*Official Election 2005 Results Thread*

[jimrtex]

Any info on Virginia an Jersey legislaures?

House of Delegates:

R 58 (-3), D 39 (+2), I 3 (+1)

GOP Gains:
6 - SW along WV border west of Radford.  Democrat incombent had very narrow lead in 2003.
99 East, Lower Potomac - Cheasapeake.  About 30% gain in seat where Democrat incumbent did not defend, so assume rural seat with longtime conservative Dem retiring.

Dem Gains:
32 N, NE Loudon.  In 2003, 52% GOP with independent 12%, 47% in 2005.
41 N, SC Fairfax.  In 2003, this was GOP vs 2 independents.  It appears that losing independent in 2003 was GOP candidate in 2005.
67 N, E Loudon, W Fairfax.   GOP uncontested in in 2003.  D 56%, R 41% (not incumbent) in 2005.
87 SE, N Norfolk city.  GOP uncontested in 2003.  2005: D 50, R 38 (non-incumbent), I 12.

Independent Gains:
68 Central.  W Richmond city, N Chesterfield, roughly equal parts in and out of Richmond, with 2400 vote margins cancelling out to produce 44 vote margin out of 27,000 margin.  Republican incumbent was unopposed in 2003.

The other two Independent seats are in rural C Virginia, one between Charlottesville and Lynchburg and areas to the east; the other between Lynchburg and Roanoke (probably includes suburbs of both).  They were not contested in 2005;  In 2003 both candidates drew a Democrat opponent.

[jimrtex]

I thought 87 had a Dem incumbent, Paula Miller? I'm pretty sure that's the case.

You're right.  I was going through the results and looking at races that seemed competitive (30-70%) and comparing with the 2003 results.

Miller narrowly won a special election in December 2004.  In 2003, it was an uncontested GOP seat.  This year, the two candidates from 2004 ran again, but there was also an independent.  Miller won 50-38-12.

[jimrtex]

which may be ultimately how the Republicans gained control of the State Legislature in '02, right?

In the 2001 Texas House of Representatives, Republicans represented a majority of Texans, yet did not have a majority of the seats.  This was because the districts were based on 11-year old census data, and did not reflect population growth in Republican-leaning areas.  In addition, the 1990 redistricting had undersized Democrat seats and oversized Republican seats in Harris County.   Simply drawing equal population districts resulted in a Republican majority in 2002.  Imagine a district with twice the population of the statewide average that is 65% Republican.   Divide it into two districts, and both will elect Republicans.  Or imagine 3 districts with the population of two average districts that are 65% Democrat.  Redraw them into 2 average sized districts, and one Democrat will have lost his seat.  

In other areas. redistricting inevitably results in drastic changes in boundaries, so that the power of incumbency is reduced.  You're not the incumbent if you didn't represent most of the voters in your new district.  Some incumbents retire, creating open seats.  Most such seats are going to choose a Republican.

It was inevitable after the 2001 House redistricting that Republicans would have majority control.

Republicans already controlled the Texas Senate in 2001.  The 2001 senate districts mostly just unkinked the boundaries that the Democrats had drawns and redrawn and redrawn and redrawn (different Senate boundaries were used in 1990, 1992, 1994, 1996, and 1998).  The Republicans picked up another senate seat in 2002.

You must be logged in to read this quote.

These allegations are not based in fact.

[jimrtex]

I've captured the current result by locality (county and independent city) so we can see where the changes are occuring.

On Wednesday there was a regular precinct out for Richmond city and James City county.  They are now included, but with a double entry (one still showing 0 votes, and one with the actual vote totals).  These two together were a +159 for Deeds.

[jimrtex]

But with Deeds ahead, now, but 159 votes.  (I did get it right that one of the precincts needing to be counted was in the city of Richmond, I still know much of my Virginia geography, don't I?  Just a question for myself, sometimes I amaze myself, I don't know why, I just do).   Wondering about the provisional ballotts though?   And of course now it will be McDonnell who will probably call for a recount, and we will have to see if Deeds lead holds.  Can someone remind me when the Attorney General, as well the Governor and Lt. Governor, on what date the AG is inaugurated in Virginia?  I'm wondering if there will be an AG inaugurated on this date?   Again shades of the Washington state Governor elections from 2004.

