"at least on the ballots that can be verified as having been cast as per the voter's intent." (Brad Friedman)
According to the official Delaware elections website with 100% reporting:
Verifiable Paper-Based Absentee Results:
CASTLE: 54.7%
O'DONNELL: 45.3%
Unverifiable Election Day E-Voting Machine Results:
CASTLE: 46.7%
O'DONNELL: 53.3%
Nonetheless, the Tea-Party/Palin/DeMint-endorsed Christine O'Donnell, who was getting trounced by the popular Castle in pre-election polls until only recently after losing twice before in her quest for a U.S. Senate seat, was declared the "winner" of yesterday's race and --- as The BRAD BLOG detailed yesterday --- nobody can prove whether the voters of Delaware actually selected her or not.
Appropriately enough for the far Rightwinger, the "victory" was 100% faith-based, since it's strictly impossible to know if even one citizen's vote cast yesterday on the 100% unverifiable e-voting machines Delware forces voters to use on Election Day was recorded accurately...
That said, while we've seen examples of similar disparities between paper-based absentee results and electronically cast results before (the unknown Alvin Greene's "victory" over Judge Vic Rawl in South Carolina's recent Democratic U.S. Senate primary comes to mind) there are logical-ish reasons --- as there always are, in every election --- to justify O'Donnell's computer-reported "victory" yesterday.
As we noted in response to a reader in comments on yesterday's Delaware item, O'Donnell received a late endorsement from Sarah Palin on September 9th, just 5 days before the election. That brought with it a surge of last-minute support from the "Tea Party" and others.
Moreover, the number of absentee ballots cast as a percentage of the total votes was quite small (1,499 absentee ballots, versus 56,083 cast on Election Day), so one should be careful of reading too much into those numbers as the bulk of absentee ballots were likely cast prior to O'Donnell's endorsements surge
...snip
Wouldn't it be nice to live in a democracy where you could be relatively sure that the reported results would match the actual intent of the voters? In such a hypothetical US political system, the pre-election polling and the paper based voting (absentee ballots, etc.) would probably match pretty closely the final result. Right now, this is just a fantasy, a dream that I feel sure I won't live long enough to see realized.
Read the whole article here:
http://www.bradblog.com/?p=8073