City of Hudson's weighed voting system under scrutiny

[jimrtex]

If there were a unit count for Tradewinds inside and Tradewinds outside the loop, we could estimate the Tradewinds population outside the loop, and subtract that from the 200 persons in 12-1012.   Then the Ward 4/Ward 5 split of the remainder could be estimated based on number of houses.

Is the 70 units for Tradewinds in the newspaper article gospel?

If so, there are 27 units in 12-1011 (per census), and 43 in 12-1012.   We can project from 59 persons in 12-1011 to get a population outside the loop.  27:43 :: 59:94.

This would give 200 - 94 = 106 persons elsewhere in 12-1011.

The housing split between Wards 4 and 5 is 25.5:12.5, which would put about 70 persons in Ward 4.

[jimrtex]

From Legal Committee, January 25, 2012 report:

"7. Census ramifications and prisoner counts were discussed. Moore spoke with Mayor Hallenbeck who is investigating coordination with the county to share work and expenses. The last prisoner census was taken in 1998. It can be done every 10 years."

This suggests the possibility that Columbia County may have been responsible for calculating the ward populations.   The population for the towns is trivial (LATFOR produced prison-adjusted numbers for all census geography).  But the wards are more complicated.  This would have led to assuming that VTD 3 and VTD 5 were identical to Ward 3 and Ward 5.

The split of the Great Northern block is pretty reasonable.  Someone then goobered the Front Street block.

From Legal Committee, August 22, 2012 report:

"The eighth and final order of business was a discussion of the status of the redistricting the City must accomplish in the wake of the 2010 Census. The Chair reported that there was some question as to the constitutionality of a weighted vote given that one ward (5th) would have greater than 40% of the that weighted vote as a result of the initial calculations. Alderman Pierro registered his discontent with the issue being raised at all, stating that a “weighted vote is a weighted vote.” The Chair responded that, in his view, at some point such a disparity among the wards would result in a vitiating of low-weight ward’s (s’) votes in some cases. The Chair reported that he had had a conversation with President Moore wherein the latter indicated that he’d sent a letter to the State Attorney General’s office seeking its opinion in light of a recent law review article on point. The conversation was tabled until the opinion could be received and reviewed."

The letter from Moore and any response from the AG should be public record.   The letter must specify the particular law review article. 

Is John Friedman related to the Hofstra law student ___ Friedman who was one of the author's of the report?

The February 13, 2013 Legal Committee discussed the Dr. Papa+ report, and decided to make no recommendation  as to weighting.   Was Dr.Papa+ provided the ward populations, or did he calculate them as part of his contract?

Incidentally, the original blog post that started this thread had an 1873 map showing the then four wards.  The link to Harry Howard was from Mill Street, and there was no road west of Underhill Pond, then known a Lake George.   Note use of Columbia Turnpike as the boundary line between Wards 3 and Wards 4.

[Torie]

Boundary lines can only be changed by referendum. The city is "trapped" for the moment vis a vis the next election with the ward lines specified by the charter.

I am going to use the lower figure for the Mill Street area of 28 since we really don't know the vacancy rate of the Mill St. houses. I know one is red tagged and another has been vacant for some time. Most of the old houses are just within the flood plain.

I think I will attach your post which gets to 70 rather than 55 for the split block between the 4th and 5th wards.  The actual numbers are not too critical, since it really does not make much sense to jigger the weighted vote numbers before the next election, at substantial expense. The odds that re-doing them will actually change how a council vote comes out are extremely low, given the current political climate.

The city attorney has been charged with writing a legal opinion on the legality of the weighted vote, and I will be, and other legal players are in the wings. It is all happening now rather rapidly. You see how much trouble you've caused Jim? 

[Torie]

If there were a unit count for Tradewinds inside and Tradewinds outside the loop, we could estimate the Tradewinds population outside the loop, and subtract that from the 200 persons in 12-1012.   Then the Ward 4/Ward 5 split of the remainder could be estimated based on number of houses.

Is the 70 units for Tradewinds in the newspaper article gospel?

If so, there are 27 units in 12-1011 (per census), and 43 in 12-1012.   We can project from 59 persons in 12-1011 to get a population outside the loop.  27:43 :: 59:94.

This would give 200 - 94 = 106 persons elsewhere in 12-1011.

The housing split between Wards 4 and 5 is 25.5:12.5, which would put about 70 persons in Ward 4.

The 4th ward part of 12-1012 has about 27 units, and if the number of Crosswinds units in 12-1012 is 43 units, that totals to 70 units. How many units are in 12-1012 per the census, and could you link to me that data?

Anyway, 27/70 = 38.57%, and 38.57% of 106 = 41 persons in Ward 4's portion of 12-1012.

[jimrtex]

Dr.Papayanopoulos was paid $1500.00 in November 2002.

I would surmise from this that the order of events was:

(1) Spring 2002: City council resolution on ward populations for 2000 Census based on ward boundaries in charter.  The resolution specifies that it was to be used for generating the weights, and also noted the issue with the Firemen's Home.

(2) Summer-Fall 2002, Papayanopoulos develops weights, presumably based on resolution populations.

(3) November 2002, Lee Papayanopoulos is paid.

(4) December 2002 through Spring 2003, equal-population proposal is advanced.

Question: Who drew this map?  It was clearly aware of the problem of splitting census blocks.

(5) Spring 2003, council approves equal-population wards.  Under state law, this is subject to a mandatory referendum.

(6) November referendum fails.

(7) Spring 2004, new weights are implemented.

Presumably, the weights were sitting on the shelf since the 2002 Papayanopoulos report.