Kaine asks Kilgore to stay on as AG until the election is settled?

[jimrtex]

I've captured the current result by locality (county and independent city) so we can see where the changes are occuring.

Correction.  Margin should be 619 rather than 579

Between 9:42 PM Thursday and 4:32 PM Friday the updates were:

Alexandria city
+1 Write In

Richmond city
+187 McDonnell
+472 Deeds
+4 Write-in

Margin is now 619

I don't know where the votes were from.  All precincts are shown as reporting, and Absentee votes are listed as two separate precincts.  Provisionals?

[jimrtex]

Margin is now 579

Who to?

McDonnell over Deeds (but see correction and additional posts)

[jimrtex]

Between 9:42 PM Thursday and 4:32 PM Friday the updates were:

Margin is now 619

Between 4:32 PM and 6:52 PM Friday (last report on Virginia State Board of lections site for Friday) an additional 21 votes were received from Richmond city.

McDonnell  +11
Deeds       +10

McDonnell over Deeds by 620 (0.032%)

[jimrtex]

Update to Saturday  9:17 AM

Margin is now 513

139 net additional votes reported from Norfolk city.

McDonnell  +19
Deeds      +126
Write-in        -6

McDonnell over Deeds by 513 (0.026%)

The State Board of Elections certifies the state results on November 28, after which a candidate has 10 days to request a recount (Deeds has already indicated that he will seek a recount).

It appears that most larger jurisdictions use some sort of DRE (direct recording electronic) so that there aren't ballots to be recounted.   Notable exceptions are Virginia Beach city which uses a punch card system that allows voter screening at the polling place for overvotes/undervotes, and Loudon county which uses mark sense ballots.

Voting Systems in Virginia

[jimrtex]

Update to Saturday  6:57 PM

Margin is now 410

Another 178 votes reported from Norfolk city.

McDonnell  +33
Deeds      +136
Write-in        +9

McDonnell over Deeds by 410 (0.021%)

The two bunches from Norfolk were 85% Deeds, vs. 61% overall.  I still don't know where the votes are trickling in from.  The Fairfax County web site indicates a total of 48 provisional votes among 262 thousand votes (Fairfax County is by far the largest jurisdiction in the state, with about 5 times the Richmond vote and 6 times the Norfolk vote.  If the Norfolk additions were from provisionals, the provisional rate would be roughly 40 times that of Fairfax County.

[jimrtex]

Results of Commonwealth of Northern Mariana Islands Governor/Lt Governor race:

Fitiali & Villagomez (Covenant)      3497 28.3%
Hofschneider & Apatang (Independent) 3371 27.3%
Babauta & Benavente (Republican)     3228 26.1%
Tenorio & Santos (Democrat)          2256 18.3%

Roughly 1600 absentee ballots to be counted on November 16.

[jimrtex]

McDonnell over Deeds by 410 (0.021%)

The two bunches from Norfolk were 85% Deeds, vs. 61% overall.  I still don't know where the votes are trickling in from.  The Fairfax County web site indicates a total of 48 provisional votes among 262 thousand votes (Fairfax County is by far the largest jurisdiction in the state, with about 5 times the Richmond vote and 6 times the Norfolk vote.  If the Norfolk additions were from provisionals, the provisional rate would be roughly 40 times that of Fairfax County.

So does this mean that there are still a lot of Fairfax County provisional ballots to be counted?

My impression is that the there were a total of 48 provisional votes in Fairfax County and that they have been counted.  Maybe their voter rolls are computerized so if someone goes to wrong polling place most can be resolved quickly, and other places they just take the ballot and let the central office resolve the vote later on.

Fairfax County canvass

You must be logged in to read this quote.

Some of the DREs may keep a record of individual ballots, but I'd be amazed if these were being mistabulated.  I suppose you could dump the individual voters and count them.  Absentee ballots have to be on paper.  The absentee vote appears to be within a few percent of the total vote in the few cities I checked.  With the margin as thin as it is, even a tiny bias towards Democrats in messing up their ballots could be crucial.  A quick sample shows that absentees are around 5%, or 100,000 statewide, presumably with about 50,000  for each candidate.  If Democrats mess up 3% of the time, and Republicans 2% of the time in a manner that the intent can be determined, that is equivalent to the current margin.

