There's a lot about the theory that's wrong, but as Scott Rasmussen notes, it might be best for Republicans if they don't overinterpret their mandate on Tuesday:
The reality is that voters in 2010 are doing the same thing they did in 2006 and 2008: They are voting against the party in power.
This is the continuation of a trend that began nearly 20 years ago. In 1992, Bill Clinton was elected president and his party had control of Congress. Before he left office, his party lost control. Then, in 2000, George W. Bush came to power, and his party controlled Congress. But like Mr. Clinton before him, Mr. Bush saw his party lose control.
That's never happened before in back-to-back administrations. The Obama administration appears poised to make it three in a row. This reflects a fundamental rejection of both political parties.
In this environment, it would be wise for all Republicans to remember that their team didn't win, the other team lost. Heading into 2012, voters will remain ready to vote against the party in power unless they are given a reason not to do so.
It's possible, of course, that the politics itself is the problem - that everybody who gains power starts to look stale and awful after two to four to six years or so, and everybody out of power starts to look like the Next Great Hope during that time. Under these circumstances, nobody's going to have a mandate - and everybody's going to overreach. We no longer have generational dynasties like the Democrats did during the FDR-to-LBJ era or the Republicans did from McKinley to Hoover. The news cycle can now be measured in minutes; the political cycles as a result can probably be measured in years or months.