Was the 2004 election stolen? No.

Claim: In rural counties in Ohio, more than 150,000 votes meant for Kerry were somehow switched to Bush.

"An examination of election data suggests widespread fraud -- and even good old-fashioned stuffing of ballot boxes -- in twelve sparsely populated counties scattered across southern and western Ohio," Kennedy writes. The counties he suspects are Auglaize, Brown, Butler, Clermont, Darke, Highland, Mercer, Miami, Putnam, Shelby, Van Wert and Warren. "One key indicator of fraud is to look at counties where the presidential vote departs radically from other races on the ballot," he writes. "By this measure, John Kerry's numbers were suspiciously low in each of the twelve counties -- and George Bush's were unusually high."

Kennedy points to vote results for Ellen Connally, a liberal Democrat who ran for chief justice of the state Supreme Court. Kennedy contends that Kerry's vote totals in the presidential race should have exceeded Connally's in the Supreme Court race in these rural counties; you wouldn't expect a relatively unknown liberal to win more votes than a well-known moderate in a rural area.

"Yet in these twelve off-the-radar counties, Connally somehow managed to outperform the best-funded Democrat in history, thumping Kerry by a grand total of 19,621 votes -- a margin of ten percent," Kennedy writes. To Kennedy, this indicates that a lot of the people who voted for Connally also intended to vote for Kerry, but their votes somehow didn't show up. Rep. Dennis Kucinich tells Kennedy, ''Down-ticket candidates shouldn't outperform presidential candidates like that. That just doesn't happen. The question is: Where did the votes for Kerry go?''

Kennedy says Kerry's votes "were fraudulently shifted to Bush." He points out that "statewide, the president outpolled Thomas Moyer, the Republican judge who defeated Connally, by 21 percent. Yet in the twelve questionable counties, Bush's margin over Moyer was 50 percent -- a strong indication that the president's certified vote total was inflated. If Kerry had maintained his statewide margin over Connally in the twelve suspect counties, as he almost assuredly would have done in a clean election, he would have bested her by 81,260 ballots. That's a swing of 162,520 votes from Kerry to Bush -- more than enough to alter the outcome."

Reality: Kennedy's pattern sounds intriguing. But as Mark Lindeman, a political scientist at Bard College, pointed out to me, the whole story dissolves when you look at results from previous elections.

Contrary to Kucinich's assertion, down-ticket candidates do indeed sometimes win more votes than presidential candidates of their own party in some places -- sometimes a lot more. In 2000, Democratic state Supreme Court candidate Alice Resnick won more votes than Al Gore in dozens of counties -- in 81 counties, which makes the 12 counties where Supreme Court candidate Connally outperformed Kerry in 2004 look not very suspicious at all. (I arrived at these numbers using Excel and Ohio's 2000 county-by-county results, available here.) If Kennedy considered Connally's 19,000 vote margin over Kerry in 12 counties a "thumping," I wonder what he'd think of Resnick's margin over Gore -- she won 126,000 more votes throughout the state than did the incumbent vice president (she won her race against her opponent, too). Tim Black, another Democratic Supreme Court candidate, lost his race, but he too managed to outperform Gore in 40 counties.

Lindeman points out that the numbers work out this way for a very specific reason -- ballots in Ohio don't list party affiliations for Supreme Court races. Kennedy finds it unlikely that someone in a rural Ohio county would have cast a ballot both for Bush and for a liberal justice like Connally. But if you consider that those voters might never have heard of Connally and had no idea she was a Democrat, there's no surprise why they might have chosen her. Therefore, Kennedy's assertion that 162,000 Kerry votes were switched to Bush falls apart.

It's worth noting, too, that a team of political scientists hired by the Democratic Party to investigate what happened in Ohio also used statistical analysis to search for any pattern of obvious shifts from Bush to Gore in the vote count. That group saw no evidence of fraud (PDF). "The tendency to vote for Kerry in 2004 was the same as the tendency to vote for the Democratic candidate for governor in 2002," their report noted. "That the pattern of voting for Kerry is so similar to the pattern of voting for the Democratic candidate for governor in 2002 is, in the opinion of the team's political science experts, strong evidence against the claim that widespread fraud systematically misallocated votes from Kerry to Bush."

They added: "Kerry's support across precincts also increased with the support for Eric Fingerhut, the Democratic nominee for U.S. Senate, and decreased with the support for Issue 1 (ballot initiative opposing same-sex marriage) and increased with the proportion of African American votes. Again this is the pattern that would be expected and is not consistent with claims of widespread fraud that misallocated votes from Kerry to Bush."

Kennedy cites parts of their report several times, but he does not mention this conclusion.

Claim: Blackwell engineered a "purge" of 300,000 voters in Ohio's major cities.

Kennedy writes that "Blackwell permitted election officials in Cleveland, Cincinnati and Toledo to conduct a massive purge of their voter rolls, summarily expunging the names of more than 300,000 voters who had failed to cast ballots in the previous two national elections. In Cleveland, which went five-to-one for Kerry, nearly one in four voters were wiped from the rolls between 2000 and 2004."

He concedes that there were "legitimate reasons to clean up voting lists: Many of the names undoubtedly belonged to people who had moved or died. But thousands more were duly registered voters who were deprived of their constitutional right to vote -- often without any notification -- simply because they had decided not to go to the polls in prior elections." Kennedy estimates that 10 percent of these 300,000 voters represented actual voters who were disenfranchised. He concludes that Blackwell's actions put 30,000 votes in the missing column.

Reality: Scrubbing the voting rolls of people who hadn't voted in prior elections isn't an arbitrary move. It's the law. Here's the relevant section of the Ohio code, 3503.19, which states that a person who "fails to vote in any election during the period of two federal elections" shall have his registration "canceled." To be sure, people who intended to vote and weren't aware of this rule could have been cut from the rolls, and you might say that's unfair. But that's an argument for a better election law, and not proof that the purges were part of a Republican election-theft plot.

Next page: Republicans couldn't have engineered long lines in Democratic precincts