Pages

Monday, April 18, 2022

The AK Redistricting Board's Amateur Decision Making Process - Plus Marcum and Binkley Reasons For 3B Votes

In my previous post, I put up the video of the meeting where the three Republican Redistricting Board members go through their reasons for choosing Option 3B.  It also includes Board member Borromeo's rejection of those assertions and a plea to Judge Matthews to just fix the map himself and not remand it to the Board because they aren't going to change.  

In the post prior to that I went through Board member Budd Simpson's reasons for supporting Option 3B - as he laid them out in Wednesday's meeting - and pointed out problems I had with his reasons.  I thought this next post I would do the same with member Marcum's and member Binkley's reasons.  (And I do that as sort of an addendum to this post at the bottom.) But it seemed to be more useful to demonstrate that their reasons do not add up to a good, professional decision making exercise.   And in doing that, try to summarize their strategy so it's easier to understand what they did and what they didn't do, in terms of good decision making.  (But I wasn't quite done with it and so I put up the video of that meeting yesterday instead.)

Using a decision making model allows me to establish the flakiness of how they - in their words and actions - decided that Option 3B was the best option.  Because flaky as it might seem to many, the districts they created, while not optimal, could be considered Constitutional by the courts.  That depends on how they interpret "as contiguous as practicable."  Since other criteria, such as deviation and compactness, are applied more strictly in densely populated urban areas than more sparsely populated rural areas, why shouldn't contiguity be treated the same?  It's much easier to be contiguous in a densely populated area.  Will they see that such an urban/rural differentiation makes sense now?  At least in Anchorage.  

But if the Courts view the maps as basically in compliance with the Contiguity requirement, the other way I see that they might block the map is  by deciding it was politically motivated gerrymandering.  Part of the evidence for reaching that  conclusion is to:

  • realize that the Board's decision making was haphazard at best,  simply the application of anecdotal evidence to 'prove' the 3B map was the best, because it's not the best on objective measures
  • wonder why the majority fought so hard for Senate pairings that ignore the obvious pairings and force together much less natural pairings and  
  • conclude there was an unspoken (by the majority) reason - getting another Republican Senate seat.


So first a brief description of what good decision making should NOT look like from a 1998 Harvard Business Review article, The Hidden Traps in Decision Making:

"So where do bad decisions come from? In many cases, they can be traced back to the way the decisions were made—the alternatives were not clearly defined, the right information was not collected, the costs and benefits were not accurately weighed. But sometimes the fault lies not in the decision-making process but rather in the mind of the decision maker. The way the human brain works can sabotage our decisions."

The article is more interested in the problems with the mind of the decision maker, and while I'm sure exploring the hidden biases in brains of the decision makers would yield fascinating results, we don't have to go there to find serious problems with the Redistricting Board's decision to select Option 3B over Option 2.  

If anyone wants to know what went wrong - and it's clear the decision was wrong from a public interest perspective - we need look no further than

  1. the alternatives were not clearly defined, 
  2. the right information was not collected, 
  3. the costs and benefits were not accurately weighed
Click to enlarge
More detailed maps with the Simpson post


  1. the alternatives were not clearly defined, 
    1. The Board settled into two map options.  
    2. One offered by the East Anchorage Plaintiffs - Option 2
    3. One offered by Randy Ruedrich - Option 3B
    4. They could have made more, but didn't.
    5. The 3B Option chose to 
      1. Pair north Muldoon (D20) with south Muldoon (D21)
    6. That left D22 as an orphan district (not connected to another district to make a Senate district)
      1. The two key options were to connect with D24 (to the north above the map) or D9.  Both were connected to other house districts, so both required at least one more change.  
      2. Majority basically decided that 24 was not available because it was paired with 23 and they weren't going to change that.  
    7. So the next choice was D9.  
So, basically, the majority had only ONE choice - pairing 22 with 9.  So, no, the alternatives were not clearly defined.  

