Judging the maps: How we did it

A different holiday crush

For members of the Draw the Lines PA team, the recent holiday season included a lot of the usual stuff – eggnog, latkes, exchanging presents, going to parties, watching football – and one new element.

Staring at maps.

Lots and lots of maps.

The deadline for our project’s first statewide mapping competition was 11:59 p.m., Friday, Dec. 14.

By 8 a.m. that next day, our intrepid chief of staff, Justin Villere (AKA Resident Tech Wizard), was hard at work downloading maps and data from the DistrictBuilder mapping platform and organizing them into his beloved Google Sheets.

By that Sunday, our task was clear, if daunting.  We had 300-plus valid entries that met the basic criteria for being validated by District Builder. Each map divided Pennsylvania into 18 contiguous congressional districts of nearly equal size.

Now, we had to winnow that pile of maps down to six: a statewide champion and a runner-up in each of three age categories, youth, higher education and adult.

For the mappers who’d clearly put many hours into their maps and personal statements, a lot was riding on our judgments.  A regional and statewide champion would win $5,000. That's $500 for regionals, $4,500 for the state title.

We owed each entry careful, serious consideration.

 A four-stage process

 First Justin and I, as chief of staff and project director, made a first cut, eliminating maps we just could not imagine putting up on the screen in the Capitol Rotunda during our Feb. 6 awards ceremony and saying, “Hey, look at how great this is.”

Also out were entries where the mapper clearly put no effort into the personal statement.   Together, that eliminated maybe a third of the 300.

Second, amid the pre-Christmas rush, internal judging panels, made up of Draw the Lines team members and Committee of Seventy staffers made a second cut of the maps. They created groups of 7-14 “semifinalists” in each of our nine judging cohorts e.g. Youth/East, Higher-Ed/West.

That left us with about 90 semifinalists – pretty darn good maps.

In the third stage, judging panels made up of our regional coordinators and steering committee members made the often-hard final determinations of who should be the regional winner and runner-up in each cohort.

They also had the option of awarding honorable mentions for maps that it really pained them to leave out of the top two, or quirky maps that showed real effort and ingenuity.

The regional winners receive $500, the runners-up $250.   Honorable mention cash awards were up to the judges’ discretion; they varied from $75 to $200.

A four-part rubric

For the regional judging, we created this rubric:

1) Look of the map - At a glance, does the map seem to follow some logic?  Does it offer reasonable solutions to the challenges posed by the state's geography and unequal distribution of population?  20 percent

2) Map metrics - Based on the provided metrics, does the map actually achieve its stated goal(s)?  (Balance among multiple goals can be valued equally with nailing one particular goal.) 30 percent

3) Personal statement -  Does the statement clearly explain the thinking that went into the map? Does it say anything distinctive, perceptive, memorable?  Does it show any creativity or extra effort?  30 percent

4) Outreach and dialogue - Were the map and statement based on any democratic dialogue or outreach on the part of the mapper(s)?  20 percent

As you can see from the percentages, we put as much stock in the thinking and democratic dialogue that went into the map as we did in the actual lines. 

This was made clear in the contest rules, so mappers who actually read the rules had a bit of advantage. 

And, yeah, we could pretty much tell who’d read the rules and who hadn’t.  (Sometimes, TLDR is not a great plan.)

On the entry form, mappers told us 1-3 goals (out of possible eight) that they were going for on a given map.  Then, judges would weigh how well the map met its goals, both by eye test and looking at the map metrics automatically generated by District Builder

Numbers that tell a tale

The metrics are:

Competitive Districts – The actual formula is bit more complicated, but the easiest way to explain this one is that, usually, if the percentages of Democrat and Republican in a district are within 10 percent, the district is deemed competitive.   In districts with a high percentage of Independents or Other Parties, the spread between D and R has to be a little tighter.  What’s a good score here? Well, the way people have sorted themselves geographically in our state, it’s actually harder than you might think to create competitive districts. The typical score here was 4 to 6 competitive districts out of 18.  A score of 8 or above we considered really nailing this goal. A few maps (and, boy, do most of them look crazy) created more than 14 districts.

Compactness -  We used the Polsby-Popper ratio, one of several complicated metrics used by professional mappers.  It’s expressed as a percentage.  Basically, it measures how close a district’s shape is to a circle.  It punishes wandering or jagged borders. Any score above 30 percent is OK on this metric.  If you got into the 40’s on compactness, you were really cooking. Only a handful got above 50 percent.

Majority-minority Districts – This refers to any district where the sum of percentages for African-American, Latino and other non-white residents tops 50 percent. It’s super hard to create more than two in Pennsylvania.

Population Equivalence – This is the spread between figure for a map’s most populous district and its least populous.  The smaller the better.   In the real world, Supreme Court rulings on the principle of one-person/one-vote require that for congressional maps this figure must be in at most the double-digits.   At Draw the Lines, we concluded this rigorous goal would be too hard and frustrating for first-time mappers, so we allowed a tolerance of plus or minus .5 percent from the perfect district size of 705,688.  On this metric, anything under 2,000 was a really great score.  Anything in the 3,000’s was very solid, and scores between  4,000 and 6,000 were common.

No robes, but good judgments

Equipped with this data, our regional judging panels dived into their task with rigor and passion.  Some judges showed up for their deliberations with stacks of charts and spreadsheets that summarized their analysis.  The discussions were intense , sometimes running on long after the 90 minutes allotted for them, as judges agonized over those final calls.

We want to thank by name all the people who brought such care and integrity to the judging:

EAST – Barbara Adams, Kate Doyle, Amanda Holt, Steven Bradley, Myra Forrest, Yaasiyn Muhammad, Tom Baldino, Karen Clifford, Nyron Crawford, Linda Breitstein

CENTRAL – Rick Bryant, Sarah Niebler, Dave Trevaskis, Jeff Cooper, Bernie Mazur, Nicholas Pyeatt, Jill Family, Corinna Wilson

WEST - Anna Batista, Martina Jacobs, MaryKay Babyak, Martina Jacobs,  Laurie Sprankle, Jason Wilburn, Lindsay Cashman and Rachel Colker

No judges who are educators judged any division where students from their school or college were entrants.

The final cut in the judging – taking 20 regional winners (we had two ties for runner-up) down to the three statewide champs and three runners-up – took place Jan. 23.

The judges were our six regional co-chairs:

Mark Schweiker, former governor and lieutenant governor of Pennsylvania.

Maureen Lally-Green, dean of the Duquesne Law School and former state Superior Court judge.

Fred Thieman, Chair for Civic Leadership with the Buhl Foundationa and former U.S. attorney for the Western District of Pennsylvania.

Mike Brubaker, a business executive from Lancaster and former Republican state senator.

The Rev. Sandy Strauss, director of advocacy and ecumenical outreach for the Pennsylvania Council of Churches.

Sharmain Matlock-Turner, president and CEO of the Urban Affairs Coalition.

Their choices for statewide honorees will be revealed at our Feb. 6 awards event in the state Capitol Rotunda. It begins at 10 a.m. and you are invited to attend.