[Torie]

As to who knew what when, that is something I am trying to avoid exploring. I think I have an idea based on various chats.   What is important is fixing it all. Look to the future, not trawling the past. That serves no useful purpose, and in the end, none of it was done really with some evil political agenda in mind (I suspect the bisecting of the new location of the Firemens' home in particular freaked some folks out that were involved). This issue will most probably be going to referendum (I already drew what I think is a sensible map), and if that fails, litigation. There is no going back. The genie is out of the bottle, in substantial part to yes, you Jim, you trouble maker you.  

Oh, yes, Crosswinds does indeed have 70 apartments.

[jimrtex]

Boundary lines can only be changed by referendum. The city is "trapped" for the moment vis a vis the next election with the ward lines specified by the charter.

I don't know that this is true.   A referendum is required if there is a structural change, such as changing the size of the council, or doing away with the popular election of the mayor.   Switching to equal-population districts required a change in how council decisions were made, since a majority was being redefined to be 6 of 11 members.

But a referendum was not required to change the voting weights to reflect population change.  In that case, the city was simply reapportioning political power, similar to changing ward boundaries.   In this case a referendum by initiative is permissible.

But, there is a specific requirement than when a city changes ward or other district lines, that are used for electing members to a county board of supervisors, then a referendum is mandatory.   The fact that this case is specifically subject to referendum, implies that other cases are not.   Since Hudson's wards are used for election to the Columbia Board of Supervisors, then Hudson is subject to a mandatory referendum.

A problem with power-based weighting vs. population-based weighting is that changing the relative populations of the Hudson wards, may require readjustment of the voting weights of all the towns as well as the Hudson wards.  This may be particularly true because four of Hudson's wards have fewer persons than every town in Columbia County (Taghkanic has 1313 persons).

There is joint responsibility for conduct of elections in Hudson.  Hudson is responsible for seeing that its charter is conformed with, and the Columbia County Board of Elections is responsible for executing elections in conformance with the city charter.

The City of Hudson is (or should be) aware that the elections have not been conducted according to its charter, and should inform the CBOE of this, as well as giving details.

At the same time, it should recognize the ambiguities in its charter.   A person does not reside a specific point within their residence.  You move about during the day, sometimes even in the yard.  It would be absurd to ask which bedroom someone slept in, or how often they fell asleep before the TV, and whether and when did they move to the bedroom, or whether they cooked their meals, or had it delivered, or went out to eat.  Even if someone is confined to their bed for 22 hours a day in a nursing facility, does not mean you can dictate that they reside at that point.  It would be a violation of equal protection to treat them differently than other persons.  Just because someone may be physically incapable of exercising their liberty, does not empower the state to take it.

Therefore, the City of Hudson, not only may, but must, provide an interpretation of its charter.  In particular, it should determine that all residents of the Firemen's Home reside in Ward 4, and should also provide an interpretation for the two Houses on Clinton that are north of 5th Street, and can not be said to live east or west of it.

The city attorney has been charged with writing a legal opinion on the legality of the weighted vote, and I will be, and other legal players are in the wings. It is all happening now rather rapidly. You see how much trouble you've caused Jim?  

Is the City of Hudson in compliance with this provision of New York's Municipal Home Rule Law (MHR), Section 10.1.a.13.c?

(c.)  As  used  in  this subparagraph the term "population" shall mean
  residents, citizens, or registered voters. For such purposes, no  person
  shall  be deemed to have gained or lost a residence, or to have become a
  resident of a local government,  as  defined  in  subdivision  eight  of
  section  two  of  this  chapter,  by  reason  of  being  subject  to the
  jurisdiction of the department of corrections and community  supervision
  and   present   in  a  state  correctional  facility  pursuant  to  such
  jurisdiction. A population base for such a plan of  apportionment  shall
  utilize  the  latest statistical information obtainable from an official
  enumeration done at the same time for all the  residents,  citizens,  or
  registered  voters of the local government. Such a plan may allocate, by
  extrapolation or any other  rational  method,  such  latest  statistical
  information  to  representation  areas  or  units  of  local government,
  provided that any plan containing such an allocation shall have  annexed
  thereto as an appendix, a detailed explanation of the allocation.

Note that the prisoner-adjustment clause was added in 2011, but this subparagraph existed before this.  Hudson is not exclusively using data from the 2010 Census (including the LATFOR prisoner adjustment).  They are making an extrapolation from that data, and additional information.  But they have not annexed a detailed explanation of the allocation.  If they had, someone might have caught the Front Street mistake, or we could at least explain where they had gone wrong.

Even if power-weighted voting is lawful, has the City of Hudson implemented its current plan in a lawful manner?

'Roxbury Taxpayers Alliance' v. Delaware County Board of Supervisors' is an interesting case.  Delaware County (NY) used power-based voting weights for election to their board of supervisors.  It was determined that only persons whose representation-share was less than their population-share had standing.   This is based on cases for equal-population districts where only persons in overpopulated districts have standing.

In the case of Hudson and its current weights, only residents of Ward 5 and Ward 1 would have standing (ignoring the erroneous population base for now).  In general, particularly for small bodies, it will be the most populous districts whose voting weight share will be reduced below its population share.  Ward 1 is somewhat of a quirk, since it and Ward 4 have the same voting weights.  With a small number of representatives a "best" fit may not be a good fit.

After determining who had standing, the court compared the voting-weight shares to population shares, and determined that they were within safe harbor limits, and dismissed the case.  This was even though the intent of using power-based weighting was to overcome some perceived defect in population-based weighting.   That is, the voting weights were deliberately not proportional to population.  In effect, Delaware deliberately violated equal protection, but the court decided it hadn't been severe enough.

