Greg Borowski:

A review of Milwaukee voting records from the Nov. 2 presidential election has found more than 1,200 ballots cast from invalid addresses in the city, including many cases in which the voter could not be located at all.

The number is a result of a detailed computer analysis by the Journal Sentinel of the city’s voter records and represents about 0.4% of the 277,535 ballots cast in the city in the hard-fought election. Some of the problems may be due to flawed record keeping, such as transposed digits or incorrect street names. Many others, however, cannot easily be explained.

The newspaper’s review, the most extensive analysis done so far of the election, revealed 1,242 votes coming from a total of 1,135 invalid addresses. That is, in some cases more than one person is listed as voting from the address. Of the 1,242 voters with invalid addresses, 75% registered on site on election day, according to city records.

Wisconsin Counties are required by law to send voter address verification cards to new voters. Evidently, this law is not enforced, according to this AP article.

Kevin Kennedy, executive director of the state Elections Board, said local officials have said prosecutors rarely follow up on the undeliverable cards, even though they could lead to charges if voter fraud was involved.
“Municipalities have complained for years the D.A.’s don’t do anything with it,” Kennedy said. “That’s the feedback we get when we hold training sessions.”
In Milwaukee, there have been “several hundred” cards returned as undeliverable, said Lisa Artison, executive director of the Milwaukee Election Commission. The cards are still arriving and have not yet been sent to the district attorney’s office, she said.
In Dane County, District Attorney Brian Blanchard said he didn’t know how many cards had been returned but noted a clerk had bundled up a 6-inch stack.

Seattle blogger Stefan Sharkansky on voting irregularities their Governor’s race:

Christine Gregoire has said that our recent election was “a model to the rest of the nation and the world.” If what she meant is that the King County Elections Office is her model of how she plans to run the state of Washington, then we should all be worried.
Is it really “good enough for government work” to count 3,500 or 2,000 more ballots than there were voters? The airlines figured out years ago how to match the number of boarding passes with the number of people sitting in the airplane. Why can’t our elections officials match the number of ballots cast with the number of voters who supposedly cast them?
I think most Washingtonians agree that it isn’t good enough for government work to decide an election by a box load of funny votes. It is not the American way for a tainted victory of 129 votes, marred by thousands of illegitimate votes, including double voters, felon voters, cemetery voters and unidentified voters, to take the place of a legitimate decision of the electorate.

Greg Borowski has a useful followup to recent discussions on Milwaukee’s unusual election numbers today:

At issue is a gap between the city’s estimate of 84,000 election-day registrants and 73,079 verification cards that were sent, as required by law.
Local bloggers and others, including talk radio hosts, have labeled the gap as evidence of more than 10,000 illegally cast ballots.
Stone has stopped short of calling the ballots fraudulent but said “it casts doubt over the 10,000 votes, who cast the 10,000 votes, where those people live and whether they were eligible to vote in the city of Milwaukee.”

More on Madison’s numbers:

In Madison, the city clerk’s office doesn’t keep a tally of same-day registrants whose addresses could not be verified. In the November election, 17,467 people registered at the polls, but city officials have no idea how many of those addresses could be verified, said Sharon Christensen, deputy city clerk.

Ed Morrissey:

Michelle Malkin points out a story that I had missed in neighboring Wisconsin, one that calls into question the veracity of its presidential-election results. Wisconsin wound up going for John Kerry by 11,300 votes in what came as a mild surprise to most observers in the Upper Midwest (via Stranded On Blue Islands). Al Gore had carried the state by a shade over 5,000 votes in 2000, and most pollsters had the race a dead heat or George Bush pulling slightly ahead in 2004. Instead, Kerry took Wisconsin by doubling Gore’s total.
How did that happen? Well, in one county — Milwaukee, a traditional Democratic stronghold — turnout increased by just under 49,000 votes, or about 10%, outstripping the nationwide increase of 6.4%. The new votes broke about 60/40 Kerry, about the trend of the county in both elections, adding a 9,000-vote margin to Milwaukee over Gore in the last election.
But here’s where the Silence Of The Cheese gets … well, stickier. According to state records, 83,000 people executed a same-day registration for Milwuakee County, which is more than 20% of all voting-age residents in the county. Now, Wisconsinites may procrastinate a bit, but in order to believe that number, you’d have to expect that 20% of the county had moved or became newly eligible within the past two years (after the previous national cycle). Not only that, but the state now reports that 10,000 of those registrations cannot be verified, a whopping 12% of all same-day registrations and almost the entire margin of victory for Kerry for the entire state.

There’s more, including some corrections to the data. Via instapundit.