FL: Rereredistricting

[Nyvin]

The lawsuit about the Fair Districts amendment limiting freedom of speech has been rightly tossed out by the courts:

http://miamiherald.typepad.com/nakedpolitics/2015/10/federal-court-rejects-lawsuit-challenging-anti-gerrymandering-law.html?utm_source=twitterfeed&utm_medium=twitter

There goes that option for Republicans...

[Torie]

The lawsuit about the Fair Districts amendment limiting freedom of speech has been rightly tossed out by the courts:

http://miamiherald.typepad.com/nakedpolitics/2015/10/federal-court-rejects-lawsuit-challenging-anti-gerrymandering-law.html?utm_source=twitterfeed&utm_medium=twitter

There goes that option for Republicans...

Except that it will be appealed presumably. It's an interesting case. The issue is sure, you can speak to anyone, but if you speak to the wrong person, it taints the map. In the end it is kind of an evidentiary issue I guess, as to whether it really tainted the map. If the case had been framed that the only evidence of taint was the conversation itself, perhaps that would be deemed an unacceptable chilling effect on speech. But I think here there is more evidence of taint, than just conversations.

[Nyvin]

The lawsuit about the Fair Districts amendment limiting freedom of speech has been rightly tossed out by the courts:

http://miamiherald.typepad.com/nakedpolitics/2015/10/federal-court-rejects-lawsuit-challenging-anti-gerrymandering-law.html?utm_source=twitterfeed&utm_medium=twitter

There goes that option for Republicans...

Except that it will be appealed presumably. It's an interesting case. The issue is sure, you can speak to anyone, but if you speak to the wrong person, it taints the map. In the end it is kind of an evidentiary issue I guess, as to whether it really tainted the map. If the case had been framed that the only evidence of taint was the conversation itself, perhaps that would be deemed an unacceptable chilling effect on speech. But I think here there is more evidence of taint, than just conversations.

A majority (8D-3R) of active judges on the 11th Circuit Court are Democrat appointed judges....but good luck with that.

[Torie]

The lawsuit about the Fair Districts amendment limiting freedom of speech has been rightly tossed out by the courts:

http://miamiherald.typepad.com/nakedpolitics/2015/10/federal-court-rejects-lawsuit-challenging-anti-gerrymandering-law.html?utm_source=twitterfeed&utm_medium=twitter

There goes that option for Republicans...

Except that it will be appealed presumably. It's an interesting case. The issue is sure, you can speak to anyone, but if you speak to the wrong person, it taints the map. In the end it is kind of an evidentiary issue I guess, as to whether it really tainted the map. If the case had been framed that the only evidence of taint was the conversation itself, perhaps that would be deemed an unacceptable chilling effect on speech. But I think here there is more evidence of taint, than just conversations.

A majority (8D-3R) of active judges on the 11th Circuit Court are Democrat appointed judges....but good luck with that.

Well First Amendment issues are hyper sensitive ones. If the case were postured right (again it probably is not here), SCOTUS might indeed take it up if the 11th Circuit got it wrong (SCOTUS is very, very pro First Amendment, particularly when it comes to speech, and that crosses the ideological divide there to a substantial extent).

[jimrtex]

The 6 base plans drawn by the legislative staff have been released:

Senate Redistricting Website

Two methodologies have been used to draw the 6 proposed base maps:

Method 1:

"In drawing districts, consistently respect county boundaries by keeping counties whole and
keeping districts entirely within counties, where feasible."

Method 2:

"In drawing districts, consistently respect county boundaries by minimizing the number of times
each county is split into more than one district, as well as the aggregate number of county splits
statewide, where feasible."

Looking at the set of maps drawn with each methodology I can't figure out the difference. Those which were said to follow method 2 appears to split more smaller counties.

[Nyvin]

Other than the Tampa area,  Senate Draft 9078 looks the best.    It's 21R-19D (around that at least).    If the Tampa-St Petersburg district doesn't hold up and is given to the Dems it'd almost be an evenly divided map (which "Florida" should be anyway).    Although it'd still favor the Republicans slightly.

[Bigby]

Other than the Tampa area,  Senate Draft 9078 looks the best.    It's 21R-19D (around that at least).    If the Tampa-St Petersburg district doesn't hold up and is given to the Dems it'd almost be an evenly divided map (which "Florida" should be anyway).    Although it'd still favor the Republicans slightly.

21 R - 19 D? That's 40 districts! I thought that Florida only had 27?

[Nyvin]

Other than the Tampa area,  Senate Draft 9078 looks the best.    It's 21R-19D (around that at least).    If the Tampa-St Petersburg district doesn't hold up and is given to the Dems it'd almost be an evenly divided map (which "Florida" should be anyway).    Although it'd still favor the Republicans slightly.

21 R - 19 D? That's 40 districts! I thought that Florida only had 27?

It's for the state Senate.

[Bigby]

Other than the Tampa area,  Senate Draft 9078 looks the best.    It's 21R-19D (around that at least).    If the Tampa-St Petersburg district doesn't hold up and is given to the Dems it'd almost be an evenly divided map (which "Florida" should be anyway).    Although it'd still favor the Republicans slightly.

