Washington, D.C. (rightcommentary.com): Strategic cross-over voting is nothing new in partisan elections. It has long been discussed that in partisan caucuses and primaries, a certain percentage, usually thought to be small percentage, of partisans “change sides” in order to disrupt and or shape the outcome of the contest in the hopes that their actions benefit their “true” partisan interest. This issue was taken to a new height, however, with Rush Limbaugh’s conception of “Operation Chaos” where his listeners (presumed to be mostly conservative and/or Republicans) will abandon their Republican tag in order to vote for Hillary Clinton. His goal was to create a situation where Sen. Clinton was being supported by enough cross-over votes to keep Sen. Obama at bay and thus cause the Democratic party nomination battle to continue all the way to the Convention in Denver in August. Limbaugh told his listeners that by continuing to “bloody” the candidates as they fight for the nomination - Republicans would gain a strategic advantage in winning the General Election in November.

Quite honestly, while I had no doubt that Limbaugh’s listeners (of which I am one) would execute “Operation Chaos,” I did doubt that it could materially change the outcome of one of the electoral contests. I disagreed with Mr. Limbaugh that Operation Chaos was a determinative factor in the Pennsylvania Primary - I believe Sen. Obama’s “guns and bibles” remark was largely responsible for her victories, combined with the fact that the demographics in Pennsylvania were stacked against an Obama victory.

However, I have spent a good portion of the day looking at the outcome of the Indiana primary and I do believe that Operation Chaos did play a determinative factor in that contest and I believe there is significant evidence pointing to the fact that Clinton would have probably lost Indiana by as much as 3-5% had Limbaugh supporters not voted for her.

It is difficult to know by what margin and by how many people were inspired by Limbaugh to change party lines and vote. We do not know how many of those voters might not have voted at all, were it not for their desire to be a part of “Operation Chaos.” However, there is some stunning evidence to suggest that as few as 18,000 and perhaps as many as 70,000 voters who would normally identify themselves as “Republican” and might have, in a close Republican contest, voted for the Republican candidates instead voted for Hillary Clinton.

Consider some of the following facts about Partisanship and Elections generally:

• “Republicans” or people who identify themselves as such - tend to vote as a cohesive block and they do not tend to change their “party ID” from election to election.
• There are considerably more registered “Democrats” than “Republicans” in almost every state. However, Democrats do not tend to necessarily vote the democratic ticket nor do they vote in the same numbers that Republican voters do.
• Republican voters tend to show greater resistance to changing their affiliations, whereas Democrats or “moderates” tend to float in and out of party affiliation. Since no general election contest requires a partisan ID, and moderates and unaffiliated tend not to participate to a high degree in partisan contests - Republicans can win general elections by maintaining a high percentage of their base and getting a large portion of the Republican-leaning unaffiliated and some right-leaning Democrats to vote Republican.

What those facts tell me:

• Over time, changes in registration in the Republican party is largely a factor of demographic changes in the state writ large, such as population growth or macro demographic changes in education, income, employment, etc.
• Once registered Republican - you’re unlikely to change that registration.
• Any sudden change in registrations or voting patterns are therefore the result of some exogenous shock or unexplained intervening event.

Let’s take a look at some interesting facts about Indiana registration and voting patterns:

