17 November 2016

Political campaigns matter #nlpoli

bialik-turnout-nov15-1New information calls for a change in perspective.

Turns out that the drop in turnout identified by the initial vote results wasn't as big a drop from 2012 as initially reported.  The folks at fivethirtyeight.com have figured it out.

 About 58.1 % of eligible voters turned out,  down  from 58.6% in 2012. The turn-out in 2000 was about 54%. Contrary to the impression some folks have,  turnout in American elections is actually up lately. The recent election may wind up having a bigger turnout than any election between 1972 and 2000.
bialik-turnout-nov15-2So what does that do to the earlier SRBP post?

Well, the main point in the post was that the two parties ran strategies that aimed to suppress the votes of the other crowd.  This was particularly true of the Republicans.

A revised map of the state-by-state change in turnout from fivethirtyeight.com  - at right - shows how relatively small changes actually took place in most states.

Bear in mind, though,  that these relatively small percentages likely translate out to a significant number of votes in some instances and, in other instances, to enough votes to shift the state from the blue column to the red column.

Case on point:  Michigan.  Shows here as having a change in turnout of less than two percent, but the real story is in the way that translated out.  On election night, it was clear that involved a large number of  Democratic voters in 2012 staying home.

The more we learn about the 2016 American presidential election, the more we will see the impact that the campaigns had on the outcome.  Seems like an obvious point but frankly it is the one element missing in most post-vote analyses of elections.