[jimrtex]

McDonnell is back up over Deeds, 446 votes.

The current total as of 12:57 pm, today, is:

McDonnell (R)   970,793   49.97%
Deeds (D)        970,347   49.94%
write-ins              1793        .09%

Monday's additions are from Louisa county (midway between Fredericksburg, Charlottesville, and Richmond).

McDonnell     +158
Deeds           +122
Write-ins             0

The 280 added votes represent about 3.6% votes beyond the original vote totals.  If this were repeated statewide, we would see around 70,000 votes trickling in.

The 158:122 is within 1% of the other votes, so perhaps these are late counted absentees.  Many jurisdiction have one or more "Absentee Vote" precinct swhich appear to correspond to congressional districts.  Louisa county does not have the special precinct.

[jimrtex]

A couple of hours later:

R F McDonnell  Republican 970,889 49.96%
  R C Deeds  Democratic 970,532 49.94%
  Write Ins    1,793 0.09%
Total:    1,943,214   

The margin is 357 votes.

The most noteworthy changes are at the bottom, I have bolded the highlights.


Since yesterday:
   McDonnell    +99
   Deeds        +185

   Margin        357 McDonnell (0.018%)



The SBE site has added a precinct (or precincts) called "Conditional Votes" to each locality.  Fairfax County includes the 47 provisional votes that I had noted from the Fairfax County's web site (actually 48 ballots, one person skipped the AG race).

There were updates from 32 localities, many ones and twos that corresponded to the conditional votes (I didn't check mosy to confirm that these were conditional votes).  There were a few more in larger localities:

Norfolk city       12: 4 M, 8 D
Charlottesville    8: 3 M 5 D
Alexandria city  14: 7 M, 8 D
Prince William     7: 4 M, 3 D
Loudoun           11: 7 M, 4 D
Henrico             19: 8 M, 11 D
Fairfax              47: 19 m, 28 D
Chesterfield      14: 7 M, 7 D
Arlington           13: 3 M, 10 D

Overall conditionals broke:

McDonnell     73
Deeds          107


There are many localities that did not (or have not) reported conditional votes.  In the localities that did report them, they were typically on the order of 1 in 3,000 to 5,000 votes, so it is possible that many smaller counties simply didn't have any.  Localities with more than 10,000 total votes that have not reported conditional votes are: Danville, Hampton, Portsmouth, Suffolk, Virginia Beach cities, and York, Washington, Stafford, Rockingham, Roanoke, Pittsylvania, Isle of Wight, Campbell, Botetourt, Bedford, and Augusta counties.  It would be surprising that Virginia Beach with 96,000 votes had no conditional votes, so I would expect some more conditional votes to show up in the next few days, including additional
votes in some localities that have already reported some.

There were two counties that reported other additional votes:

Henrico County (Richmond suburbs)
  +31 McDonnell
  +30 Deeds

This includes 19 conditional votes notes above, so there were
42 additional vote that were:
   +23 McDonnell
   +19 Deeds

Accomack County (Eastern Shore):
  +0 McDonnell
  +56 Deeds

No conditional votes here, and there are only 7,605 votes in the county, which McDonnell carried with 54.5% of the vote.

[jimrtex]

As of 5:07 pm Tuesday

Margin McDonnell +347

Changes:

Botetourt (+2)
    Write-in +2

Fluvanna (-54)
    McDonnell  -32
    Deeds        -22

[jimrtex]

Edit: Left some stuff in from previous message.

As of 4:07 Wednesday

Margin McDonnell +353

Changes:

McDonnell +1 (Fluvanna +1)
Deeds       -5  (Louisa -2; Rockbridge -3)
Write-In    +1 (Albermarles +1)

[jimrtex]

What's going on in Fluvanna County?

The -54 votes was actually yesterday's episode that didn't get edited out.  Today was just the extra vote for McDonnell.

