Lying (to ourselves) with statistics: “PVI” and the “DFL Index”…

If you’ve been hacking about politics long, no doubt you’ve heard about the “Index” of a district. On the federal level, the “Cook Political Report” has been cranking these out for years, essentially averaging out the presidential vote by party for the past couple elections, as explained here. Minnesota’s Democratic Farmer-Labor (DFL) party does something similar, but tries to keep it’s recipe secret just because they’ve added a few herbs and spices like the vote for statewide offices in the district. For those of us that don’t have access to the party’s innermost secrets like the “DFL Index”, Tony Petrangelo has produced a pretty damn good substitute here.

So the Cook PVI tells me that my congressional district, Minnesota’s 7th, is a hopeless R12, predicting that a republican should defeat a democrat by a 62% to 38% margin.  My legislative district, 22A, rates an R+8 by Tony’s calculations, but he hasn’t updated after the disastrous 2016 elections. Sounds like I should throw away my campaign stuff and find a new hobby!

But apparently several DFL candidates didn’t get the memo on how hopeless MN CD7 was and ran anyways… Like Congressman Colin Peterson who has held the seat for two decades now. Or Senator Amy Klobuchar, who won with 61% of the vote in 2012, or Senator Al Franken and Attorney General Lori Swanson who squeaked out victories in 2014 in CD7 despite the democratic debacle around them. What the…?

The problem is that we’re melting down a rich data set of election results into an average “Index” that tells us almost nothing. In so doing we let one candidate who dragged down the whole ticket label a district hopeless for democrats, despite strong evidence to the contrary. And following the DNC and DFL’s mythology that the voters fall into vast camps on the left and right and victory goes to the party who best gets their camp to the polls, we ignore the nearly as large camp of “swing voters” who usually decide elections in MN CD7 and other rural districts.

So let’s look at the real numbers:

MN CD7 has about 509 thousand folks 18 and older, per the Census 2015 estimates. There’s only around 20 thousand foreign born adults in the district, and many of them are now citizens. There are around 40 thousand felons barred from voting in Minnesota, but most of them reside in the metro districts… So we have around 500 thousand voters in MN CD7. In the three elections since redistricting, the base GOP vote was sounded by Torry Westrum at around 110 thousand, barely lowing the bar below Kurt Bills 112 thousand tally of the GOP loyalists. On the DFL side Hillary Clinton lowered the bar defining the DFL base to 105 thousand. So the GOP and DFL bases are each barely 20% of the voters… Damn, gonna have to redefine the whole field operation!

DFL, meet the swing voter… In 2012 69 thousand voters cast a vote for Romney, then crossed over to give Klobuchar a total of over 200 thousand votes. The GOP district wide slate was weak in 2014, but 5 thousand voters gave the GOP U.S. Senate candidate their vote, then crossed over and voted for DFLer Peterson.  In 2016 208 thousand voters gambled on Trump, then 51 thousand crossed over to hedge their votes on DFLer Peterson. So truth be told, up to 71 thousand voters, a slice of the electorate almost as big as the DFL and GOP bases, are deciding the elections in CD7.

About now some grizzled DFL field organizer interjects “But turnout still wins elections”. But turning out a DFL voter only wins us one vote, swing a swing voter to the DFL and they count twice, because you’ve denied the GOP a vote. But the “base strategy” is half right… Of those 500 thousand or so eligible to vote, only 334 thousand bothered in 2012, 242 thousand in 2014, and 338 thousand in 2016. Yup, there are over 160 thousand votes up for grabs in MN CD7, even more in a non presidential election year, and other that we know who didn’t vote, we have no idea who or what would get them to vote!

DFL and DNC, delete that failed “Index” software off your computers. Just to be safe, crush the hard drive, then throw it in an acid bath then irradiate it just to be sure! We need to quit trying to distill the rich landscape of voting data points into simplistic indexes and instead use these mountains of data to pick which candidates and strategies will win over swing voters and non voters to our party. Our DFL candidates are more than the party’s “index”, and no longer should Hillary’s dragging the “index” down be used as an unscientific excuse to deny them the state and national party’s support!



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s