Count every vote, again
Advertisement
Text size: small | medium | large
By Published by The Editorial Board
Published: November 25, 2008
Most elections aren’t close. Most of the time, it’s clear on Election Night who won and who lost — and that’s the end of it.
That can’t be said of the 5th District congressional race between Republican Virgil Goode and Democrat Tom Perriello. Once the votes were canvassed during the days following the election, Perriello held a 745-vote lead.
On Monday, the Virginia State Board of Elections certified Perriello’s win over Goode, and Goode served notice that he would ask for a recount. Both actions make a lot of sense.
Since the vote canvass was completed earlier this month, Perriello has been acting like the winner. He has formally thanked his supporters, named a transition team that includes several locally prominent leaders and attended new member orientation in Washington.
Perriello has done that knowing that Goode could — and probably would — ask for a recount.
“We respect Virgil Goode’s right to ask for a recount and we are confident the recount will not alter the outcome of this race,” Jessica Barba, communications director for the Perriello campaign, said in a news release Monday. “The official vote canvassing process that took place in the days immediately following the election was extensive, thorough and overseen by representatives of both parties. A re-tabulation of the results of the vote canvass will result in what the Virginia State Board of Elections certified today — that Tom Perriello has been elected as the new congressman for the 5th District.”
Goode owes himself — and his supporters — this recount. Perriello’s 745-vote margin gives him just 50.08 percent of 316,862 votes cast. Goode was 0.23 percent behind Perriello with 49.85 percent.
It’s understandable if Goode doesn’t want to close out a 12-year congressional career with a loss that close. When the sun came up on Wednesday, Nov. 5, Goode was leading by 446 votes. During the canvass, 3,200 votes changed with “the largest changes coming from cities and counties that voted for my opponent,” Goode said in a statement Monday.
“In an election this close — with less than one-quarter of 1 percent separating the two candidates, and with literally thousands of vote changes taking place during the post-election canvass process — a recount is an important protection for voters,” Goode wrote.
Now is the time to clear up lingering doubts about the 5th District race, and the only way to do that is for the recount to proceed. So let’s count those votes one more time — and make sure the totals are correct.
Page 1 of 1
Post a Comment
The commenting period has ended or commenting has been deactivated for this article.
Can't find what you're looking for? Try our quick search:
Reader Reactions
Posted by ( sulu69 ) on December 01, 2008 at 9:16 pm
So Gordie where would yu like to meet? With all those numbers I know that you need to get out. We could go shopping together for starts.
Report Inappropriate Comment
Posted by ( Gordie ) on December 01, 2008 at 7:31 am
sulu69 The linear model is the most straightforward, but the quadratic model adds in an x^2 (x squared) term and is a much better fit statistically, and it accommodates expected diminishing returns for large amounts of money.
For the quadratic version of the revised model take x = total contributions divided by 100,000. The 100,000 is just a scaling factor. You’re right - calculate y = predicted returns = model score
y = -0.0768*x^2 + 2.3223*x + 34.631
The linear model will use
y = 1.0379*x + 36.653
Once you get the data in columns in Excel, you can do regressions using the scatterplot chart, and right click to add trendlines and options. That’s the easiest way to get the regression equation with one explanatory variable (contributions).
When you have more than one explanatory variable, it makes sense to use more versatile software. But also, there are lots of new considerations - for example - if you want to introduce a new variable into the model, then you need to examine whether or not it is correlated with an existing variable and so on. I tried using the quotient of the contributions divided by the MFI as a single new variable and that is the test model (there were some others). And I did try some multivariate models.
As for your blueness variable, that definitely sounds like a good thing to add. But I don’t know how hard it would be to derive such a measure. It would be easier to figure in whether or not an incumbent is being challenged or not.
What I should have emphasized is that none of this is fixed in stone and is only for getting a better sense of what goes on - not predictions. Not only are many other variables huge factors, some of them only matter in particular elections (e.g., marriage amendment). Then each election has its individual features. Also unique was as noted, much of Perriello’s out-of-state contributions was a vote against bigotry by many people.
Report Inappropriate Comment
Posted by ( Gordie ) on November 29, 2008 at 5:54 pm
sulu69 the idiosyncrasies of the mentally incompetent have a tendency to believe that thru their slow brains and eyes compare others to be exactly as they are when the opposite is the correct sexual preference of the individual they assume is gay. Though in reality “sticks and stone may break ones bones name will never hurt”. The idiosyncrasies of assuming definitely makes an ??? of the person assuming.
Report Inappropriate Comment
Posted by ( Gordie ) on November 27, 2008 at 9:12 am
Before the editorial staff makes such a comment “Count every vote, again” they should know that is not possible. Since electronic touch screens have no paper trail and just a tabulation of which candidate got what vote, all that can be done is look at those tabulation tapes again. Those tapes where gone over by just about everyone already and I am sure the courts can only say tabulate them again. Unless when the registrars testify the court find their statement acceptableand does not say to do a re-tabulation.
As far as the optic scanners, they run those thru the machine again, that machine is programed to look at only that race and programed to set all under and over votes aside as well as write in votes, which will be visually checked.
Folks this is not Florida and don’t expect to see someone with a magnifying glass checking chads.
Report Inappropriate Comment