• Indiana is very much a “Red State” that has largely maintained a large Republican base. Over half of both the federal and state elected seats are held by Republicans.
• In 2004 - Bush won Indiana by 60%, or roughly the same margins as held in Federal offices (House and Senate).
• In 2008, approximately 1.93 million voters were registered as Republicans, compared to 2.63 million Democrats (however, as I explained, you can have more “registered” Democrats and still have a red state if they don’t vote.)
• From 2004 to yesterday, approximately 23,000 new voters registered. While I realize that at the Indiana Secretary Of State website - they say something like 300K… the reality is they must have had people move out of state and move in - register - etc. In 2004, approximately 4.296 million registered voters were counted, and in 2008, about 4.319 million registered voters were counted. This is only a difference of 23 thousand.
• Given the number of “new registered” voters according to the Indiana Secretary of State - we can estimate about 182 thousand new Republicans registered in Indiana.
• In the primary yesterday, about 1.265 million democrats voted. About 410 thousand Republicans voted. In 2004, about 469 thousand Republicans and 317 thousand Democrats voted in the Primary. In 2000, 322 thousand Republicans and 455 thousand Democrats voted in the primaries.
• In the primary yesterday, 260 thousand more people who identified themselves as “Democrats” voted than those who identified themselves as Democrats voted for Kerry in 2004. Consider that for a moment, 260 thousand more than in the GENERAL ELECTION. Quite honestly, this fact alone should be of considerable concern to Republicans as this fact has been repeated over and over in “Red States” around the country during the 2008 primary/caucus season.

What can we deduce from these facts:

• For the first time in 3 contests, the number of people voting Republican went down, and it went down by 10% almost, even when voting surged for Democrats by a factor of four. Realizing that since McCain is the nominee there may be reasons for Republicans not to go to the polls… even when considered as percentage of registered to voted in the primary for 2004 (a similar situation where Bush in the nominee) it represents a pretty dramatic drop.
• It would be difficult to explain this drop by saying Republicans “stayed home” (you’ll see why when I get to the next section on exit polling) - or at least difficult to explain it alone by saying Republicans stayed home.

Now finally, exit polling data from MSNBC is quite telling (but not entirely consistent) with what we observed:

• Only 8 percent of those who responded to the poll said this was their first time voting.
• Only a quarter (roughly) said this was their first time voting in a primary.
• 61% of the voting pool identified themselves as “moderate” or “conservative”
• 16% identified themselves as “Conservative” (read here - REPUBLICAN). Think about that - 16% of those surveyed who voted in the DEMOCRAT primary were considered by themselves to be CONSERVATIVE!
• More than half of those who identified themselves as “Republicans” went to Hillary

Okay, so from all of this, I’ve made the following conclusions:

• Hillary won by only 18 thousand votes. It is unlikely that her “get out the vote” drive alone was sufficient to get her over the hump since the majority of voters were not new voters and the majority of liberal, moderately liberal, voters went to OBAMA.
• About 50 thousand Republican votes seemed to evaporate… without much to explain it besides Operation Chaos.
• 16% of the vote of people who were not “Registered” republican but who considered themselves republican voted - over half of those votes went to Hillary.

I think it is likely that Hillary received anywhere between 30 and 50 thousand votes from either registered Republicans or individuals who would otherwise vote Republican in the general election. This amounts to her receiving between 3 and 5% more votes than she would have otherwise received had those individuals not voted for Clinton.

I think this may be the best instance of where strategic voting actually did have an impact. There are some reasons for this:

• Being a red-state, the Republican contest was unimportant as the nominee was chosen, and switching sides would not change allocations for resources for the parties.
• Indiana’s demogrpahics (population on race, ethnicity, etc.) did not favor Obama like it did in North Carolina. While I think in North Carolina republicans voted for Hillary there as well - it was not sufficient to change the outcome in a state where the overwhelming majority of voters identified with Obama.
• The ticket was split in Indiana between Clinton and Obama - so influence from strategic voting was maximized.

While I don’t believe that the “Limbaugh Democrats” (as some pundits are calling them) can influence many contests… I think in situations like we had in Indiana, they can definitely throw Hillary a lifeboat and get her over the hump in these closely contested races.

I think Limbaugh can legitimately claim his strategy and the people who carried it out did impact the election in Indiana - but I welcome other analysis. This is largely anecdotal and kinda “obi-wan kenobi” science - in using your gut in looking for unexplained patterns. As we get better data down the road - perhaps the picture will be clearer and my hypothesis falsified… but I don’t think so.