Here is the weekend recap for the gubernatorial vote count -- and the definitive explanation why Dino Rossi has almost certainly won.
The simple answer is that Rossi ended the day with a 1,920 vote lead and it's just not realistic for Gregoire to recover that lead from the 41,623 remaining uncounted ballots. That would require her to take a 4.61% lead in the remaining ballots where only 40% of them are in counties that favored her.
Most of the week I was projecting that Rossi would win by about 4,400 votes. That was based on the assumption that the uncounted ballot would break in the same proportions to the previously counted ballots for their respective counties. Given Gregoire's larger than expected lead in Friday's King County returns, I'll revise my projections to give a range of outcomes based on a range of assumptions. The best case I can come up with for Rossi gives him a lead of 2,480 in the uncounted and the best case for Gregoire gives her a lead of 753, meaning that Rossi's final margin of victory would range from 1,167 to 4,400. More likely, I expect Rossi's final margin to be in the ballpark of 1,700 (if the remaining ballots swing like the most recently counted ballots) to 3,400 (if they average out like all of the previously counted ballots).
I estimate this by looking at the Secretary of State's estimate of the uncounted ballots for each county, and also at the percentage spread between the two major candidates during three different counting periods, which I've been tracking in a spreadsheet --
(1) the days after the election, Nov 2-5, poll ballots plus some absentee
(2) the first four days of the past week, Nov. 8-11, mostly absentee ballots
(3) Friday, Nov. 12, absentees and a larger number of provisionals.
I also note the "Latest" spread for each county. Not every county reported results during all three periods, so I'm most interested in the Latest period during which the county reported.
In addition to the three periods, I calculated the minimum and maximum of these three spreads for each county, and the overall spread for each county for all counted ballots.
The table below shows the uncounted votes in each county and also calculates Gregoire's (Rossi's) lead by multiplying the spread under each scenario by the uncounted votes. Gregoire's leads are in Black, Rossi's in (Red). In the interest of space, I don't show all of the columns. The two most realistic scenarios, I think, are where (a) a county's uncounted votes would reflect all of its previously counted votes [column labelled "Overall"] , or (b) a county's uncounted votes would reflect the most recently counted votes for that county ["Latest"]. I also show the projections based on the first week of counting, and the Min and Max values for each county.
The "Total Uncounted" row shows the lead for the uncounted ballots only, the "Rossi's Victory" shows his final victory margin including the current lead.
Note that nearly all of the difference between the "Latest" and the "Overall" scenarios is explained by two counties -- King and Whatcom. Friday's large number of provisional ballots in King County swung sharply toward Gregoire. We can concede that the remaining ballots in King County may break the same way. Whatcom returned only 140 ballots Friday, but they gave Gregoire a 20% spread, instead of the 2% spread she enjoyed in that county's 85,000 previously counted ballots. It would probably be optimistic for Gregoire to assume that the 2,000 remaining ballots would also give her a 20% spread. Even if we assume that both King and Whatcom's remaining ballots go the same way that Friday's ballots went, Gregoire would still need to pick up a 1,700 lead somehow. Where would these extra votes come from? One possibility is from Thurston County, which didn't report Friday and is home to Evergreen State College, so a potential source of pro-Gregoire provisionals. But enough to give Gregoire an 85% lead in Thurston County's 2,000 remaining ballots, when she's held only an 8% so far? Doesn't seem plausible.
Weirder things could happen, and the Democrats seem determined to sue their way out of their statistical difficulty. Stay tuned.
| County | Uncounted | Nov. 2 - 5 | Nov. 8 - 11 | Nov. 12 | Min | Max | Average | Latest | Nov. 2 - 5 | Latest | Nov. 8 - 11 | Nov. 12 | Min | Overall | Max |
| Rossi's Victory | (3,339) | (1,676) | (3,641) | 174 | (4,400) | (3,403) | (1,167) | ||||||||
| Total Uncounted | 41,623 | (1,419) | 244 | (1,721) | 2,094 | (2,480) | (1,483) | 753 | |||||||
| Adams | 25 | -38.53% | -29.94% | -38.53% | -29.94% | -38.27% | -29.94% | (10) | (7) | (7) | (10) | (10) | (7) | ||
| Asotin | 0 | -15.53% | -27.95% | -27.95% | -15.53% | -16.00% | -27.95% | 0 | 0 | 0 | 0 | 0 | 0 | ||
| Benton | 2,200 | -37.46% | -42.00% | -42.00% | -37.46% | -38.06% | -42.00% | (824) | (924) | (924) | (924) | (837) | (824) | ||
| Chelan | 500 | -28.21% | -30.49% | -35.47% | -35.47% | -28.21% | -28.98% | -35.47% | (141) | (177) | (152) | (177) | (177) | (145) | (141) |
| Clallam | 250 | -6.44% | -12.54% | -12.54% | -6.44% | -7.26% | -12.54% | (16) | (31) | (31) | (31) | (18) | (16) | ||
| Clark | 2,200 | -8.04% | -8.05% | -8.29% | -8.29% | -8.04% | -8.04% | -8.29% | (177) | (182) | (177) | (182) | (182) | (177) | (177) |
| Columbia | 285 | -33.92% | -33.92% | -33.92% | -33.92% | -33.92% | (97) | (97) | (97) | (97) | (97) | ||||
