio for clues as to who will dominate next Tuesday. If Santorum wins Ohio, Tennessee, and Oklahoma, and is competitive in Georgia, that establishes him as king of “flyover country”, while Romney becomes the candidate of the East Coast, West Coast, the Mountain West, and (we believe) Illinois.

Congressional/Statewide contests

Generic Congressional vote: 43-42 Republican/Democrat (14 day rolling average)

Statewide/Congressional primaries are about to start soon, as Ohio holds its primary next week. In March, the Congressional stage will begin to take shape, as 15 additional states will have their filing deadlines conclude during that time. In the immediate future, North Carolina closes its filing today, while tomorrow is Arkansas and Nebraska. Thus far, nine states have concluded Congressional qualifying.

There were two Congressional retirements since last week: moderate senator Olympia Snowe (R-Maine), (who was elected in the 1994 GOP landslide; and California Republican Congressman David Drier, who was elected in 1980. That now brings the retirement count to 38 House members (22 Democrats and 16 Republicans) and 10 Senators (7 Democrats and 3 Republicans).