There could conceivably be an issue of whether there is a safe harbor when a more proportional result could be achieved.  In the case of weighted voting, you can get an exact match by simply using the population as the voting weights.

Delaware County has since abandoned power-based weighting, and their voting weights are simply the population divided by 10 (and rounded).

In Hudson, the voting-weight share deviates further from population share.  Ward 1 is 14.13% underrepresented.   It is a quirk that Wards 1 and Wards 4 are underrepresented in their voting share.  I think it is probably related to the fact that Ward 5 and any other ward represent a majority of the population.  The majority for Wards 5+1 or 5+4 is a bare majority, but a majority nonetheless, while Wards 5+2 or 5+3 are close to 60% majorities - there is a lot of waste.
To block Ward 5, all four other wards would have to vote together.

In effect, Ward 1 and Ward 4 are as powerful as Wards 2 and 3 when population-based weighting is used.   So Ward 1 and Ward 4 have their voting weights reduced such that their power is not as great.

[jimrtex]

If there were a unit count for Tradewinds inside and Tradewinds outside the loop, we could estimate the Tradewinds population outside the loop, and subtract that from the 200 persons in 12-1012.   Then the Ward 4/Ward 5 split of the remainder could be estimated based on number of houses.

Is the 70 units for Tradewinds in the newspaper article gospel?

If so, there are 27 units in 12-1011 (per census), and 43 in 12-1012.   We can project from 59 persons in 12-1011 to get a population outside the loop.  27:43 :: 59:94.

This would give 200 - 94 = 106 persons elsewhere in 12-1011.

The housing split between Wards 4 and 5 is 25.5:12.5, which would put about 70 persons in Ward 4.

The 4th ward part of 12-1012 has about 27 units, and if the number of Crosswinds units in 12-1012 is 43 units, that totals to 70 units. How many units are in 12-1012 per the census, and could you link to me that data?

Anyway, 27/70 = 38.57%, and 38.57% of 106 = 41 persons in Ward 4's portion of 12-1012.

89 in 12-1012.   If we subtract the 43 units in Crosswinds, that leaves 46.   I count about 38.   I would think that the most likely to be multi-family would be some in the pseudo-block.   The houses on the south side of Harry Howard are fairly small, and those along Clinton aren't particularly large.

[jimrtex]

As to who knew what when, that is something I am trying to avoid exploring. I think I have an idea based on various chats.   What is important is fixing it all. Look to the future, not trawling the past. That serves no useful purpose, and in the end, none of it was done really with some evil political agenda in mind (I suspect the bisecting of the new location of the Firemens' home in particular freaked some folks out that were involved). This issue will most probably be going to referendum (I already drew what I think is a sensible map), and if that fails, litigation. There is no going back. The genie is out of the bottle, in substantial part to yes, you Jim, you trouble maker you.  

Oh, yes, Crosswinds does indeed have 70 apartments.

I don't suggest malevolence.  There are a number of points of carelessness.

The City of Hudson has failed to provide detailed information about the allocation of the split blocks.   That is an explicit requirement of state law.  If they had provided the details of the allocation of the Front Street block, it might have been caught.  We don't know for sure that is the source of error.  They could have got Front Street correct, but then added some blocks north of Warren to Ward 1.  It is quite possible that the mistake was because of someone unwilling to believe that Ward 1 had so little population.   4 aldermen could be apportioned to Ward 5, and one each to Wards 1 and Ward 4, and Ward 1 would still be overrepresented.

The big error is the Front Street block.  I had at one time figured that they had simply flipped the two wards.  But when I tried to reproduce my reasoning, they ended up being way off.  At that point, I was convinced I was doing something really stupid, but couldn't find it.

I just had a thought.   What if someone had started to create an equal-population map.  A logical change would have been to move the entirety (or most) of the Front Street block into Ward 1.  Ward 2 with a corrected population has more than 1/5 of the population.   Taking the Front Street block avoids directly crossing Warren, reduces the expansion of Ward 1 to the east, and avoids having to expand Ward 4 to the west.  But then they tried to back it out, but forgot the Front Street block.  This would indicate a mild level of culpability since they weren't doing what they had a direct reason for doing.

The other mistakes are not so obvious.   You might recall that we originally told you that the populations of Ward 3 and Ward 5 exactly matched the population of VTD 3 and VTD 5, but that they had to split VTD 1-2-4 to because it represented the composite of three wards.  It was only when I started checking the 2002 numbers, that I realized that previously there had been a quite reasonable allocation of Front Street then, but that VTD 3 and VTD 5 were not coincidental with Wards 3 and Wards 5, and that the 2002 allocation had adhered to the charter.

Anyhow, I'm not seeking an admission of culpability or fault, simply an acknowledgement of fact.  Then the obvious thing to do is simply generate a new set of voting weights based on the actual populations of the wards.   This does not require a referendum, and could be in place for the 2015 city elections.

Certainly correcting the voting rolls for 2015 is feasible.

That leaves the Firemen's Home and the two houses north of 5th Street on Clinton.  You can not reduce a place of residency to a point.  Are you going to start sawing people in half Solomon-Torie?  It is quite rational to interpret a building that is divided by a line to be one side or the other of that line for electoral purposes.

This has no effect on whether or not Hudson switches to equal-population wards, or whether power-based weighted voting is unconstitutional in general or in the particular case of Hudson.

[Torie]

If there were a unit count for Tradewinds inside and Tradewinds outside the loop, we could estimate the Tradewinds population outside the loop, and subtract that from the 200 persons in 12-1012.   Then the Ward 4/Ward 5 split of the remainder could be estimated based on number of houses.

Is the 70 units for Tradewinds in the newspaper article gospel?

