I've been using the term "Rove margin" in a number of comments now, and one Kossak (Oxon) liked it so much he asked me to define in a diary (and a hat tip to you!).
Rove Margin (n.): A number such that, if a GOP candidate's likely votes are greater than it, but less than 50%, it is feasible to steal the remaining votes needed to win. A Rove Margin depends on the electorate's credulity, and therefore can only be approximated.
Like the Rove Margin, this diary is subjective. So far as I know, no one has ever admitted to using the concept behind the Rove Margin - not surprisingly, since it describes a federal felony. I'm offering a speculative explanation for what we've been seeing in the past 6 years, and especially as it's being uncovered now. Bear with me.
The usual rule in high-level politics is to run to the extreme in the primary and then run to the center in the general. This rule is based in part on the understanding that one needs a reasonably broad consensus in order to govern effectively.
Bush (and Rove) threw this rule away in the 2000 election. Instead of a broad mandate, they went for 50%+1. Part of the reason for this is that they were much more interested in power than in governing, effectively or otherwise.
(A side effect of the 50%+1 strategy is that it only works as long as you are successful. When things go sour - and they always will - there is no store of general good will to draw on. This is one reason Bush is at 28% and slipping; he was never interested in extending his majority because he never expected things to go sour. But I digress.)
A more important side effect of the 50%+1 strategy is that it either springs from or encourages the mindset that the other 49% do not matter, which can easily lead to the next step, which is that the other 49% have no legitimate concerns, including their concern that their votes should actually count. In fact, the 50%+1 strategy quickly comes to depend on making sure that not all of the other 49%'s votes are counted. Ands that's where the Rove Margin comes in.
The Rove Margin is used when allocating election fraud resources. It is applied to candidates who are below 50%+1 in the polls, and who are not likely to get there (acutally, the line is probably set at 55% or so, to account for poll margin of error and to add a safety factor). On the other hand, they are not so far behind in the polls as to have no chance of winning. This is where the Rove Margin comes into play. If a candidate gets to within this margin, then, through use of voter suppression, voter intimidation, election machine manipulation, and similar tricks played on and just before election day, the candidate can acquire the remaining votes needed to reach 50%+1. (Thus, in 2006, the GOP gave up on BLackwell in Ohio but put effort into the Montana and Virgina senate races, both of which had results which did not square with the exit polls.)
The key is credibility. The margin has to be small enough that the press and the public will accept the results, even though the final pre-vote poll and the exit polls predicted the opposite result. This is what makes the Rove Margin a subjective number rather than something mathematically calculable, but it does give new meaning to this quote by Rove just before the last election:
Rove claimed that the polls "add up to a Republican Senate and a Republican House."
"You may end up with a different math, but you're entitled to your math," Rove said. "I'm entitled to 'the' math."
As it turns out, he got "the" math wrong, particularly in Virginia and Montana; he underestimated the level of the country's rejection of the GOP. But this shows the paradox built into the Rove Margin: The higher the level of disgust with the GOP, the wider the Rove Margin needs to be in order to keep the GOP in control, but that same level of disgust also means the country will be less willing to accept a questionable election result, and that means a narrower Rove Margin.
I'm sure that Rove recognizes the paradox, and I postulate that this is one reason for the currently unraveling USA scandal. If the public can be made to see the Democrats as perpetrators of voter fraud, then they will accept a wider Rove Margin, because now they will also factor in those accusations when the time comes to explain away the discrepancies between the polls and the election results.
Rove has another problem too: States are finally starting to put in meaningful safeguards to ensure that the vote is counted correctly. While this might allow for a wider Rove Margin in that the public will be more likely to trust the count, it also makes it more difficult to rig the count. (No electronic or other voting system is perfect, but optical scanning is currently the best, because it creates a paper record that can be used in a non-electronic fashion to verify the electronic count.) So, again, Rove needs another strategy before he can put the Rove Margin into play. Suborning the USAs was a risk - and he may have lost his gamble, since its exposure means that any indictments brought in 2008 will likely not be seen as credible and may even elicit a sympathy vote. But Rove was in a bind.
Awwww.