  1. [This loose 1. is here because I haven't figured out how to do lists that I can break for a moment and then continue on Blogger.  I even tried to make them white so you can't see them, but they are independent.  There will be more at 3.  If anyone has a suggestion I'm listening.]
  2. the right information was not collected, 
    1. The only information that the Board collected in any sort of organized way in the whole process was related to maps.  They used the census data and the Autobound software organized that data for the Board
    2. The Board collected anecdotes, personal preferences, justifications, but did not pursue collecting data that would help verify which of these vague notions about the Senate seats was accurate
    3. The Board was given more anecdotes, personal preferences, and justifications via the public testimony (I've offered a methodology for evaluating that here)
    4. The Board received some actual data and information via the public testimony 
    5. The majority did not study the data and information that came in.  Rather, they picked things that supported their preference and ignored data that didn't (example:  JBER students going to ER High School.  They cited Lance Pruitt's assertion that ER High School wouldn't exist without JBER.  They ignored Denny Wells statistics that showed more JBER residents live in areas of the bases zoned for West High and Bartlett.  And they never compared how many D9 students went to high schools in D22 or vice versa.)
    6. This was a giant gap for the Board.  While they hired technicians for the technical mapping data gathering and organization, the only other professional decision making expertise used for evaluating how well they met the non-numerical criteria for redistricting was to hire the VRA expert. Most of that debate was hidden from the public and I'm guessing led Marcum to pair D22 with D20 because the VRA expert said pairing D22 with D21 (north Muldoon) wouldn't work because of the diversity of that district.  That would have been the majority's ideal pairing because it would have forced popular Democratic Sen Bill Wielechowsk into an Eagle River district.  But that's speculation on my part.  What's key is that they did not remotely follow the three steps of decision making outlined in the excerpt above.


  1. the costs and benefits were not accurately weighed
    1. The majority never compared the two basic options which were:
      1. Pairing  22 and 24  versus pairing 22 and 9
    2. Instead, they
      1. took pairing 22 and 24 off the table from the beginning and
      2. thus made it impossible, in their minds, to pair 22 and 24
      3. used anecdotal information to argue the benefits of pairing 22 and 9
      4. used anecdotal information to argue why 23 could not be separated from 24
      5. used anecdotal information to argue why 23 could not be combined with 17 
      6. only looked at data that favored what they wanted and disfavored what they didn't want
    3. I say anecdotal because there were reliable or valid numbers available to evaluate,  that could be put into tables that neatly outlined factors that would help them compare how well each pairing met the constitutional criteria. (Actually the only criterion for Senate seats is contiguity, but that didn't stop them from talking about socio-economic integration (9SEI)when they thought it would help their cause.  They pointed out the primacy of contiguity when the proponents of Option 2 discussed SEI.
    4. They never gathered objective numbers to fill into the non-existent cost/benefit chart (in this case perhaps advantages/disadvantages chart)

This was not professional decision making.  This was just marshaling claims and assertions to back up a decision that clearly had already been made:  To keep Senate seat L (D23 and D24) intact.  

Why all this effort to pretend that pairing D22 with D9 was the only option they had?  And was far superior to D22 and D24?    And what was the problem with pairing D23 with D17 ( downtown)? In Marcum's words, "Choosing option 2 is an intentional attempt to break up that natural pairing [23/24].  JBER should be with Chugiak." 

But these aren't dumb people.  
The only explanation that makes sense to me is one that the Court already found them guilty of:  political gerrymandering.  The proclamation plan would have given them (and Eagle River) control of two Senate seats.  The Court specifically broke up one of those (D21/D22).  Pairing D22 with D21 gave Eagle River control over a Senate seat that was half a swing district. Pairing D22 with D24 would force them to give up control over another swing district. (D23)  Eagle River would end up with just one solidly red Senate Seat.  So they had to pair D22, not with the obvious match Eagle River seat D24 - a perfect Senate seat by all the normal criteria - but with D9, across the mountains and with no adjoining neighborhoods.  And that would force pulling D9 apart from D10, another reasonable community of interest.  

And that's what they did. Since this doesn't make logical sense from a redistricting perspective and they had to manipulate data to pretend that it did, one has to ask why?   

Is it possible there is another explanation?    Given the attempted gerrymandering the first time around and the fact that this time the three Republican Board members teamed up to support a map made by Randy Ruedrich and voted for that map over the strenuous objections of the other two Board members, it's pretty compelling.  



Addendum:  Based on my notes from the Wednesday, April 13, 2022 Board meeting.

Board Member Bethany Marcum's Reasons for supporting Option 3B

[Board member's comments in black, mine in red]


Marcum:  I’m very uncomfortable with Option 2 because it moves JBER and links it with D17.  It makes the least sense for any possible pairings.  Downtown is the arts and tourism center, that's not what makes up JBER.  It is used to wake up?? the military community.  Choosing option 2 is an intentional intent to break up that natural pairing.  JBER should be with Chugiak.  