21 R - 19 D? That's 40 districts! I thought that Florida only had 27?

It's for the state Senate.

Oh, I apologize. I thought that we were still discussing the House map.

[jimrtex]

Other than the Tampa area,  Senate Draft 9078 looks the best.    It's 21R-19D (around that at least).    If the Tampa-St Petersburg district doesn't hold up and is given to the Dems it'd almost be an evenly divided map (which "Florida" should be anyway).    Although it'd still favor the Republicans slightly.

21 R - 19 D? That's 40 districts! I thought that Florida only had 27?

It's for the state Senate.

Oh, I apologize. I thought that we were still discussing the House map.

During discovery for the congressional map, it was also discovered that there had been partisan intent in drawing the senate map. I read the deposition of the senate map-drawer.

He drew the initial base map. Somehow, some Republican map drawers got a copy of the map. They would make modifications and submit them as public plans. The senate map-drawer would incorporate them into the senate map. When the map was explained to the legislature, it would be that District XX was from Public Plan YY, and it would do something useful for County CC, masking the intent to draw a more Republican map. There was circumstantial evidence, such as the senate map-drawer only keeping Draft 9, and the map the Republican map-drawer being numbered as 9, with lots of similarities.

In a press conference, the lawyer for the LWV plaintiffs said that it might never be known how communication worked. And the lawyer for the senate would every so often tell the senate map-drawer to "not answer" a particular question.

This led to overturning of the congressional map, and the senate agreed that the senate map should also be redrawn. I suspect that since it had been the senate that messed up led to the House being unwilling to accept any senate changes to the congressional map during the special session.

A special session to produce a new senate map opens on Monday. 6 base plans were produced by the senate and house staff map-drawers, and have been released this week. All their drafting sessions were recorded, and they had no outside contact, other than with lawyers (and attorney-client privilege has been waived).

[Torie]

The 6 base plans drawn by the legislative staff have been released:

Senate Redistricting Website

Two methodologies have been used to draw the 6 proposed base maps:

Method 1:

"In drawing districts, consistently respect county boundaries by keeping counties whole and
keeping districts entirely within counties, where feasible."

Method 2:

"In drawing districts, consistently respect county boundaries by minimizing the number of times
each county is split into more than one district, as well as the aggregate number of county splits
statewide, where feasible."

Looking at the set of maps drawn with each methodology I can't figure out the difference. Those which were said to follow method 2 appears to split more smaller counties.

Isn't method 2 what was used for the Congressional district drawing that Muon2 pointed out, where you don't get penalized for the second and additional chops of counties? The first method where it refers to keeping districts wholly within counties does seems to treat the first and additional chops in the same manner, as something to be avoided where possible.

[jimrtex]

The 6 base plans drawn by the legislative staff have been released:

Senate Redistricting Website

Two methodologies have been used to draw the 6 proposed base maps:

Method 1:

"In drawing districts, consistently respect county boundaries by keeping counties whole and
keeping districts entirely within counties, where feasible."

Method 2:

"In drawing districts, consistently respect county boundaries by minimizing the number of times
each county is split into more than one district, as well as the aggregate number of county splits
statewide, where feasible."

Looking at the set of maps drawn with each methodology I can't figure out the difference. Those which were said to follow method 2 appears to split more smaller counties.

Isn't method 2 what was used for the Congressional district drawing that Muon2 pointed out, where you don't get penalized for the second and additional chops of counties? The first method where it refers to keeping districts wholly within counties does seems to treat the first and additional chops in the same manner, as something to be avoided where possible.

I couldn't understand the difference. The senate and house have their own websites, and generate their own reports. I'm not certain that they actually chose a metric, but rather put a lot of metrics, and that may have accidentally guided their choice of rules.

[jimrtex]

If Florida were to switch to weighted voting, it would be a lot easier to draw maps that respect county boundaries.



I determined the total entitlement for counties entitled to less than one full senate vote (1/40 of the total population or 470,056, and rounded to the nearest integer (14.356 => 14).

The other 26 (= 40 - 14) senators were apportioned among the larger counties, using the harmonic mean as the divisor, since the goal is to minimize the disparity among the districts. Monroe was considered part of Miami-Dade for this purpose.

The smaller counties were split by Polk-Orange, into a northern area (9 districts) Lake and northward; and a southern area (5 districts) Osceola and southward. The districts in the northern area ended up being a little bit oversized, so I tried to balance their populations. In the southwest area there is an alternative that is a bit-more population balanced, but at the expense of compactness.

[muon2]

The map jimrtex created seems like an excellent starting point to find a plan the chops the fewest counties. Following the example of Miami-Dade+Monroe, I appended some counties to the large counties to bring districts within 5% of the quota. The remaining small counties were then balance to provide for all districts within 5% of the quota.

[jimrtex]

The map jimrtex created seems like an excellent starting point to find a plan the chops the fewest counties. Following the example of Miami-Dade+Monroe, I appended some counties to the large counties to bring districts within 5% of the quota. The remaining small counties were then balance to provide for all districts within 5% of the quota.

The base maps may well have followed a similar procedure. For example, they have two districts covering Duval+Nassau, and 12 districts covering the southeastern four counties: Palm Beach, Broward, Miami-Dade, and Monroe.