If so, there are 27 units in 12-1011 (per census), and 43 in 12-1012.   We can project from 59 persons in 12-1011 to get a population outside the loop.  27:43 :: 59:94.

This would give 200 - 94 = 106 persons elsewhere in 12-1011.

The housing split between Wards 4 and 5 is 25.5:12.5, which would put about 70 persons in Ward 4.

The 4th ward part of 12-1012 has about 27 units, and if the number of Crosswinds units in 12-1012 is 43 units, that totals to 70 units. How many units are in 12-1012 per the census, and could you link to me that data?

Anyway, 27/70 = 38.57%, and 38.57% of 106 = 41 persons in Ward 4's portion of 12-1012.

89 in 12-1012.   If we subtract the 43 units in Crosswinds, that leaves 46.   I count about 38.   I would think that the most likely to be multi-family would be some in the pseudo-block.   The houses on the south side of Harry Howard are fairly small, and those along Clinton aren't particularly large.

46/89 = 51.69%. 51.69% x 106 = 54.78.  So is that the number we are back to?

[Torie]

Oh, as the Firemens' Home, sure a couple of bedrooms might be split. Flip a coin.

I understand that the weights can be fixed as best they can (probably still not legal for complex issues but yes, as best they can, hiring a consultant at some expense to do it all). But that just affects Council Vote weights, not voters voting. So unless the wrong weights affect a Council vote outcome, they are irrelevant. I doubt that they will ever be relevant in the next year given the local politics. Sure if the referendum fails (yes, not necessary for changing the weights, but apparently necessary for changing the lines, and therein is the rub), the weights will need to be corrected ASAP. If not, the system will become even more legally vulnerable, and the city would be risking that the court will not bother with just ordering a narrow fix, and not interested given the attitude in struggling to find the system, and the assumption that aldermen from the same ward vote randomly vis a vis each other is tolerable, or otherwise not caring about the Banzhaf issue, considering it legally dead (it isn't, it just isn't a panacea anymore); rather the court would just throw up its hands,  just toss the whole thing out. So it may become potentially relevant, but not now in my view. The city can't afford to spend money on this, and should not.

[jimrtex]

If there were a unit count for Tradewinds inside and Tradewinds outside the loop, we could estimate the Tradewinds population outside the loop, and subtract that from the 200 persons in 12-1012.   Then the Ward 4/Ward 5 split of the remainder could be estimated based on number of houses.

Is the 70 units for Tradewinds in the newspaper article gospel?

If so, there are 27 units in 12-1011 (per census), and 43 in 12-1012.   We can project from 59 persons in 12-1011 to get a population outside the loop.  27:43 :: 59:94.

This would give 200 - 94 = 106 persons elsewhere in 12-1011.

The housing split between Wards 4 and 5 is 25.5:12.5, which would put about 70 persons in Ward 4.

The 4th ward part of 12-1012 has about 27 units, and if the number of Crosswinds units in 12-1012 is 43 units, that totals to 70 units. How many units are in 12-1012 per the census, and could you link to me that data?

Anyway, 27/70 = 38.57%, and 38.57% of 106 = 41 persons in Ward 4's portion of 12-1012.

89 in 12-1012.   If we subtract the 43 units in Crosswinds, that leaves 46.   I count about 38.   I would think that the most likely to be multi-family would be some in the pseudo-block.   The houses on the south side of Harry Howard are fairly small, and those along Clinton aren't particularly large.

46/89 = 51.69%. 51.69% x 106 = 54.78.  So is that the number we are back to?

46 is the number of housing units excluding Crosswinds.
89 is the number of housing units total.
106 is population excluding Crosswinds (based on the same population per household as 12-1011.
200 is total population.

46/89 x 200 would give a population estimate outside Crosswinds is 103.   94 + 103 is 197 so 3 persons disappeared.   The reason is that the persons per housing unit for Crosswinds (at least the inner part) is 2.19, while the persons per housing unit for 12-1012 is 2.24, even though about half the housing units are in Crosswinds.   Assuming the two parts of Crosswinds have the same household size, then the household size for the area outside Crosswinds is 2.30.

To allocate the non-Tradewinds part of 12-1011, you really need the number of housing units in both parts, and they must total 46.   You can't just count those in Ward 4.

I can find 42:

2 on 6th Street north of Clinton (Ward 5)
15 on north side of Clinton.
  12 to east of 5th Street (Ward 5)
  2 north of end of 5th Street (Ward 4 or Ward 5?)
  1 to west of 5th Street (Ward 4)
5 west side of 5th Street (Ward 4)
0 on north side of Washington.
4 on east side of Short/Harry Howard (Ward 4).
Underhill Pond
16 on south side of Harry Howard (including two before turning corner) (Ward 4)

Where are the other 4 housing units?   The second house/building north of Washington on Harry Howard is extremely long, and has a huge parking lot.  If it were in Texas it could be a Baptist or independent church.

A couple of houses on 5th Street are huge, and could easily be subdivided.  A problem in the NE is that double deckers can be split vertically, so that they don't look like duplexes.  It is possible that there are garage houses, with some detached garages set way back.  Do you have access to voter rolls organized by street address? 

46 Housing units: 106 population
14 HU: 32.26 (Ward 5)
2 HU: 4.60 (Ward 4 or Ward 5)
26 HU: 59.91 (Ward 5)
4 HU: 9.22 (unkknown).

Ward 5 (minimum) 32.
Ward 5 (maximum, both 5th street houses and 4 missing) 46
Ward 4 (minimum) 60
Ward 4 (maximum) 74.

