PA Senate Map that is Judicially Enforceable

View Map on Dave's Redistricting (

This Pennsylvania (PA) 50-seat state Senate map is submitted by Concerned Citizens for Democracy (CCFD), a PA nonprofit 501(c)(3) association. The map was created by applying CCFD’s straightforward nonpartisan redistricting method for redistricting that results in fair, judicially manageable and enforceable districts, a stipulation sought by the Supreme Court of the United States. A Senate map using the same CCFD method received an award from the Draw the Lines competition in 2019. Electoral district maps can be readily prepared using a 5-step method that is transparent and fair. The method applies a judicially enforceable standard that reduces the risk of legal disputes. The CCFD method abides by the four (4) principles stipulated in the Pennsylvania Commonwealth’s Constitution. Districts are compact, contiguous, equal in population, and minimize boundary splits of political entities – “Unless absolutely necessary no county, city, incorporated town, borough, township or ward shall be divided….” The following process summarizes the CCFD 5-step method: 1. Divide the state into roughly equal population districts using the largest political subdivision in the first instance. These are usually counties. 2. Add or subtract smaller political subdivisions in a compact manner beginning at the common border of the largest political subdivisions to begin equalizing population among districts. 3. Repeat the procedure in Step 2 with each level of smaller political subdivisions (e.g., municipalities, wards, precincts). 4. Create compact districts as electoral districts are formed. Then measure compactness and the number of split political subdivisions. 5. As a final step, make sure the district complies with the VRA. The CCFD method is simple. Simple inputs. Simple procedure. This simplicity yields transparency that all Pennsylvanians can witness and understand. This simplicity and transparency are what guarantees a method that the courts can examine and attest to producing an enduring standard for the creation of any and all redistricting maps and plans. Most importantly, what is NOT required to create districts are any information on the following: • NO demographic data including race, age, etc. • NO partisan data on party registration or election history • NO incumbent elected officials’ addresses • NO community of interest data, for example school districts • NO physical geography, such as rivers or ridge lines Through further independent mapping research, CCFD determined that for PA Senate redistricting, a +/-2% population variance from the target district population is a practical population range (260,054 target from 2020 Census with +/- 5,201 persons) to achieve the balance of representative fairness and manageable map drafting. CCFD’s unique and novel procedure is to transfer whole political subdivisions in stepwise series along the boundary of two districts and add only layer-by-layer to achieve population equality. The process starts with collecting the 67 PA counties into roughly equal population district: 50 districts for the Senate map. Next, whole townships and wards are swapped between districts aiming to equalize the districts’ population. The township/ward swaps are only made in linear steps along the districts’ boundaries. Adhering strictly to the boundary prevents the creation of meandering tentacles favored by those who gerrymander for political advantage. This method is easily observable by citizens, all political parties and actors, and the judges in the courts. The Senate map was constructed using Dave’s Redistricting App (DRA), an online open redistricting tool being used by many researchers and interested citizens. DRA’s partisan and demographic layers were not consulted during map design. Only population totals and required redistricting geography (counties, townships, cities, wards, and precincts) were used. Ward geographies are not present in DRA. Ward boundaries were downloaded from the LRC’s official redistricting data release. Examples in the Senate map submitted herein demonstrate the method and its simple application. One example that is easy to see lies in north central PA. The boundary between districts 48 and 35 within Potter County shows how one north-south line of townships (rectangular Sharon Twp. at the north edge and triangular Portage Two. at the south) were added from district 35 to 48. Finally, to achieve the equal population within the allowable variance, three additional townships were moved from 35 to 48; these being Oswayo, Hebron, and Eulalia from north to south. The embedded boroughs of Oswayo and Coudersport move as well. Another example is the three districts in Luzerne and Lackawanna Counties – districts 27, 28 and 29. Two borders of the middle district 28 demonstrate the CCFD construction method. On the northeast and southwest boundaries of district 28 the townships were added in straight linear layers to produce the most compact district, even with challengingly irregular township shapes. Identical use of layers can be seen in southeast PA in Bucks (districts 7-8-9), Chester (districts 15-16), Montgomery (3-ways districts 9-10-11), and in southwest PA in Fayette (districts 38-39), Westmoreland (districts 38-40) and Mercer (districts 46-49). Some unique cases can be observed. In Butler County, linear layers could not be used due to the large population of Butler city in the center of the county, so a square, compact boundary was necessary (districts 45-47). Likewise, the large population of the city of Lancaster at the center of Lancaster County required layering around that population center (districts 17-18-19). The CCFD method in these situations is still appliable, and judicially comprehensible and manageable. Only if the districts’ populations exceed the variance band would townships or wards be split along precinct lines. In the Senate map submitted herein, only within the cities of Philadelphia and Pittsburgh are ward split. No other townships or wards are split statewide. Remarkable. The CCFD method applies in Pennsylvania’s largest cities – Pittsburgh and Philadelphia. In Allegheny County, even though Pittsburgh’s population exceeds that of a target district, wards in the southeast portion are moved layer-by-later into the adjoining district (districts 42-41). The other districts surrounding Pittsburgh are similarly created with layers of whole townships on district boundaries (districts 42-41-43-44). In Philadelphia, layer-by-layer movement of wards create compact districts (districts 6-5-4-3-2-1) The CCFD methods produces high statewide values for important metrics of redistricting criteria ( DRA metrics normalized to a 0 to 100 scale: proportionality = 91, splitting = 74, compactness = 69). Minority representation (= 45) is sufficiently strong that only minor editing may be needed at the end of the drafting process to achieve VRA objectives. Though the DRA Competitiveness rating is low (=24) there is significant balance is the post-drafting partisan analysis (DRA Blue Lean = 20, Red Lean = 20, Even = 10 districts). The districts created likely already comply with racial requirements of the Voting Rights Act (VRA). If not, if racial, ethnic or communities of interest are split, one can fine tune them at the precinct level to achieve these redistricting objectives. CCFD claims its redistricting method is simple. How simple? In the summer of 2019, two CCFD college interns used this method to create 203-seat PA House maps for Draw the Lines in a few short weeks. In fact, one finished theirs during an all-nighter. The result - these students tied for 1st Place in the Higher Ed Division beating all other college entries in that competition. The CCFD method is that elementary. By using this simple method, splits in political units are minimized and districts remain compact. Consequently, it is a comprehensive solution to partisan gerrymandering that the courts can understand, manage, and enforce.

Quantitative Analysis

Number of Counties Split
Total Number of County Splits
Reock compactness
Polsby-Popper compactness
Overall Deviation
Average Absolute Deviation