[jimrtex]

As of 5:27 Wednesday

Margin McDonnell +352

Changes (since 4:07)

All in Tazewell county
McDonnell  0
Deeds       +1
Write-In    +3

[jimrtex]

As of 3:12 Thursday

Margin McDonnell +340

Changes (since Wednesday 5)

McDonnell  -10 (Fauquier -10)
Deeds        +1 (Nottoway +1, Tazewell -1, Radford city +1)
Write-in      +1 (Suffolk city +1)

My understanding is that the State Board of Elections does their canvass on November 28, after which Deeds has 10 days to file for a recount.

The recount is conducted by a recount court consisting of the chief judge of the Circuit Court of the City of Richmond (because it is a statewide election), and two judges appointed by the Chief Justice of Virginia.  It appears that the general intent is that recounts would be conducted in each locality under oversight of the recount court, but if they found it necessary they could get everything trucked down to Richmond. 

Voter eligiblity is not an issue in a recount.  Optical scan and punch card ballots are not hand counted unless they are write-ins, overvotes, undervotes, or write-ins.  There are currently 1800 write-in votes.  Write-in votes are ordinarily not tabulated in Virginia unless over 5% of all votes, or more than apparent winner, so there might be write-in votes for Deeds and McDonnell.

A contest of the election would be conducted by the General Assembly, with a final determination made by a joint session.

[jimrtex]

As of 4:47 Friday

Margin McDonnell +345

Changes (since Thursday 3:23)

McDonnell  +2 (Fauquier +4, Carroll -2)
Deeds        -2  (Carroll -2)
Write-in     +1 (Fauquier +1)

Edited to add 1 Write-in vote.

[jimrtex]

Northern Mariana Islands Governor

Count with off-island absentee votes counted:

Governor&Lt Governor    Party        E-Day Abs  Total Pct
Fitial & Villagomez     Covenant     3497  312  3809 28.1%
Hofschneider & Apatang  Independent  3371  339  3710 27.3%
Babauta & Benevente     Republican   3228  382  3610 26.6%
Tenorio & Santos        Democrat     2256  186  2442 18.0%

The Covenant party appears to be a reform offshoot of the Republican Party.  For example, the campaign site of Benigno Fitial includes among his political activities the Republican Party and Bush for President  in addition to founder of the Covenant Party.  Heinz Hofschneider, the independent gubernatorial candidate is a Republican member of legislature.

[jimrtex]

I always saw Republicans as being in a Covenant. 

I believe the agreement defining the relationship between the US and the CNMI is called a Covenant.

[jimrtex]

Certified Result

Margin McDonnell +323

Changes (since November 18)

Chesapeake city (+89)
   +33 McDonnell
   +56 Deeds

Suffolk city (+5)
   +3 McDonnell
   +2 Deeds

[jimrtex]

Heartbreakingly close. What's the chances of it going the other way?

The article mentioned the previous statewide recount when Doug Wilder was elected governor.  In that case, 113 votes were trimmed off a 7000 vote margin.

It might be possible that because of the closeness of the election that the recount will be more closely contested (eg in the previous recount, even if 50 times as many votes were found, Wilder would have won).

Virginia election law provides that machine-countable ballots be recounted by machine, with only overvotes, undervotes, and write-in ballots visually examined.  There are at least 40,000 undervotes (difference between Governor's race and AG race), but many of those are likely to be deliberate.  There are also 1801 write-in ballots (which have not been counted other than as being write-ins).

Typically, recovered undervotes (ballots where a voter choice is determinable, break pretty much the same way as regular ballots.  But this rule of thumb works because of the typical margins of close elections.  If there is a 1% margin in an election 50.5% to 49.5%, and another 1% of ballots are added that split 40-60, the margin of victory is only cut to 50.4% to 49.6%.  But with the margin under 0.02%, Deeds would only need about 1/10 of 1% all votes cast splitting 60-40 in his favor.  1/10 of 1%, is around 2000 ballots, or about 5% of the undervotes, or slightly more than the total number of write-ins.

[jimrtex]

The Virginia State Board of Elections has added some files to their web site apparently related to the impending AG recount.  I assume these will be treated as recommendations to the recount court, rather than necessarily binding.

Updated Recount Standards

Lots of examples of valid/invalid ballot markings.  One curiousity is that a write-in vote for a candidate on the ballot is not considered valid.

Virginia Recount & Handcount Ballot Examples (Appendix A to Recount Standards)