This can be improved on if you can find the 4 missing HU, or Solomon-Torie adjudicates the location of the two houses on the end of 5th Street at Clinton.

A reasonable argument could be made that we should not do specific allocations such as when we assumed that the two sections of Crosswinds are comparable.  Because that opens up questions of housing size, and whether there are childrens toys, etc.  A simple method based on housing units in the block may be the best (with the exception of when there is an identifiable group quarters population).

Of course we are only demonstrating the approximate magnitude of error.  I would expect the City of Hudson to make their own allocations, and document their methodology.

[jimrtex]

Oh, as the Firemens' Home, sure a couple of bedrooms might be split. Flip a coin.

Arbitrary and capricious and violates right to privacy.   Maybe Hudson could force the building to be demolished because it interferes with their mathematically perfect line.

I understand that the weights can be fixed as best they can (probably still not legal for complex issues but yes, as best they can, hiring a consultant at some expense to do it all). But that just affects Council Vote weights, not voters voting.

An equal protection claim is based on a voter having an equal opportunity to affect legislation.

If you had a 5 equal-population districts, but then discovered that one of them actually had 25% of the population, it is an equal protection violation, even if redrawing the districts would not change any outcome.   A court might defer a remedy, if there was not enough time to make the change, or other possible consequences such as voter confusion.

If you have a weighted voting scheme, and one ward had its voting weight set based on the assumption that it had 20% of the population, but then discovered that it actually had 25% of the population, it is an equal protection violation, even if correcting the weights would not change any outcome.   But in this case, there is less reason to defer any remedy.  No voter is going to be confused by by the change in the weight, and it can even be done between elections.

You must be logged in to read this quote.

Judge Jim grants summary judgement to the plaintiffs.   Given that the voters of Ward 2 are the most harmed, we even have a VRA violation.  Hudson should be lassoed in under Section 3.

You must be logged in to read this quote.

The city can not win an equal protection case; and a VRA claim would be harmful to Hudson's reputation.   They would be wise to be pro-active.

If the erroneous population numbers are the responsibility of Dr.Papa+, he is likely responsible to correct them free of charge.  The cost of calculating new weights is much cheaper than developing a new plan, and holding a referendum on it.

Judge Jim rules that the population error is an equal-protection violation, and there is no reason to address the other claims.  A federal court would have no reason to undermine the wishes of the people when they rejected equal-population districts at a referendum.   So it would order new weights, and order Hudson to pay the cost.

[jimrtex]

If we use your estimated ward populations (in your spreadsheet), but reverse your division of the Firemen's home

Ward 1: 593
Ward 2: 1471
Ward 3: 1076
Ward 4: 787
Ward 5: 2476

And use the following weights:

Ward 1: 95, 95
Ward 2: 211, 211
Ward 3: 155, 155
Ward 4: 113, 113
Ward 5: 356, 356
President 190

We get better results:


Ward            Vote    Swing R.Pop.  R.Pow    Dev.  
Ward 1            91     106   9.26%   9.45%   2.01%
Ward 1            91     106
Ward 2           211     258  22.97%  22.99%   0.09%
Ward 2           211     258
Ward 3           155     182  16.80%  16.22%  -3.47%
Ward 3           155     182
Ward 4           113     134  12.29%  11.94%  -2.83%
Ward 4           113     134
Ward 5           356     442  38.67%  39.39%   1.87%
Ward 5           356     442
President        190     202
Majority 1026


Range of Deviation 5.48%
Maximum Absolute Deviation 3.47%
Standard Deviation 2.30%
Weighted Standard Deviation (RMS) 2.18%

Than with the current weights and erroneous population.


Ward            Vote    Swing R.Pop.  R.Pow    Dev.  
Ward 1            95     128  12.03%  11.51%  -4.28%
Ward 1            95     128
Ward 2           185     228  20.01%  20.50%   2.49%
Ward 2           185     228
Ward 3           180     196  17.84%  17.63%  -1.17%
Ward 3           180     196
Ward 4            95     128  11.32%  11.51%   1.66%
Ward 4            95     128
Ward 5           364     432  38.81%  38.85%   0.10%
Ward 5           364     432
President        190     260
Majority 1015


Range of Deviation 6.77%
Maximum Absolute Deviation 4.28%
Standard Deviation 2.38%
Weighted Standard Deviation (RMS) 2.00%

As an added benefit, the President's share of critical votes is much closer to the 1/11 that he would have under an equal-population, unweighted system.

[Torie]

Yes, if someone files a lawsuit, the weights would have to be changed pending the next election (which will take a lot of time and some money (they won't just accept your numbers), resulting very probably in no change in any council vote outcome. That won't happen. It's all sound and fury signifying nothing. And if the Firemens' Home is not split, and there is a close election, because the nursing home folks (some of them), voted in the wrong ward, then there will be a lawsuit. Yes, if the lines could be changed without referendum, they should be, but alas the Council on its own cannot change them. Again, on this matter, the city is trapped. What needs to be done is get all of this on the ballot for the next election, probably via the petition route. Then everything can be fixed.  Stay tuned.

[Torie]

In the 4th ward's portion of block 12-1012 (that is the block number on the spreadsheet that you sent me), there are 29 units. 29/46 x 103 = 64.93.  29/89 x 200 = 65.17.  65 is the number. We are done. It's been fun.

[jimrtex]

Yes, if someone files a lawsuit, the weights would have to be changed pending the next election (which will take a lot of time and some money (they won't just accept your numbers), resulting very probably in no change in any council vote outcome. That won't happen. It's all sound and fury signifying nothing. And if the Firemens' Home is not split, and there is a close election, because the nursing home folks (some of them), voted in the wrong ward, then there will be a lawsuit. Yes, if the lines could be changed without referendum, they should be, but alas the Council on its own cannot change them. Again, on this matter, the city is trapped. What needs to be done is get all of this on the ballot for the next election, probably via the petition route. Then everything can be fixed.  Stay tuned.

