Of course the entire Michigan vote was a charade. Former Michigan Senator Donald Riegle compared it to Soviet elections: "a sham" she, Bill, and her supporters "rigged to give the nation the impression that she's the leading candidate in Michigan." In an October 2007 New Hampshire Public Radio interview that every delegate should hear, Clinton justified her staying on the Michigan ballot by explaining, "this election they're having is not going to count for anything." As Michigan Public Radio commentator Jack Lessenberry pointed out, fewer than 600,000 voted in the state's Democratic primary, compared to 867,000 Republican votes (and 2.5 million votes for John Kerry in 2004). So Michigan and Florida are among just a handful of states where Republican turnout exceeded that of the Democrats in this year's primaries. The vast bulk of Michigan Democrats stayed home, with way fewer voting than in far smaller states.
Michigan Democrats who felt the vote was meaningless were in fact encouraged to cross over. Since everyone (including Clinton) said the results wouldn't count, Daily Kos founder Markos Moulitsis suggested Michigan Democrats vote for Mitt Romney to prolong the Republican race, keep the Republican candidates at each others' throats, and perhaps help nominate the presumably less-electable Mitt Romney. In other words, pretty much what Rush Limbaugh and his allies ended up doing, except that unlike Indiana and Ohio, Michigan had no laws even theoretically prohibiting such an action. No one thought Clinton would have the chutzpah to retroactively claim a Soviet-style victory. Other progressive bloggers picked up on the idea as well. I also heard it discussed on my local Air America affiliate. These strategic voters, whether inspired by the blogs or self-initiated, combined with other Democrats who simply figured McCain was less fundamentalist than Huckabee and had more substance than Romney's empty-suit puffery, so would still be a better choice for America if the Democratic candidate lost. Had all of them stayed in the Democratic primary and voted against Hillary, it would have tipped her to 48 percent.
Kos and his DailyKos site do valuable work, and I suspect Limbaugh would have launched his campaign without the precedent. But in both cases, Clinton benefited. As soon as McCain had clinched the nomination, Limbaugh and allies like Laura Ingraham began encouraging Republicans to further an increasingly nasty Democratic fight. Obama had been gaining legitimate support from Republicans simply inspired by his message, sick of Bush, and therefore open to changing. A conservative Mormon accounting professor me saying "Paul, you aren't going to believe this. Obama is way too liberal for me, but I'm going to vote for him because I think he has integrity." Post-Limbaugh, those who switched included a substantial number of Republicans trying to disrupt the Democratic primary. As the Boston Globe reported, "In Ohio and Texas on March 4, Republicans comprised 9 percent of the Democratic primary electorate, more than twice the average GOP share of the turnout in the earlier contests where exit polling was conducted. Clinton ran about even with Obama among Republicans in both states, a far more favorable showing among GOP voters than in the early races." A Wall Street Journal story found similar results. Twelve percent of Mississippi's Democratic primary voters were Republicans, breaking three to one for Clinton, but 31% of that group said she wasn't honest and trustworthy, which hardly suggested they'd be voting for her come November. In Indiana, Huffington Post staff reporter Sam Stein points out that seven percent of those who voted for Clinton in the primary say they wouldn't vote for her in November. But without those voters, she wouldn't have had her two-percentage-point victory.
This resonates with my experience. I've gotten emails from people throughout these states who've described coworkers, neighbors, or friends who they witnessed bragging or laughing about being part of Rush's crossover legions. I've read many similar first-person accounts on various blogs. Given the size of Limbaugh's audience and that of his allies in this effort, it seems perfectly conceivable that he shifted 2-3 percent of the vote in each of the states that he targeted. Those jumping on the Daily Kos campaign in Michigan had a far smaller megaphone, but operated under similar strategic assumptions. They were hardly Hillary supporters, or they'd have voted for her, but the numbers suggest their votes made a difference, helping give her the "victory" she now clings to.