(This seems more like RBH and Michael’s thing than mine, but worthy of discussion. – promoted by Clark)

In thinking about the way politics are reported in this state, I realized that there was no shorthand way to refer to the relative partisan leanings of one county versus another.  Sure, anybody that follows Missouri politics knows that Jackson county is Democratic and Greene county is Republican, but what about the other 113 counties in the State?  And can’t we get more precise than that?

This issue led me to spend some time tabulating what I am calling a “MO PVI” for each county.  The MO PVI is based off Charlie Cook’s “Cook PVI” to describe the relative partisan leanings of each Congressional House District in the country.  Explanation and analysis is below the jump.

The great thing about the Cook PVI is that it gives a single number and lets you make a quick and dirty comparison of the relative partisan leanings of each house district.  I assume everyone is familiar with what Cook PVI is and how it is calculated.  Click here for a primer on the concept.

That’s all well and good for folks following federal elections, but what about a similar concept for state-level elections?  Currently, the closest thing you hear mentioned to a PVI is a certain state legislative district’s Democratic Performance.  This number, expressed as a percentage of the vote (i.e. 48%, 51%, 38%, etc.) is roughly a calculation of the percent of the generic Democratic vote of that district.  This is useful information, but there are limitations to this number:

1. As far as I know, this info isn’t really public.

2. Often, statewide campaigns run by non-credible candidates are included in the calculation (think Mitch Hubbard v. Robin Carnahan), which can result in a skew against whatever party put up the non-credible candidate.

3. The numbers themselves don’t give much information regarding the relative partisan leanings of the house district.

In my humble attempt to remedy this lack of information, I present to all of you the MO PVI for each county in Missouri.  As you can see from this spreadsheet, I have taken the difference in vote percentage between the Republicans and Democrats for the 2000 and 2004 Governor elections, compared that number for each of the elections to the statewide results, and then averaged the two to come up with the PVI.  Some observations:

1) The 2000 governor’s election seems to have been from a different era.  Holden won many outstate counties, while Talent almost won in St. Louis County.  I can’t even comprehend how that could happen today.

2) Partisanship drastically increased from 2000 to 2004.  The Republican counties, on average, got much more Republican, and vice versa for Democratic counties.

3) This is pretty obvious, but Missouri is in a rather unique situation where the last three gubernatorial elections will all be open contests with no incumbents running in the general election.  This probably makes the numbers a little more “true,” as it gets rid of any incumbent effect.

4) Here are the top ten Democratic counties in the State, according to MO PVI:

    St. Louis City D+55.4

    Kansas City*   D+47.1

    Pemiscot       D+20.1

    New Madrid     D+11.0

    Ray            D+10.2

    Saline         D+9.9

    Mississippi    D+9.3

    Boone          D+7.5

    StL County     D+7.0

    Clay County    D+5.9

*Kansas City reports its election results separately from Jackson County, which is why it is treated separately here.

Nothing too shocking here, although I was surprised to see the Bootheel trio so high.  I wouldn’t have figured Saline so high, either, which is probably explained by an inordinate amount of love they had for Bob Holden.

5) Top ten Republican counties:

    Barton     R+49.7

    Newton     R+41.7

    Jasper     R+40.4

    McDonald   R+39.8

    Putnam     R+39.4

    Dade       R+34.6

    Wright     R+34.5

    Mercer     R+34.4

    Douglas    R+34.3

    Taney      R+29.9

Welcome to southwest Missouri, with a little North Central thrown in to keep things interesting.

6) I also included a “trend” column to see where we are losing strength.  Here are the counties that lost the highest percentage of Democratic vote from 2000 to 2004:

    Clark    -37.3% Democratic loss

    Lewis    -30.4

    Marion   -30.3

    Scotland -30.0

    Monroe   -27.4

    Shelby   -27.0

    Ralls    -25.2

What the hell happened to all the Democrats in Northeast MO in just four years?!  Have we found them yet?

Here’s where the Dems picked up the most, percentage wise, from 2000 to 2004:

    KC       +16.9

    Grundy   +16.2

    Perry    +15.2

    StL Co.  +13.2

    St. Chas +9.6

    StL City +9.4

    Platte   +8.3

Looks like increased strength in the cities and suburbia, with two rural outliers thrown in.

Does this add to the political discourse?  I don’t know, but I thought it was interesting enough to spend some time on it today.  I’ll leave it up to you all to determine if “this is anything,” as Letterman might say.  I’d be especially interested in hearing about any errors I made.  Most of these numbers “feel” right to me, but outside of KC and NW Mo my knowledge is limited.  If, however, this is something that has been done already, I do not want anybody to mention it, because I have spent too much damn time on this today for it all to be for naught.

If somebody out there is good with maps, it would be cool to see this data in that format.