The Common Council has an affirmative obligation to follow the city charter, state law, and US Constitution.

The current plan violates the New York Municipal Home Law.

"i.)  The  plan  shall  provide substantially equal weight for all the voters of that local government in the allocation of  representation  in  the local legislative body."

"(c.)  As  used  in  this subparagraph the term "population" shall mean
  residents, citizens, or registered voters. A population base for such  a
  plan  of  apportionment shall utilize the latest statistical information
  obtainable from an official enumeration done at the same  time  for  all
  the  residents,  citizens, or registered voters of the local government.
  Such a plan may allocate, by extrapolation or any other rational method,
  such latest statistical information to representation areas or units  of
  local  government, provided that any plan containing such an allocation
  shall have annexed thereto as an appendix, a detailed explanation of the
  allocation."

The Common Council will avoid any lawsuit about the Firemen's Home by declaring that for electoral purposes that it is entirely west of the projected ward boundary.

Do any current registrations for the residents of the Firemen's Home contain a room number?

You have studiously avoided the issue of the two houses on Clinton, that are north of the end of 5th Street.  Or are you still trying to determine which point in these houses that the inhabitants reside in?  Remember, that you have to provide both a current location, and a location on April 1, 2010.   If a resident of a split house were to be elected alderman, would he have to resign whenever he eats in the kitchen.   Are you going to dock his salary whenever he goes to the bathroom in another ward within his house?

Torie's standard is arbitrary and capricious and denies equal protection to the residents of the Firemen's Home and the two houses on Clinton.  It also is in violation of the 26th Amendment.

The City of Hudson may resolve ambiguity in the charter by passing a resolution.   It does not change the ward boundaries by declaring that the Firemen's Home is west of the boundary.

Should I send my spreadsheet with the revised voting weights to you or to the city directly?

[jimrtex]

In the 4th ward's portion of block 12-1012 (that is the block number on the spreadsheet that you sent me), there are 29 units. 29/46 x 103 = 64.93.  29/89 x 200 = 65.17.  65 is the number. We are done. It's been fun.

I count 18 units north of Underhill Pond.

13 units facing north with light colored roofs, including one "2".
1 red roof set back from road.
1 gray with light colored box, with mark in driveway to its west.
1 unit on corner
2 units facing west



For the pseudo block 12? units.

4 units on Short Street, but how do we know that the large structure has 3 units.  The real estate annotation is two.   What is the nature of the structure on the northeast corner of Washington and Short?  Could the Census Bureau have considered it to be a housing unit even if it much of the building is commercial?

6 units on 5th Street.

2 units on Clinton, west of the center line of 5th street (extended).

1 unit on Clinton divided by center line of 5th street (extended)



For completeness the remaining portion of the census block (eastward on Clinton and around the corner on 6th) needs to be inventoried.

[Torie]

The red roof structure isn't a house.

[Torie]

Yes, if someone files a lawsuit, the weights would have to be changed pending the next election (which will take a lot of time and some money (they won't just accept your numbers), resulting very probably in no change in any council vote outcome. That won't happen. It's all sound and fury signifying nothing. And if the Firemens' Home is not split, and there is a close election, because the nursing home folks (some of them), voted in the wrong ward, then there will be a lawsuit. Yes, if the lines could be changed without referendum, they should be, but alas the Council on its own cannot change them. Again, on this matter, the city is trapped. What needs to be done is get all of this on the ballot for the next election, probably via the petition route. Then everything can be fixed.  Stay tuned.

The Common Council has an affirmative obligation to follow the city charter, state law, and US Constitution.

The current plan violates the New York Municipal Home Law.

"i.)  The  plan  shall  provide substantially equal weight for all the voters of that local government in the allocation of  representation  in  the local legislative body."

"(c.)  As  used  in  this subparagraph the term "population" shall mean
  residents, citizens, or registered voters. A population base for such  a
  plan  of  apportionment shall utilize the latest statistical information
  obtainable from an official enumeration done at the same  time  for  all
  the  residents,  citizens, or registered voters of the local government.
  Such a plan may allocate, by extrapolation or any other rational method,
  such latest statistical information to representation areas or units  of
  local  government, provided that any plan containing such an allocation
  shall have annexed thereto as an appendix, a detailed explanation of the
  allocation."

The Common Council will avoid any lawsuit about the Firemen's Home by declaring that for electoral purposes that it is entirely west of the projected ward boundary.

Do any current registrations for the residents of the Firemen's Home contain a room number?

You have studiously avoided the issue of the two houses on Clinton, that are north of the end of 5th Street.  Or are you still trying to determine which point in these houses that the inhabitants reside in?  Remember, that you have to provide both a current location, and a location on April 1, 2010.   If a resident of a split house were to be elected alderman, would he have to resign whenever he eats in the kitchen.   Are you going to dock his salary whenever he goes to the bathroom in another ward within his house?

Torie's standard is arbitrary and capricious and denies equal protection to the residents of the Firemen's Home and the two houses on Clinton.  It also is in violation of the 26th Amendment.

The City of Hudson may resolve ambiguity in the charter by passing a resolution.   It does not change the ward boundaries by declaring that the Firemen's Home is west of the boundary.

Should I send my spreadsheet with the revised voting weights to you or to the city directly?

The two houses for addresses in the 400's on the north side of Clinton St were counted by me.

