Wednesday, April 28, 2010

Political Tides

The state puts out some pretty detailed statistics on voter registration, including specific data on folks who change their party registration.   So I poked at that data a bit and came up with the time series below depicting the trends for the two major parties in terms of who is switching their voter registration over the last year+.

This deserves a lot more explanation than anyone wants to read but...   This graphic compiles the ratio of folks switching from Democrat to Republican and shows it as a ratio to the number switching from Republican to Democrat. Normalized to 100, so a value of 100 means the number switching from D to R and number switching from R to D are equal.  The higher the number, the more folks are switching from R to D compared to the other way around.  Numbers under 100 mean more folks are switching from Democrat to Republican.  Put simply:  high = good for Democrats  low = good for Republicans.

I summarized the weekly data into 4 week segments to get rid of some weekly volatility. I did the same for both the state as a whole, and then for Allegheny County by itself.  What isn't obvious is that the raw number of folks switching their registration is pretty volatile over the course of the year.  So some of the data points are calculated on just a handful of voters (such as in the period just after an election), while at other times there are a lot of voters (such as just before registration deadlines going into an election) being captured in this. Nonetheless you can see some of the reason D's might have cause for concern looking at fall elections since this is what I come up with.

So if you are a D, don't despair entirely.  No time to break it out the same way, but it looks to me that D's have had, and continue to have, a pretty decisive advantage in capturing the largest fraction of new voters as comapred to the party switchers captured above. Party switchers, by definition here, have to have been registered already and are not new registrants.  How it all nets out in the long run I dunno?? What else might be impacting these numbers?  I was curious if there might be some impact of the whole story from 2008 when Rush Limbaugh encouraged Pennsylvania Republicans to register as Democrats in order muck with the primary election... Remember that?  Maybe there is some residual impact of that in these numbers, but again who knows?


Anonymous MH said...

I'm voting for the guy who puts state legislators in jail.

Wednesday, April 28, 2010 8:41:00 AM  

Post a Comment

<< Home