They set a 4% maximum deviation range constraint. I do not know the legal basis for this. Before 2010, the Florida Constitution did not require population equality for legislative districts. Any requirement would be under the 14th Amendment, or possibly something that the SCOFLA found in the Florida constitution. There was a review process prior to 2010, and perhaps they included a maximum deviation.

The 2010 amendment added these tier two standards:

"(b) Unless compliance with the standards in this subsection conflicts with the standards in subsection (a) or with federal law, districts shall be as nearly equal in population as is practicable; districts shall be compact; and districts shall, where feasible, utilize existing political and geographical boundaries."

A 10% deviation range does not conflict with federal law. Since the main reason that the SCOTUS has permitted that range is to permit utilization of existing political boundaries, it would be reasonable to infer that the voters wished to expand the deviation limit. Before 2010, not only was there no equality requirement, there were no other standards. The SCOFLA could have imposed a tighter limit, just as there has been a supposition that there is no minimum limit for congressional districts since they could be drawn to the census block level.

The 2010 standards also said:

"(c) The order in which the standards within subsections (a) and (b) of this section are set forth shall not be read to establish any priority of one standard over the other within that subsection."

Arguably setting a 4% deviation limit is giving priority to population equality over respect for political boundaries. Use of county boundaries is made less feasible through adoption of an impractical, possible ad hoc, equality standard.

To achieve a 21:19 majority would require the support of senators that represent 20.6 / 40 of the population (51.5%) vs. the nominal 52.5%.

We can compare the effect of the two standards:

Your map

Large counties with whole number of districts:

Pinellas: 2

Large counties with whole districts plus one fragment extending outside the county (Texas House rules).

Duval: 1+
Volusia: 1+
Orange: 2+
Brevard: 1+
Hillsborough: 2+
Lee: 1+
Palm Beach: 2+
Broward: 4+
Miami-Dade(+Monroe): 5+

Split Large Counties (no whole district)

Polk

Split Small Counties(*)

Marion
Sarasota

Thus there are two excess splits, plus the one non-optimal for Polk County.

(*)These are the most likely splits, and have the benefit of splitting fairly large counties.

Base Map 9072

Duval: 1+
Orange: 2+
Lee: 1+
Palm Beach: 2+
Miami-Dade(+Monroe): 5+

Split Large Counties (no whole districts):

Brevard

Large Counties with whole districts + excessive fragments.

Broward 4+2 fragments
Polk 1+2

Large Counties with fewer whole districts than possible + excessive fragments

Pinellas 1 + 2 fragments
Hillsborough 1 + 3 fragments
Volusia 0 + 3 fragments

Split Small Counties:

Okaloosa
Marion (3 fragments)
Manatee

Voting Rights Districts

There are 9 minority opportunity districts:

6 are in Miami-Dade+Broward, including a Black opportunity districts straddling the Broward-Dade county line. Under your map it would be shifted about 50,000 persons south, which is unlikely to flip it.

The other three are in Jacksonville, Orlando, and Tampa-St.Petersburg.

Your map would have no effect on the Jacksonville district. It might have an effect on the Orlando district. In 9072 it is drawn to the northwest corner of Orange County, but that is partially for compactness reasons, since the Orange County fragment attaches to Osceola in that map. The portions of Orange County that would attach to Lake County could wrap around Orlando. Moving the Orlando district southeast could improve its performance (it would now be in the west central part of the county.

The final district currently crosses Tampa Bay. The issue is whether the St.Petersburg portion of the district is necessary to maintain it as a Black opportunity district may be at issue.

Your map would presumably place the eastern (and perhaps northeastern portion) of Hillsborough County with western  or northern Polk County. This may then cause a split of the black population in Hillsborough, which is not really on the western edge of the county, unless you consider the bay to be the western edge of county, that makes the jump to St.Petersburg more plausible. It could be that the black population fails the Gingles Test since you have two separate black communities with a bay between them.

An Alternative Map in north central Florida would put Sumter, Hernando, Citrus, Levy and Dixie in the single member district, this would move the split to Alachua, but produce two relative compact districts: Marion, Putnam, plus some of Alachua; and most of Alachua with the rural northern counties, with a small decrease in population equality.

[muon2]

The Orange opportunity SD 12 in 9072 is 35.9% BVAP and 21.2% HVAP+HBVAP. In my version shifting it SE in Orange gets 32.1% BVAP and 27.7% HVAP+OVAP. The minority performance does rise slightly and the chop count goes down for the Lake SD, but at the cost of compactness and greater inequality.

The Tampa-St Pete opportunity SD 12 in 9072 is 34.5% BVAP and 26.5% HVAP+HBVAP. In my version putting it entirely in Hillsborough gets 21.7% BVAP and 32.3% HVAP+OVAP. There is a definite drop in minority performance and again 9072 has lower inequality, but my plan would be substantially more compact and it avoids the traveling chop of Pinellas and Hillsborough.

[jimrtex]

The Orange opportunity SD 12 in 9072 is 35.9% BVAP and 21.2% HVAP+HBVAP. In my version shifting it SE in Orange gets 32.1% BVAP and 27.7% HVAP+OVAP. The minority performance does rise slightly and the chop count goes down for the Lake SD, but at the cost of compactness and greater inequality.