I don't understand how the Council can waive a wand and declare that all for the Fireman's home is in one ward or the other without changing the boundary itself, which it cannot except by referendum. Sometimes a line will split a home or here, we have a room or two or three perhaps so split. The court will be fine with putting its denizens therein assigned to vote, and be counted, in one zone or the other. The effect is de minimus. Stuff happens.

[jimrtex]

The red roof structure isn't a house.

It is a 2-story garage for 68 Harry Howard?

The three houses before the corner are 44, 46, 48 Harry Howard
50, 52, 56 are skipped, then
56, 58, 60, 62, and the two unit is 64-66, then
68, 70, 72, 74, 76, 78
and then 82-84 which is single family (but looks like it may have been converted at some time in the past).
and then the last unit is 86 Harry Howard

Crosswinds is 100 Harry Howard.

The tax rolls have the Harry Howard houses in the 4th Ward, and Crosswinds and 106 Harry Howard and 120 Harry Howard in the 5th ward.   So at least in this area, the tax rolls are correct.

[jimrtex]

New York state did not participate in delineating VTD's for the 1980 Census (when they were called election districts).  They did participate in 1990, by which time the VTD name had been adopted.

The 1990 Census was the first in which Census Blocks were defined for the entire country, and there would have been a requirement that VTDs be comprised of whole blocks.  The VTDs in Hudson have not changed since 1990, so somewhere maps were exchanged and the VTD boundaries changed.

I noticed on the image of the Columbia Triangle, that the tip census block is the traffic island between Columbia Street, Columbia Turnpike, and Prospect Avenue.  It nonetheless has a population of 1.   It also has a housing unit count of 1.   Perhaps the Census Bureau does not require a physical structure for a housing unit, or it could be shopping cart, tent, or backpack.

This is somewhat different from the block northwest of 2nd and State Street, north of the public housing tower.  There the census bureau had counted 0 persons, but LATFOR added 1.  LATFOR got the last address of prisoners prior to their incarceration from Department of Corrections records. They then used a commercial service to convert addresses to coordinate locations (lat/long), and then to census blocks.

If a prisoner did not have a last address, or it was out of state, or it could not be located, he was simply removed from the census count for the prison location.

[jimrtex]

In the 4th ward's portion of block 12-1012 (that is the block number on the spreadsheet that you sent me), there are 29 units. 29/46 x 103 = 64.93.  29/89 x 200 = 65.17.  65 is the number. We are done. It's been fun.

Based on Tax Records:

6 Harry Howard (1)
4 Harry Howard (3)
2 Harry Howard (on tax rows as "Det Row Bldg")
(Harry Howard starts at Washington).

The following are in 12-4000, south of Washington

84-90 Short (1)
80 Short (1)
442-444 Prospect (1)
77-79 N 5th (Apt)
81-83 N 5th (Apt)
87 N 5th (Apt)
89-93 N 5th (1)

Going north of Washington, so back in 12-1012

101 N 5th (1)
103 N 5th (1)
107-109 N 5th (1)
111-113 N 5th (1)
115-199 N 5th (2)

So we match your picture to here.

496 Clinton (2) This is an extra unit.
498 Clinton (1)

Everything up to here is on the tax rolls as 4th Ward.

500 Clinton (2) This is the house north of 5th.  On the tax rolls it is in the 5th ward.

Based on the principal of not taxation with representation, these persons should vote in the 5th ward.

502 (2), 510(2), 512(1), 516-518(1), 520(1), 522(1), 528(1), 530(1), 534 (1), 548(1).

So 14 units on the 5th ward portion of Clinton.

123 N 6th (1) This is on the corner of 6th and Clinton.
131 N 6th (1)
135 N 6th (1)

Summary for 12-1012

43 Units Crosswinds Apt
17 Units South of Harry Howard, north of Underhill Pond
10 Units in (Clinton)-Harry Howard-Washington-5th Pseudo Block, 4 on HH, 6 on 5th.
3   Units on Clinton west of 5th (on Tax Rolls as being in 4th Ward)
14 Units on Clinton at or east of 5th (on Tax Rolls as being in 5th Ward)
3   Units on 6th Street north of Clinton.

90 Total units (vs census 89).   I say we go with 90 for allocation.

If we don't differentiate the Crosswind Apartments.

Ward 4: 30 units
Ward 5: 60 Units

Total Population 200

Ward 4: 67 persons
Ward 5: 133 persons

If we do differentiate the Crosswind Apartments

Then based on 12-1011 (59 persons in 27 units) we can estimate 94 persons in the 46 units in 12-1012 portion of Crosswinds.

This leaves 106 persons in the remainder of 12-1012.

Ward 4: 30 units.
Ward 5: 17 units.

Population of 106:

Ward 4: 68 persons.
Ward 5: 38 persons (134 with Crosswinds)

So it really doesn't make a difference.

The houses in Columbia Triangle, on the south side of Columbia Street, west side of Paul Avenue, and north side of Columbia Turnpike are all listed in the Tax Rolls as 5th ward.

So all the areas on the east part of town that were shifted relative to the charter, are in the correct wards in the Tax Rolls.

[jimrtex]

Inconsistencies.   If inconsistent with the charter, it is highlighted in red. 

I have not found online voting rolls.  It is possible that voting is inconsistent throughout an area, such as the two houses on Harry Howard which vote in Ward 4, but for all other purposes are considered part of Ward 5.  One clerk might consult the CBOE map, while another uses the tax rolls, or checks where neighbors are placed.