It was hard to sit through these comments without putting my hands over my face.  She’s just pulling words and ideas out of the blue here.   Let’s look carefully at the majority’s defense of 3B strategy.  

The first step is to assume the Chugiak/Eagle River(24) and JBER/Government Hill (D23) Senate district is untouchable.  That’s because this is the last district left for them to pull an extra Republican Senator from.  In the same way that the courts found linking (ER) D22 and south Muldoon (21) was political gerrymandering because it ‘cracks’ D21, this pairing ‘cracks’ D23.  


How?  Eagle River is solidly White, comfortably middle class, and Republican, Trump Republican.  It’s the voters who elected Lora Reinbold  and Jamie Allard.    D23 is, in general, lower income, and far more diverse.


Both Simpson and Marcum refer to D23 as the JBER district although 1//3 of the population does not live on JBER.  Other factors:  many military either don’t voter or they vote in their home states.  The JBER precincts had fewer people voting in the last few elections (including 2020) than the non-JBER districts even though they outnumber them 2-1.   So this is a perfect district to pair with a strongly Republican district.  Higher income white neighborhoods have a higher voting percentage than lower income diverse neighborhoods, so Eagle River will dominate Senate races.  As has been pointed out at different times in this process, when a Muldoon house district was paired with ER in the 2011 redistricting process, a popular Senator, and the only black in the State Senate at that time, was handily defeated by ER voters.  


It would appear the Republican majority of the Board is working hard to keep this district for themselves.


So, basically, they start out by putting it to the side, not even to be considered, as they look for a partner house district for a D22 Senate pairing.


But D23 is NOT just JBER.  It’s 1/3  off base, Downtown-adjacent neighborhoods.  


And if JBER and downtown are so different they can’t be paired, why did the Board  make a house district that does just that:  puts downtown in with JBER?  You can’t have it both ways.  It was ok as a house district, but not as a Senate District?


And none of the majority have considered the reverse problem of lumping Government Hill folks in with Eagle River.  If it’s bad one way, then it should be bad the other way.  But that doesn’t help their case.


I’ve been listening to the Redistricting Board debates since December 2020.  I don’t recall people talking about Downtown as the arts and tourism center of Anchorage before and I’m not sure why she thinks the military have no interest in the arts.  They don't go to the museum? They don't go to the Performing Arts Center?  She didn’t mention that downtown is probably the bar/tavern center of Anchorage too because she knows it would be harder to keep a straight face saying the young soldiers don’t spend time in the downtown bars.  


While Marcum has taken umbrage when people have characterized her intent, she has no problem jumping to the conclusion that pairing D23 with downtown is an intentional attempt to break up “the natural pairing” of JBER and Chugiak/Eagle River.  See, it’s just such a natural pairing that it’s off the table when we make these adjustments.  And I suspect that Marcum has convinced herself of this.  The problem is that pairing D22 with D24 is the most natural pairing to be had.  It’s not just me saying that.  Lots of people did, including Dr. Chase Hensel and Dr. Phyllis Morrow.  Dr. Hensel was the expert witness for the East Anchorage plaintiffs.  The two submitted about six pages of testimony about why the two ER districts were a community of interest, the crux is this:


“Because a large data set informs the question of whether and to what degree a population constitutes a community of interest, it can be a judgment call as to where the boundaries of a community of interest lie. In the Eagle River case, however, there is no question: all the signposts point in the same direction.”    


You can read it all in the public testimony listed for April 7-April 8 on pages 327 - 332.  


In that same file there is testimony (pp. 312-316)  from Doug Robbins who offers a long list of references to “Chugiak Eagle River” by the Municipality, by businesses, by Eagle River organizations, to make the point that we all know that Chugiak Eagle River is the most natural pairing.  




Marcum:  Looking back at E Anchorage lawsuit.  Challenged K and L.  L - 23/24, not found to be invalid.  Both are proposals we are not considering.  Both addressed K issue the same way.  It’s what the E Anchorage plaintiffs wanted, satisfactory.  Those individuals still very involved pushing one plan over the other.  Why are they investing themselves in this?  I have to conclude there is political motive.  


I’d have to go back and review the East Anchorage plaintiffs’ suit before commenting on what it did or didn’t do.  