| Cowlitz | 50 | 0.90% | -4.29% | 1.79% | -4.29% | 1.79% | 0.31% | 1.79% | 0 | 1 | (2) | 1 | (2) | 0 | 1 |
| Douglas | 200 | -32.63% | -33.84% | -33.33% | -33.84% | -32.63% | -32.70% | -33.33% | (65) | (67) | (68) | (67) | (68) | (65) | (65) |
| Ferry | 5 | -19.51% | -3.49% | -19.51% | -3.49% | -19.09% | -3.49% | (1) | (0) | (0) | (1) | (1) | (0) | ||
| Franklin | 250 | -34.92% | -44.47% | -44.47% | -34.92% | -36.01% | -44.47% | (87) | (111) | (111) | (111) | (90) | (87) | ||
| Garfield | 0 | -31.26% | -46.91% | -46.91% | -31.26% | -32.25% | -46.91% | 0 | 0 | 0 | 0 | 0 | 0 | ||
| Grant | 50 | -37.21% | -40.15% | -40.15% | -37.21% | -37.24% | -40.15% | (19) | (20) | (20) | (20) | (19) | (19) | ||
| Grays Harbor | 500 | 0.19% | -9.32% | -6.69% | -9.32% | 0.19% | -0.65% | -6.69% | 1 | (33) | (47) | (33) | (47) | (3) | 1 |
| Island | 850 | -7.33% | -11.48% | -11.56% | -11.56% | -7.33% | -8.04% | -11.56% | (62) | (98) | (98) | (98) | (98) | (68) | (62) |
| Jefferson | 25 | 17.76% | 20.41% | 30.32% | 17.76% | 30.32% | 18.08% | 30.32% | 4 | 8 | 5 | 8 | 4 | 5 | 8 |
| King | 11,000 | 16.98% | 21.65% | 26.67% | 16.98% | 26.67% | 17.58% | 26.67% | 1,868 | 2,934 | 2,382 | 2,934 | 1,868 | 1,934 | 2,934 |
| Kitsap | 2,500 | -1.17% | -3.61% | -0.53% | -3.61% | -0.53% | -1.38% | -0.53% | (29) | (13) | (90) | (13) |
Posted by Stefan Sharkansky at November 13, 2004
01:41 AM | Email This
Comments
1. Anybody want to place guesses, educated or not as to when the litigation over the election will be over?
(I would bet good money the State Dems will sue and sue and sue until the last possible second into the new year because they can and don't want to relinquish power without a fight - besides, they're hijacked by the state employees' unions.) Thanks for all you do, Stefan. Posted by: Josef on November 13, 2004 02:13 AM
2. I have two fears about the remaining ballots.
Fear number 1. What if only 50% (or there abouts) of provisional ballots in Rossi counties are counted, while 90% (or there abouts) of provisional ballots in Gregoire counties are counted? Rossi might lose. Does anyone know if Rossi is as aggressive in pursuing provisional ballots in Rossi counties as Gregoire is in her counties? Fear number 2. Regarding the remaining provisional ballots in King County (an to a lesser extent, Whatcom and Thurston Counties), what if Gregoire contacts these voters and asked them how they voted. If they voted for her, she assists them in getting their votes counted. If they voted for Rossi, she ignores them, or worse, gives them false information as to how to get their votes counted. Such a tactic could greatly increase Gregoire's margin in King County. What's the likelihood that Gregoire will use such a tactic? Does Rossi have a counter tactic for this? I hope Rossi's team has thought though all the contingencies. I'm praying for a Rossi victory.
3. Could it be that the provisionals in King County are being counted by precinct? That would be how they are turned in on election night. The variation in results might be predicted by looking at what precincts have provisionals outstanding vs the counted results in those precincts.
Thank you for all the fine analysis. Posted by: John on November 13, 2004 08:10 AM
4. I don't want to rain on your Republican parade (actually, I would want to, I just don't think I can), but keep in mind that Maria Cantwell gained about 1500 votes in the recount.
Posted by: David Goldstein on November 13, 2004 09:02 AM
5. I estimate that if King County and only King County reports its remaining ballots, Gregoire will hold the lead by about 1,200 votes. Of course that would leave another 5,475 ballots to count in Gregoire friendly counties and 25,148 left in Rossi friendly ones. The real question isn't whether Rossi will win or not, it's whether or not he'll have enough votes to avoid a recount.
Posted by: Nathan on November 13, 2004 09:12 AM
6. This table looks very odd in Safari/Mac OS X. The 'bold' parts look like two non-bold sets of the same text, but not _quite_ lined up right. Really quite wild.
Reading the source quickly led into MS auto-generated code by the mile, so I don't see a quick fix. The std table on the post a couple down looks great though. Posted by: Al on November 13, 2004 01:37 PM
7. The tabling looks funky in Firefox, as well. Who's using IE these days? ;)
Posted by: Mike Christianson on November 13, 2004 10:37 PM
8. Forgot something: these votes are being counted by (mostly Democrat) UNION "workers" -- Boss Sims says it's Christy by 226 --
Posted by: Lew on November 14, 2004 05:14 PM
9. Regarding my previous post prediction "Boss Sims says Christi by 226" -- make that about 2200 -- 2800 to avoid run off --
Note: King County just "found" thousands of extra ballots -- ya, right...I wonder what would happen if 500-600 republicans stormed the county elections offices? (Democrats are a filthy, nasty bunch of people --liars all, corrupt to the core....) Posted by: Lew on November 15, 2004 05:10 PM
10. The main thing I notice about Democrats is that they usually resort to name calling and lack class. Rossi is winning because he presented a positive message that provides direction and thats what the people want in a leader.
Posted by: Dinah on November 17, 2004 07:32 PM
Post a comment
|