South Front Street:

Charter: Ward 1
VTD: 1-2-4
CBOE map: Ward 1
Voting: Ward 1 (should have S Front St addresses)
2000 weight: Ward 1
2010 weight: Ward 1
Tax roll: Ward 2 (but Hudson Terrace is a single property with a 15 N Front St address)

North Front Street:

Charter: Ward 2
VTD: 1-2-4
CBOE map: Ward 2
Voting: Ward 2 (should have N Front St or State St addresses)
2000 weight: Ward 2
2010 weight Ward 1 and Ward 2 (allocation error)
Tax roll: Ward 2

5th Street Notch (areas west of 5th Street that sometimes are treated as being in Ward 5: including houses South of Harry Howard and north of Underhill Pond; (Clinton)-Harry Howard-Washington-5th pseudo-block; Washington-Short-Prospect-5th block; 496 and 498 Clinton.

Charter: Ward 4
VTD: 5
CBOE map: Ward 5
Voting: Ward 4
2000 weight: Ward 4
2010 weight: Ward 5
Tax roll: Ward 4

Crosswind Apartments 100 Harry Howard

Charter: Ward 5
VTD: 5
CBOE map: Ward 5
Voting: Ward 4
2000 weight: Not built, but Ward 5.
2010 weight: Ward 5
Tax roll: Ward 5

Harry Howard outliers 106 and 120 Harry Howard

Charter: Ward 5
VTD: 5
CBOE map: Ward 5
Voting: Ward 4
2000 weight: Ward 5
2010 weight: Ward 5
Tax roll: Ward 5

Columbia Triangle: Bounded by Columbia Street, Paul Avenue, Columbia Turnpike.

Charter: Ward 5
VTD: 3
CBOE map: Ward 3
Voting: Ward 3
2000 weight: Ward 5
2010 weight: Ward 3
Tax roll: Ward 5

[jimrtex]

Message broken in two to get under size limit.

You must be logged in to read this quote.

I'm not sure I understand your question.

I think you are asking  something related to this:

Let's start with 5 aldermen with equal weights of 20, which total 100 votes, and a majority of 51 is required.  It still requires 3 aldermen to pass anything, and while a majority of 51 is required, any winning combination will actually have 60 votes.

There are only 2⁵ or 32 combinations of 5 members, including 5:0 and 0:5 unanimous votes.   ^5! / _{3! 2!} or 10 combinations will be 3:2 votes.   3 members are critical to each of these 3:2 combinations, making 3 x 10 or 30 vote changes that are critical, with each member the critical change 6 times.   As we would expect, the voting power of each member is 6/30 or 20%.

Now let's change the weights some.   A 29, B 20, C 20, D 20, E 11; Total = 50, majority = 51.   The voting weight for A is now 45% larger than the average, and a combination of (A,B,C) will have 69 votes, while a combination of (C,E,F) will have 51 votes,

Nonetheless, it still requires 3 members to pass any resolution since the strongest pair (A,B) only has 49 votes, and any 3-member combination will succeed since the weakest triplet (C,D,E) has 51 votes.  The voting power of each member is still 6/30 or 20%.  If the voting weights were proportional to population (eg District A has 29% of the population, while District E only has 11%), then there is a large discrepancy between population and voting power.

To actually change the voting dynamics, we need to make some 2-member combinations succeed, or equivalently require four-member combinations fail.   So let's tweak the weights to A 31, B 20, C 20, D 20, and E 9.  Now any two-member combination of A and any of B, C, or D succeeds, and a combination of E and two of (B,C,D) fails.

Suddenly, A is critical to 12 combinations.  It is critical not only to (A,B), (A,C,), and (A,D), but also to any of the 6 3-member combinations it is a member of (If A leaves (A,B,E) it fails), and the 3 four-member combinations that include A,E, and 2 of B,C,D.

B, C, and D are each critical two 4 combinations.  B is critical to (A,B), (A,B,E), (B,C,D), and (B,C,D,E).

E is critical to zero combinations.  There are no combinations, where it would pay off to get E to switch his vote, whether through convincing arguments, cajoling, bribery, or coercion.  E has no voting power.

There are 24 critical changes, A is critical 12 of them or 50%, while B, C or D are each the critical  vote 4 times or 17%.   E is critical 0% of the time.  This clearly is not acceptable.

So let's try (A,B,C,D,E) = (29, 22, 20, 18, 11)

Critical swings are now (8,8,4,4,4) or on a percentage basis: A 29% (on target); B 29% (way overpowered); C and D, 14% each and underpowered; and E 14% also overpowered.

The best I can come up with is (A,B,C,D,E) = (30, 21, 21, 21, 10) .  Critical swings are (9, 5, 5, 5, and 3) for a total of 27,  Power shares are (33%, 19%, 19%, 19%, 11%).   If the populations shares were 29%, 20%, 20%, 20%,and 9% this is not too bad, though the relative error for A is 15%, and for B, C, and D it -7%, and E 1%.  The deviation range is 22% which is pushing what a court might accept.

Problems with weighted voting and small bodies include few voting combinations, and few critical combinations.  In our example, there were only 32 combinations and somewhat fewer critical changes.   With 27 critical swings total, each represents 3.7% power share, which is quite coarse in terms of resolution.   And where we had 3 districts with the same population (and it wouldn't really matter if they had populations of say 1538, 1511, and 1487), many of the combinations are equivalent (A,B), (A,C), and (A,D) for example are equivalent in their population share, number of members, and relative voting strength between the members.  There are only 16 non-equivalent sets of combinations:

Empty set.
(A), (X), (E)
(AX), (AE), (XX), (XE)
(AXX), (AXE) (XXX), (XXE)
(AXXX), (AXXE), (XXXE)
(AXXXE)

Here X represents any one of B, C, or D; with no duplicates within a combination.

I didn't answer your question, but I think you may be able to re-ask it now.