The Tampa-St Pete opportunity SD 12 in 9072 is 34.5% BVAP and 26.5% HVAP+HBVAP. In my version putting it entirely in Hillsborough gets 21.7% BVAP and 32.3% HVAP+OVAP. There is a definite drop in minority performance and again 9072 has lower inequality, but my plan would be substantially more compact and it avoids the traveling chop of Pinellas and Hillsborough.

Have you submitted your map to the state of Florida, or were you using DRA?

[muon2]

The Orange opportunity SD 12 in 9072 is 35.9% BVAP and 21.2% HVAP+HBVAP. In my version shifting it SE in Orange gets 32.1% BVAP and 27.7% HVAP+OVAP. The minority performance does rise slightly and the chop count goes down for the Lake SD, but at the cost of compactness and greater inequality.

The Tampa-St Pete opportunity SD 12 in 9072 is 34.5% BVAP and 26.5% HVAP+HBVAP. In my version putting it entirely in Hillsborough gets 21.7% BVAP and 32.3% HVAP+OVAP. There is a definite drop in minority performance and again 9072 has lower inequality, but my plan would be substantially more compact and it avoids the traveling chop of Pinellas and Hillsborough.

Have you submitted your map to the state of Florida, or were you using DRA?

I didn't know they were still taking submissions. I think I still have an account on their website with the software. I would also have to know where the 4% tolerance comes from before submitting anything.

[jimrtex]

The Orange opportunity SD 12 in 9072 is 35.9% BVAP and 21.2% HVAP+HBVAP. In my version shifting it SE in Orange gets 32.1% BVAP and 27.7% HVAP+OVAP. The minority performance does rise slightly and the chop count goes down for the Lake SD, but at the cost of compactness and greater inequality.

The Tampa-St Pete opportunity SD 12 in 9072 is 34.5% BVAP and 26.5% HVAP+HBVAP. In my version putting it entirely in Hillsborough gets 21.7% BVAP and 32.3% HVAP+OVAP. There is a definite drop in minority performance and again 9072 has lower inequality, but my plan would be substantially more compact and it avoids the traveling chop of Pinellas and Hillsborough.

Have you submitted your map to the state of Florida, or were you using DRA?

I didn't know they were still taking submissions. I think I still have an account on their website with the software. I would also have to know where the 4% tolerance comes from before submitting anything.

I don't know (on either question).

The special session started today. I started to fill out a comment form about using a broader deviation range, and got distracted and started drawing a map with the senate website. I have roughed in the counties, reserving district numbers for the large county. I have split Duval County, and have about finished a division of Alachua.

It is possible that equality is discussed in the initial SCOFLA decisions in 2012. The legislative maps are subject to automatic review, and this was done in 2012. The Senate map was rejected at that time, as the SCOFLA got its first chance to interpret the new redistricting standards. The senate then drew the current map.

My recollection is that the SCOFLA thought that they had split too many cities and counties, and had not used major highways and roads. For example, they had split the panhandle north-south to create a coastal district and an inland district. The SCOFLA specifically rejected COI if it superseded tier 2 criteria (political boundaries, compactness, etc.).

In the recent congressional districting trial this was references with regard to the Palm Beach division of 21/22. The Romo plaintiffs had argued for the east/west split (beach vs. inland) based on COI, but the final decision was based on compactness. You can use COI, but only after you have met all the other standards.

It is conceivable that the legislature had adopted a tighter standard in order to force additional county splits.

[jimrtex]

The Orange opportunity SD 12 in 9072 is 35.9% BVAP and 21.2% HVAP+HBVAP. In my version shifting it SE in Orange gets 32.1% BVAP and 27.7% HVAP+OVAP. The minority performance does rise slightly and the chop count goes down for the Lake SD, but at the cost of compactness and greater inequality.

The Tampa-St Pete opportunity SD 12 in 9072 is 34.5% BVAP and 26.5% HVAP+HBVAP. In my version putting it entirely in Hillsborough gets 21.7% BVAP and 32.3% HVAP+OVAP. There is a definite drop in minority performance and again 9072 has lower inequality, but my plan would be substantially more compact and it avoids the traveling chop of Pinellas and Hillsborough.

Have you submitted your map to the state of Florida, or were you using DRA?

I didn't know they were still taking submissions. I think I still have an account on their website with the software. I would also have to know where the 4% tolerance comes from before submitting anything.

This is SCOFLA opinion in Reapportionment 1, which was its review of the initial legislature plans in 2012.

SCOFLA Reapportionment 1

I have got a lot further. It is pretty comprehensive once you get through the part where they justify their role.

The SCOFLA interprets the minority right provisions as being equivalent to Section 2 and Section 5 of the VRA. Their understanding of Section 2 is based primarily on Ginglex, and thus requires the presence of a compact majority minority district.

Their interpretation of Section 5 is that preclearance was to ensure non-retrogression and protects both majority-minority and crossover/coalition districts. They approved a house district that crossed from Tampa to St.Petersburg to Bradenton and then squiggled to Sarasota.