 I have learned over the years that people often project their own thinking and actions on to others.  The Supreme Court found the majority Board members had politically gerrymandered the map.  And so with no proof offered other than they are proposing Option 2, Marcum concludes it’s political.  I’m guessing that she knows that her preference against the common sense pairings, is for political reasons, then opposing her choice must also be for political reasons.  There are other reasons to do things, like fairness and equity in voting.


I didn’t gerrymander.  Here’s why I support pairings.


Courts ruling on Senate K - key response created Muldoon Road district.  


When you put 20 and 21 together, you are left with 22 empty.  The only pairing is D9.  


This is exactly what Simpson said.  The other part of Eagle River (D24), what is known by so many organizations and people as Greater Chugiak and Eagle River, is not an obvious pairing?  Only if you’re locking D24 with D23.  The rational way to proceed is to look at the two adjacent districts - D24 and D9.  When you do that and measure the pairings by every criteria (other than political advantage) then the obvious choice is D24 and D22.  And since it is so obvious, Marcum and Simpson have simply taken that option off the table.  D24 and D23 is a done deal, end of discussion.  So all that is left is D9.  



That leaves 10 without a partner.  ???  That leaves 14 stranded and requires a new pairing.  Take two primary midtown roads.  Four remaining districts 23/24   17/18 same as current    11/12  no changes.  15/16   four changes that result from responding to court ruling and four that remain the same.  



In comparison, Simpson had a set of points and arguments that he went through.  Like Marcum, pairing D22 with D24 was never an option.  So there were never any comparisons of that pairing with the D22/D9 pairing.  Because everything they said about how compatible D22 and D9 are - non-urban neighborhoods, mountainous, wild life encounters (as though those don’t also happen in the Anchorage bowl), fire and snow, service areas for roads - all that is true and a much better match with D24 than D22, as former ER representative/senator Randy Phillips testified.


So let’s move on to Chair John Binkley’s reasons.



Binkley:  Thank you Bethany.  My position.  Then the second round.  Outpouring of public testimony.  Shows me Alaskans are engaged.  Seven different public hearings.  Heard from 100 Anchorage residents.  Heartening that Alaskans care.  People are supportive or opposing one of the other. We step back it’s our job to replace Sen K  - concerned with our pairing of 22 and 21.  Heartening that both proposals repaired that.  It is noteworthy that is how we solved the problem.   


No dispute with them on the process. 


Heard from people that  D22 and D24 should be paired.  Those people are articulate about how ER Chugiak, Peters Creek are closely tied together.  But as Budd pointed out earlier, the two Republican Senators and former Sen President all testified they should be together.  There are factions in the Republican Party who think they should be paired.  Budd mentioned another member of the administration.  Also supported 22 and 24 together.    I take it seriously and those are legitimate moves by people


Let’s give him points for even talking about the D22 and D24 option.  He did not rule it out from the get go.  He likes to ramble a bit and seems to have gotten off the prescribed talking points.


But also heard similarities between 22 and 9.  Both more rural, larger lot sizes.  Single family homes.  Served by road service districts.  Share the Chugach state park.  Close to mountains, deal with wildlife, wildfire danger.  Could be important.  Most house districts are compact and larger districts large rural districts on the outskirts.  Also heard that ER and Hillside were once in a single house districts - met a higher standard.  


The judge said that the Board members’ simple preference doesn’t outweigh the preponderance of public opinion.  Binkley has given us NO hard data to support his feelings about this.  There is no serious comparison of the two possible Senate pairings.  Just words with little or no factual support.  And in the past Eagle River didn't have a big enough population to form two house seats so the population was shared with other house seats.  


I'd also note that nine of the 2021 map districts border Chugach State Park either directly or border other park land that spills into the park.  And no district in town is more than 15 minutes from Chugach State Park.  


JBER in District 23 one of the most compelling.  Extends from D23 into D24.  If underlying house districts different it could have been done differently.  


The only one who recognizes something about how D23 was drawn.  But he doesn’t explain what he means.  Perhaps if they hadn’t made the district 1/3 off base, downtown adjacent?  He also doesn't mention that Government Hill in D23 extends from D23 to D17 (downtown).


Really active and retired military reside in 24 and have that connection to 23.   Direct highway connection Arctic Valley and closer to town.  Also heard testimony to JBER and N Muldoon.  Also legitimate.  Not an option presented to us.   


