Those trying to discount the theory are missing several things and obviously didn't read the full billion page thread.
1. If Romney does so much better in the larger areas, as stated, then why didn't polling data from rural, urban, and suburban areas from three different states show this? Instead it showed that all of the candidates in the race at the time enjoyed almost the same percentage of support from all areas. Another possibility is that larger areas are more likely to be counted by electronic voting machines.
2. You shouldn't see the "mirror image" between Romney and Paul when there are multiple other people running. In the fullbillion page thread, not one detractor could come up with a plausible explanation for that.
Site your sources. We aren't simply talking about "rural" "suburban" and "urban" either. I predicted exactly where Romney would be strongest before the election. That's because I knew enough about the state, where the rich people lived, to make that determination. I said something like "people are going to cry fraud when they see how well Mitt Romney does in Southern Fairfield County. And he did get between 80-85 percent on the "gold coast", compared to 67% statewide.
Now, if you give me 3 sets of results, the connecticut results, with the graph including all votes, a graph of those 5 towns, and a graph with everything but those 5 towns, you'll see the "everything but those 5 towns" being a lot flatter.
And remember, I predicted this beforehand, I'm not choosing these results because the graph looked "wrong". The richest people, the people who work for Fortune 500 companies, all them, they like Romney and they don't like Ron Paul. And that little fact, or theory, really explains the results as well as "vote flipping".
I've heard flippers talk about how everywhere in the US it's pretty much the same. People in rich suburbs and people in urban areas think the same, so Candidate X and Candidate Y should have the same percentage in both areas. Which just shows complete stupidity on their part. People who haven't the slightest clue about politics are claiming vote flipping. I'm not saying that there isn't vote fraud like this. It's just that claiming that the lines should be flat in a Romney v Paul race, or one where they're major factors in the race is stupid.
Perhaps they've advanced, and gotten somewhere?