They might not be willing to accept a Hillsborough district that will elect a Democrat, just because Blacks vote for a Democrat in the general election.

They seem to accept the federal 10% deviation range limit, but won't actually say so. In previous cycles, they declined to set a tighter standard. The deviation for the House plan is 3.97%, and the SCOFLA noted that was slightly larger than in 2002. But there were no explicit constraints in 2012.

This is also a case where the range is simplistic. Your range is 7.2%, but 38 districts are under 2.7%.

[muon2]

Is this better? I increased the chop count by 1 to decrease the range from 7.2% to 5.3%. With 2 more chops in Hillsborough and Broward I could reduce it to 3.8%.

[jimrtex]

Is this better? I increased the chop count by 1 to decrease the range from 7.2% to 5.3%. With 2 more chops in Hillsborough and Broward I could reduce it to 3.8%.

Depends what you mean by better.

I started listing to the recordings of the map-drawers.

They started their first meeting talking about minority districts, particularly in the northern part of the state. They decided that they would try to draw the black-opportunity district in Hillsborough, but might have to go into Pinellas. They noted that the SCOFLA in the congressional case noted that FL-14 was non-performing, so that the excursion into St. Petersburg might not be really necessary, since it would not be retrogression, while at the senate level it might be.

In the House map, where there are a number of minority districts which are quite non-compact where they were able to collect enough minorities in a smaller district within a county, but could not for Senate or Congressional districts. There are 120 House districts, vs 40 senate districts and 27 congressional districts.

They passed over Jacksonville quickly. In Orlando, they indicated one question that they would have to consult with lawyers was whether a Hispanic majority district was now necessary, now that the current map demonstrated that it could be done.

They seem to understand the difference between the methodologies - but I don't. That may be a refusal to understand by me, who hasn't bought into cover and pack rules. They appear to have decided to draw several alternatives with one methodology, and then go to the other.

They then discussed the overall approach. (1) was to draw the minority districts. While that was OK for the congressional remedial map where the SCOFLA had directed their attention to certain districts, they generally agreed to start from a clean map. (2) The second was to start at one end, and start drawing. This has an an appeal in Florida since you can easily draw districts in the panhandle, and put off the southeast districts where you have the complexity of the minority districts.

Then one suggested that they block the counties off into groups that could have a whole number of districts, based on the deviation from the memo. I'm pretty sure the 4% was an unnecessary constraint, given the other constraints. It is only when you have no constraints, where 10% is too large. In splitting counties, I have tried to equalize the population of districts, within that block of counties, which gets you better overall equality than would be anticipated from a simple 5% limit.

They agreed they could find these "sandboxes" with a spreadsheet, Paint, and few hours on the Atlas. (they didn't actually mention the Atlas nor Paint, but suggested that this map could be produced informally, and that they could use District Builder to produce an equivalent map to explain the process to legislators (eg the deviation might show as 470,000 to large for an area to be divided into 2 districts.

When I did it, I would split your areas into separate districts. For example, Lake was 12, and Orange was 13, with 14 skipped. When I started dividing Orange, I drew 14 as the black opportunity district, and augmented 12 from Orange County.

The first discussion was paused, as one of three went to get his spreadsheet that could be used to add county populations uo.

They also discusses where they might go from decision points, which may have been increased because they had to work with such a small deviation limit. On the other hand, they appear to think the deviation would give them lots of flexibility as compared to the congressional map. The SCOFLA opinion in Reapportionment 1 contrasts Karcher v Daggett, with standards for legislative redistricting. But with respect to legislative districting, they refused to acknowledge the 10% deviation from White v Regester. Reapportionment 1 was before Tennant, and in the recent map drawing it doesn't appear that they considered increasing deviation more than one person.

They covered the minority districts in the southeast, and acknowledged that SD-39 was a challenge because it is neither Black majority or Hispanic majority. In Houston, there could be additional Hispanic opportunity districts, except a 30% Black, 60% Hispanic, 10% Anglo district, is a performing Black opportunity district. If you took in a large Anglo population, you would make it a Republican district, so you would end up packing black voters in order for Hispanics to actually elect a candidate of choice.

They noted that Hendry County had been one of 5 counties covered by Section 5, and that was one reason both congressional and senate districts had been extended west. They felt that was no longer necessary.

I have completed the division of Duval, Alachua, and Volusia, and am working on Orange County.

Cities like Apopka and Ocooe appear to be similar to some cities in the Dallas area such as Grand Prairie, Mesquite, and Garland, where there is a census tract or two in each of the farm town central areas, that has remained even after the suburbs have swamped the area.

I started out adding to Lake County from the north and was doing fine, until I figured out how to turn on the racial composition. So I started drawing the black district, clicking on the 30+ VTDs, and finding that I was going to have to go west towards Apopka and Ocoee. I wasn't getting enough population for a district, so I was dropping down into the 20s and teens where it helped compactness.

When you go east in Orlando, the black to white gradient is quite sharp, similar to in Cincinnati. So I finally gave in on coming into Orange County from Lake in two parts, and have come in from the south, forming a C shaped district. The other district is reasonably compact for a minority opportunity district that I was searching for population for. The remaining eastern Orange district is almost rectangular, but no where a Hispanic majority district.