He’s saying here, I think, that he heard compelling testimony that pairing the JBER with N. Muldoon was another option, but it was “not an option presented to us.”   Excuse me?!  You are chair of the Board.  It’s the Board’s job to create the maps.  Why didn’t you pursue that option yourself if you thought it viable?  Are you saying the Board’s hands are tied? Unless someone else offers an option they can’t entertain it?  


But, of course, the East Anchorage plaintiffs’ preference was to pair the two Muldoon districts, so they didn't  offer it to you.  And the other option was a map which both member Marcum and Randy Ruedrich prepared independently (according to Marcum.)  But if you start with the assumption that  the D24 and D23  pairing is untouchable,  then it wouldn’t be hard to come up with the rest of the maps exactly the same.  


I don’t find compelling the idea of JBER with downtown Anchorage. For 13 years I’ve had a condo here and  been in that district for work and with Alaska RR and in my experience the downtown area part of D23 is dominated by professional services.  


Hearing this, member Borromeo jumped in and said something like, “You deferred Anchorage to Marcum because you said you didn’t know Anchorage well, and only now you are telling us you had a condo in Anchorage for 13 years?


Just moments before Marcum said that Anchorage is the arts and tourism center of Anchorage.  Now it’s portrayed as dominated by professional services.  But no data, no statistics to back up either claim.  Just feelings, personal experience in the past.  I’m not saying that there aren't art galleries and a museum downtown or that there aren’t professional services.  But why does this make the district incompatible with the Base?  There actually wasn't much testimony from the base.  The only person I recall is retired Air Force doctor Felisa Wilson and she said people on the base use their nearest gates and the base should be connected to the part of Anchorage nearest their gate.


Military is a community of interest


Did everyone notice how all three pointed out the commonalities between the pairings they favored and the lack of commonalities between the pairings they opposed?  And how they are long on rhetoric and short on documented evidence?   


I’d point out one more thing.  They’ve spent a lot of time saying that every part of Anchorage is socio-economically integrated to dismiss such comments for house districts and senate pairings they opposed.  But now that’s pretty much their whole argument.  Even though Socio-Economic Integration is a criterion for House districts, but not Senate districts.  Though the related concept of community of interest is considered in Senate districts.


I believe we have two good options.  I’m more comfortable with Option 3B.  I plan to support.


"I feel more comfortable with", not, ‘the preponderance of evidence clearly shows 3B is the plan to support.”  Because the preponderance of evidence goes the other way.  And they did their best to not have any sort of direct, professional comparison of the two key options:  22/24 ER and 22/9 ER/Hillside.


I’d make one final point:  Although the Board members cited the extensive public testimony, there was no serious analysis of that testimony.  I offered a methodology for that there.


Nor did the Board do any serious evaluation of their options.  The Option 2 folks seemed to have gathered more specific reasons.  The Option 3B people, even Simpson’s organized list of reason - had sound bytes and talking points that were based on personal preference rather than any hard data that compared the potential Senate pairings.  And as I said, early on, their strategy was to

Claim Senate District L (23 and 24) as a done deal, the ideal pairing that shouldn’t be touched.

They claimed D23 as the JBER district, even though 1/3 of the district lives off base around downtown and other north Anchorage neighborhoods.  


So their only option left is pairing 22 with 9.  




Me again in black.  There are just so many details that could be added in.  As it is, I'm trying the patience of all but the most obsessed redistricting folks.  The mass of data makes it hard for people who haven't followed closely to see the forest for the trees.  


[UPDATE April 18, 2022, 6pm:  Someone did text me some suggestions for getting into the html code to make my own fix for the numbered list.  Thanks!   Also someone sent to me:  



"Must Read Alaska" is written by a former Republican Party Communications Director in Alaska and runs a sensationalist right wing blog on Alaska politics even though now she apparently lives in Florida.  While Board member Simpson said the fact that some Republican current and former Senators opposed Option 3B proved 3B wasn't partisan, Ms. Downing's headline here seems to acknowledge the obvious.]



No comments:

Post a Comment

Comments will be reviewed, not for content (except ads), but for style. Comments with personal insults, rambling tirades, and significant repetition will be deleted. Ads disguised as comments, unless closely related to the post and of value to readers (my call) will be deleted. Click here to learn to put links in your comment.