Limbaugh Democrats for Hillary

Nimh,

http://www.mydd.com/story/2008/3/12/12154/1623

Some number-crunchers have figured out that the Republican cross-over voting cost Obama about 10 delegates in MS alone.

sozobe wrote:

Mark Blumenthal, from Pollster.com:

Mark Blumenthal wrote:

Second, the open Mississippi Democratic primary -- the first to be held after John McCain secured his nomination -- included 2-3 times as many Republicans (12%) as the other states. And those Republicans supported Clinton by a 3-to-1 margin

Hmm.. so if we say, for simplicity's sake, that every last Republican Hillary voter was a "Limbaugh voter", that would be 3/4 of 12% = 9%.

But that's not quite right, since even in previous primaries there were some Republican cross-over votes for Hillary. In Alabama, for example, on Super Tuesday (before the Limbaugh appeals), where Republican cross-over voters made up 5% of the total vote, they split evenly between Obama and Hillary. So it will be a little less.

Let's take Blumenthal's lead and peg the percentage of Republican crossover voters in previous primaries at 4-6%. Let's consider that the 'regular' or legit cross-over vote (after all, on Super Tuesday there was still a fierce Republican contest on, not many Republicans will have wasted their vote on Hillary back then). So now in Mississippi, too, you would have had some 5% "legit" Republican crossovers who split evenly between the two Democrats, like they did in Alabama. That makes 2-3% "legit" Republican crossover votes for Hillary and 2-3% "legit" Republican crossover votes for Obama.

That covers almost the full Republican Obama vote, but leaves you 6-7% extra Republicans for Hillary, which is exactly the number by which the Republican cross-over vote increased. That 6-7% would then be, approximately, the Limbaugh vote.

Which is very close to what I calculated above, when I wrote, "let's assume that ... "Limbaugh Republicans" made up some 5% of total Dem primary voters yesterday." A sizable vote, but not enough to meaningfully change the breakdown by race of the vote.

E.g.: Let's do the calculation again but account for 6.5% rather than 5% of Limbaugh voters. Assuming they were are all white, that makes for 13.5% of the white vote. If you discount those, you are left with Obama getting 26 out of every remaining 86.5 "legit" white primary voters = 30%. His share of the white vote goes up from 26% to 30%.

Cycloptichorn wrote:

Nimh,

http://www.mydd.com/story/2008/3/12/12154/1623

Some number-crunchers have figured out that the Republican cross-over voting cost Obama about 10 delegates in MS alone.

The details sure look interesting - the point about "thresholds" that were near and narrowly missed is well taken.

But the conclusions that the diarist goes to from there - well, I dont even know where to begin.

Eg: dismissing race as basis for the animus of white / Clinton voters towards Obama. "It wasn't race! It was Rush!", he exclaims. But come on. Even discounting all Limbaugh voters, Hillary still got 70% of the white vote (see above). Nuff said.

Moreover, the phenomenon of the "bitterness quotient" among Hillary towards Obama voters being bigger than vice versa predates Limbaugh's actions. Look at the [updated] chart I posted [..]. Dissatisfaction with the prospect of Obama as nominee among Hillary voters was up in the 60s as well in Alabama, ARkansas, Oklahoma and Tennessee - all back on Super Tuesday, before the Limbaugh appeals and the surge in Republican participation in the Dem primaries.

The "Limbaugh Republicans" phenomenon in the Dem primaries has been real, for sure, but the math just doesnt add up for it to significantly impact either the race issue or the Hillary-ites' bitterness issue that the media is highlighting. The author says the media is wrong in playing up those issues because "It wasn't race! It was Rush!"; well, that's just not true, the numbers for that assertion are not there. And I think it's dangerous for Obamaites to dismiss these real concerns and try to reduce it all to external spoiling, because that leaves them ill-prepared to deal with them if and when Obama does become the nominee.

nimh wrote:

The "Limbaugh Republicans" phenomenon in the Dem primaries has been real

[..]

And I think it's dangerous for Obamaites to dismiss these real concerns and try to reduce it all to external spoiling, because that leaves them ill-prepared to deal with them if and when Obama does become the nominee.

I agree with this. "Real but minor" is about my take at this point.

Four days ago, I posted about what I called "the bitterness quotient":

nimh wrote:

I was curious whether, now that the campaign is lasting so long, the number of respondents saying they'd be dissatisfied with the other candidate as nominee was going up - and if so, who would be the more dissatisfied ones, Hillary or Obama supporters.

Finding out the latter isn't entirely straightforward, it involves a bit of recalculating. But the results are surprising/interesting:

• The "bitterness quotient," to give it a flippant label, did increase over time, but it was mainly right between the SC/FL primaries and Super Tuesday, with surprisingly little deterioration since;
  
• Hillary voters would be more dissatisfied with Obama as nominee than vice versa (something I hadnt expected);
  
• Differences from region to region are much larger than the difference over time. In five Southern states, Hillary voters would be significantly more dissatisfied with Obama than voters of either candidate would be with the other anywhere else. A race question? Alternative explanations welcome.

Here's the table with results:

OK, now Mississippi has voted, I wanted to post an update of the table. Here it is. Note: the "bitterness quotient" was 58% among Obama voters, the highest so far. And it was a record-setting 72% among Hillary voters.





Notice the remark I added to contextualise the percentage of Hillary voters who said they'd be dissatisfied with Obama as nominee in Mississippi.

In the Jed Report post Cyclo linked in on the previous page, the author wrote:

Quote:

One of the key takeaways from the media yesterday was that Hillary Clinton's supporters didn't like Barack Obama as much as Barack Obama's supporters like Hillary Clinton. They implied racial animus was at the root of this division [..].

They thing they left out was that a full quarter of Clinton's supporters were Republicans! And the Republicans didn't like her either! Because they were gaming the process!

It wasn't race! It was Rush!

I wanted to take this possibility into account. After all, Republicans did apparently make up an extraordinary 24% of Hillary voters. And the fact that 13% of Hillary voters said that they'd be dissatisfied if she were the actual nominee does suggest such voters played a role.

However, a quick recalculation shows that when you actually discount such "Limbaugh voters" and then revise the numbers accordingly, the impact they had turns out not to significantly change the overall picture. Even if you discount every last Republican cross-over voter for Hillary (some of whom will have been genuine), the "bitterness quotient" among the remaining Democratic and Independent Hillary voters remains a sizable 63%. That's similar to other Southern states and higher than in any non-Southern state.

Here's the calculation for that:

how bout an executive summary.