[jimrtex]

This is my map. There are four public plans by 3 persons on the House web site. I have yet to submit my plan, but I think I understand how to do so. The following is a screen shot of the PDF produced by MyDistrictBuilder. It is supposed to be producing a complete set of maps, but it is taking a while.



The map is based on muon2's proposed whole-district regions with the exception of North Central Florida where I made the one-district region along the coast, and the two-district region inland, splitting Alachua County.

(1) Escambia and Santa Rosa. (deviation from ideal -4.5%)

(2) Bay, Gulf, Holmes, Okaloosa, Walton, and Washington. (-1.0%)

(3) Calhoun, Franklin, Gadsden, Jackson, Jefferson, Leon, Liberty, and Wakulla. (-3.9%)

(4, 5) Alachua, Baker, Bradford, Columbia, Dixie(*), Gilchrist, Hamilton, Lafayette, Madison, Marion, Putnam, Suwannee, Taylor, and Union. (-2.6%)

(*) Dixie is technically not self-contiguous. As the Suwanee River goes south of Lafayette County, it meanders to the west and north to touch the Lafayette County line, before heading south. This isolates a tiny corner of Dixie from the bulk of the county. To maintain district contiguity, this sliver has been placed in District 4.

This required a split of Alachua County, with the larger portion in the northern district (SD-4). I therefore assumed that Gainesville would be in that portion. I started adding to the southern district (SD-5), and then snapped the boundary to the Gainesville city limits. To get sufficient population requires wrapping around Gainesville on both the east and west. Including Newberry in SD-5, reduces the wrap around effect a bit, and allows for a more relaxed following of the irregular Gainesville city limits. In general, when splitting counties, I strived for a high degree of intra-county population equality, to make up for the broadened extra-county limits.

(4) Alachua (part, 78% of county), Baker, Bradford, Columbia, Dixie(*), Gilchrist, Hamilton, Lafayette, Madison, Suwannee, Taylor, and Union. (-3.0%)



(5) Alachua (part 22%), Marion, and Putnam. (-2.2%)



(6, 7) Duval and Nassau (-0.3%)

I started by adding the parts of Duval County that aren't in Jacksonville to Nassau County in FL-6. I then began adding in areas to make the district contiguous. To get to near the ideal population I added in larger VTD's (presumably less dense). As I got near the target population I noticed that the inner border was close to I-295. I started placing the boundary on I-295 and found that it was very close to the correct population. I-295 dips into Clay County, and then when it emerges to the east leaves a relatively small strip south of I-295. I shifted off of I-295 to provide better connectivity. This explains why the portion of I-295 not used is on the southeast quadrant.

(6) Duval (part, 46%) and Nassau. (-0.1%)



(7) Duval (part, 54%). (-0.4%)



The BVAP for my FL-7 is 37.9% vs. 42.7% for the comparable district in base map 9072. The base map does a better job of race sorting: 43%:11% for the two Duval districts in the base map, against 38%:14% for my map, albeit with more irregular borders.

(8 ) Clay, Flagler, and St. Johns. (+1.4%)

(9) Citrus, Dixie(*), Hernando, Levy, and Sumter. (-1.1%)

(*) The tiny discontiguous portion of Dixie is in FL-4.

(10, 11) Seminole and Volusia (-2.4%)

Both counties are near the ideal population size, with only a small transfer from Volusia to equalize the two. The cities of DeBary and Orange City, just north of the county line, and to the west side of I-4 were added to the Seminole-based FL-11.

(10) Volusia (part, 93%). (-2.6%)

(11) Seminole and Volusia (part, 7%). (-2.3%)



(12, 13, 14) Lake and Orange (+2.4%)

The intent was to create one district (SD-12) in Lake County, plus a portion of Orange County, and two districts (SD-13, 14) wholly in Orange County. I initially tried to come into Orange County from the northwest, filling in the inner corner of Lake County. This worked out OK for SD-12, but removed too much of the black population. The black population in Orange County is not monolithic, and there are pockets in cities like Apopka and Ocoee.

So I extend SD-14 into the black areas of Apopka and Ocoee, and then included the entirety of those cities. To make up for the population loss in SD-12, I came into Orange County in the southwest, below Lake Apopka. It might be possible to to get the black population of SD-14 higher by having SD-12 come into Orange County in two or three places, but that offends my sensibilities of compactness.

No effort was made to place all of Orlando in one district, though I did keep it out of SD-12. Winter Garden is wholly in SD-12; Apopka, Ocoee, and Maitland in SD-14, and Winter Park in SD-13.

After drawing SD-12 and SD-14, SD-13 is simply the remaining eastern part of the county.

(12) Lake and Orange (part, 16%) (+2.2%)



This includes the southwest part of Orange County.

(13) Orange (part, 42%) (+2.6%)



This is the eastern part of Orange County.

(14) Orange (part, 42%) (+2.2%)



This is northwestern and north central Orange County. My version has a 34.4% BVAP and 18.9% HVAP. The equivalent version in base map 9072 is 35/9% BVAP and 18.9% HVAP.

(15, 16) Brevard, Indian River, and Osceola (+1.1%)

To place one district wholly in Brevard requires a split of Palm Bay. Rather than a north-south split, I have opted to include the western part of the city, as part of a fill-in of the Indian River-Osceola inner corner.

(15) Brevard (part, 88%) (+1.3%)



(16) Brevard (part, 12%), Indian River, and Osceola (+0.8%)



(17) Pasco. (-1.1%)

(18, 19) Pinelllas (-2.5%)

A simple split to equalize population and avoiding city splits. FL-18 on the north includes Largo, while FL-19 on the south includes Seminole and Pinellas Park.

(18) Pinellas (part, 50%). (-2.2%)



(19) Pinellas (part, 50%). (-2.8%)



(20, 21, 22, 23) Highlands, Hillsborough, and Polk. (+2.7%)

I tried to create a black opportunity district in Hillsborough County, but failed. It was made harder by the need to create another district to the west of it. All base maps maintain the cross-bay district, which can only be justified on the basis of race. The bridges into St.Petersburg are from the Interbay Peninsula, which a black district must skip along with St.Petersburg. If a map were drawn with dots representing where people lived, it would be obvious what was being done. The base map is 34% BVAP and 24% HVAP. My district is 29% BVAP and 20% HVAP.

After drawing FL-20 and FL-21, the two districts in Hillsborough County, I drew FL-23 which begins with Highlands plus an eastern chunk of Polk. There are a number of cities along the eastern boundary of the county, which I eventually followed to the northern end of the county. and then started working along the southern part of the county. This leaves the area around Lakeland in FL-22 with the eastern part of Hillsborough.

I have just realized that a full district could be drawn in Polk, with a connector in the southern part of the county between Hillsborough and Highlands. It would be ugly, but would create the maximum whole districts in every large county (if we consider Monroe to be an extension of Miami-Dade).

It would also be possible to rearrange the regions, with Hillsborough, Manatee, and Sarasota in one region, and the other from Polk to Charlotte. This would move a split from Sarasota north to Manatee.

(20) Hillsborough (part, 39%). (+2.6%)



The district includes northwest Hillsborough, the Interbay Peninsula, and a strip along the eastern edge of the bay to get enough population. Hillsborough only has three cities, and I made no attempt to keep Tampa whole.

(21) Hillsborough (part, 39%). (+2.7%)



The district includes most of the northern part of Tampa, and the city of Temple Terrace. It has a BVAP of 29.1%, and HVAP of 20.1%.

(22) Hillsborough (part, 21%), and Polk (part, 36%) (+2.5%)



The district includes the eastern remainder of Hillsborough County, including Plant City, and the northwestern part of Polk County, around Lakeland.



This is zoomed in on the central portion of the district.

(23) Highlands and Polk (part, 64%). (+2.8%)



FL-23 is the eastern and southern part of Polk County, plus Highland County.

[jimrtex]

The panhandle region include SD-1, SD-2, and SD-3.



The northeast region includes SD-4 through SD-9



The Orlando-Central East Coast region (Brevard, Indian River, Lake, Orange, Osceola, Seminole, and Volusia), showing SD-10 through SD-16. The northern parts of Volusia, and southern parts of Indian River and Osceola are clipped.



The Tampa Bay region (Hillsborough, Pasco, Pinellas, and Polk), showing senate districts SD-17 through SD-23. SD-23 also includes Highlands (partially shown).



The Southwest region includes SD-24 through SD-27, and the Southeast region SD-28 through SD-31.



Broward, Miami-Dade, and Monroe include SD-32 through SD-40.

[jimrtex]

(24, 25) Charlotte, DeSoto, Hardee, Manatee, and Sarasota (-1.6%)

It is conceivable to put Manatee and Charlotte in the same district, with Sarasota taking a portion of Manatee or Charlotte. but I used the simpler solution of extending from Manatee south into Sarasota. I split the two eastern counties of Hardee and DeSoto, because that gave me a little wider berth around the city of Sarasota in FL-24.

(24) Hardee, Manatee, and Sarasota (part, 30%). (-1.6%)



FL-24 includes Manatee, Hardee, and the city of Sarasota in Sarasota County.

(24, 25) Charlotte, DeSoto, and Sarasota (part, 70%) (-1.7%)



FL-25 consists of Charlotte, DeSoto and the major portion of Sarasota County, including North Port.

(26, 27) Collier and Lee. (+0.0%)

Cape Coral and Fort Myers were placed in FL-26, with Bonita Springs and the southern part of the county in FL-27.

(26) Lee (part, 76%). (+0.1%)



FL-26 includes Cape Coral and Fort Myers

(27) Collier and Lee (part, 24%). (-0.0%)



FL-28 includes Collier County and the southeastern part of Lee County.

(28) Martin, Okeechobee, and St. Lucie. (-1.3%)

(29, 30, 31) Glades, Hendry, and Palm Beach (-2.7%)

I began with SD-30 drawing northward in the built-up area including Boca Raton, Delray Beach, Boynton Beach and areas to the west of these. I had initially intended to build the second district in the county from the Martin County line towards the south. This would have left the the district including Glades, Hendry, and western Palm Beach County, with a peninsula through to the coast. This was not working out well, so I slid SD-31 to include West Palm Beach, Riviera Beach, Greenacres, and Lake Worth in the north central part of the county. SD-29 includes Glades and Hendry counties, the west central part of the built-up area including Wellington and Royal Palm Beach, and an area in the north including Jupiter and Palm Beach Gardens. These areas are not directly linked, but are reasonably geographically compact.

By accident, SD-31 has a higher minority concentration than the the base map, at BVAP 24% and HVAP 27%.

(29) Glades, Hendry, and Palm Beach (part, 31%). (-2.5%)



FL-29 includes the central part of Palm Beach County around Wellington and Royal Palm Beach, the northern area including Jupiter and Palm Beach Gardens, as well and Glades and Hendry counties.

(30) Palm Beach (part, 35%). (-2.6%)



FL-30 includes Boca Raton, Delray Beach and Boynton Beach.

(31) Palm Beach (part, 35%). (-2.9%)



FL-31 includes West Palm Beach, Greenacres, Lake Worth, and Riviera Beach. It is 24% BVAP, and 27% HVAP.

(32 through 40) Broward, Miami-Dade, Monroe. (+2.1%)

In Broward County, I was mainly trying to make a stack of SD-32 through SD-34, respecting city boundaries, with SD-35 being the start of black opportunity district crossing into Miami-Dade.

(32) Broward (part, 27%). (+2.0%)



SD-32 is Broward County north of Fort Lauderdale, including Coconut Creek, Deerfield Beach, Coral Springs, Margate, Pompano Beach, and North Lauderdale. The district also includes the virtually unpopulated inland area of the county.

(33) Broward (part, 27%). (+2.2%)



SD-33 includes Fort Lauderdale, and cities to the west, including Tamarac, Lauderhill, Sunrise, Oakland Park, and Lauderdale Lakes. The hook on the western end is formed by Sunrise as it wraps around Plantation. SD-33 is 36.2% BVAP and 16.4% HVAP. The base map 9072 has a Broward district that is 50.3% BVAP and 17.0% HVAP, but which ignores city limits. In that map Pompano Beach is diced between four senate districts, and Fort Lauderdale is also divided.

(34) Broward (part, 27%). (+1.1%)



SD-34 includes areas south of Fort Lauderdale, including Hollywood, Dania Beach and Hallandale Beach, and areas to the west, including Plantation, Davie, and Weston. Its somewhat irregular shape is to permit creation of the black opportunity SD-35 which  straddles the Broward-Miami-Dade county line. The dangles that hang down from the western part of the district are part of Davie and unincorporated areas. SD-34 is the least populated of the districts in the Miami-Dade-Broward area. Since the overall deviation for the area is a bit high (+1.9%), I tried to keep the populations of the individual districts about equal, while still respecting city limits.

(35) Broward (part, 27%) and Miami-Dade (part, 7%). (+2.3%)



The Broward County portion of SD-35 includes Pembroke Pines, Miramar, Southwest Ranches and Cooper City. The Dade County portion includes Miami Gardens and Opa-locka, and unincorporated areas to the west. The district has a BVAP of 41.0% and HVAP of 35.3%. The comparable district in base map 9072 is BVAP of 42.3% and HVAP of 35.0%.

In Miami-Dade, after completing SD_35, I drew in SD-40 coming in from the south, until I had sufficient population. SD-40 also includes Monroe County and the barely populated western part of Miami-Dade. Population-wise it is a north-south district, with a long tail down the Florida Keys. At this point, it appeared that the remainder of the county might be divided into four quadrants. Miami and Coral Gables almost made up the perfect population for a district, and in effect delimited the other three districts. Note, the districts should probably be renumbered into a greater north-south order.

(36) Miami-Dade (part, 19%). (+2.4%)



SD-36 is west of Coral Gables and is mostly comprised of unincorporated areas. The largest cities are Pinecrest and South Miami. The HVAP is 80.4%.

(37) Miami-Dade (part, 19%). (+2.0%)



SD-37 is in the northeast corner of the county. Its somewhat irregular shape is because of it being comprised of whole cities, and wrapping around Miami (which is entirely in SD-38), and squeezing between Hialeah and Miami in order to pick up enough population. It includes Aventura, North Miami Beach, North Miami, and Miami Beach. It has a BVAP of 32.1% and HVAP 41.3%.

(38) Miami-Dade (part, 19%). (+2.1%)



SD-38 is Miami and Coral Gable and a bit more to get it up to the ideal population. It has an HVAP of 69.8%.

(39) Miami-Dade (part, 19%). (+2.4%)



SD-39 is northwestward from Miami. It includes Hialeah (almost half the district), Miami Lakes, and Doral, and a lot of unincorporated areas. It has an HVAP of 91.0%.

(40) Miami-Dade (part, 16%) and Monroe. (+2.0%)



SD-40 is south of Miami. It includes Homestead, Florida City, Cutler Bay, and Palmetto City, and unincorporated areas. In addition, it includes the minimally populated areas in the western part of the county, and Monroe County (mainland and Florida Keys). The district is 51.5% HVAP and 18.5% BVAP.