There's definitely a lot to that theory. There are a number of fairly significant states where Romney won, which could have been taken by Santorum had Gingrich not been splitting the vote. There's even talk that the wealthy donor who is more or less single handedly funding Gingrich's campaign at this point is doing so to stop Santorum.
Also, since 1976, no incumbent president has been reelected when the unemployment rate is at or over 7% I don't know if there is enough time to get the rate back under 7% between now and November.
Statistics like this are meaningless bullshit that news networks use to fill airtime.
The unemployment rate is "meaningless bullshit" as it applies to and impacts presidential politics? Wow. Um, ok.
No. Statistics like "only X out of Y Presidents between the years E and F were re-elected when statistic A was at B%" are meaningless. The sample size as far as presidential elections go is just far too small, and the number of variables too great and differing, for any statistics like that to mean anything in that sort of context. I explained that in my post about this.
It's like when CNN go on about their exit polls where 37% of men in Jackson county who are left handed and put cream in their coffee chose Romney over Santorum. It's just chatter to fill time.
Unemployment obviously matters. It's currently lower than it was when Obama took office. That's what matters.
Including the 1976 election, we've only had 3 incumbent presidents lose re-election in that timeframe; Ford, Carter, and Bush senior. Three elections each with a large, differing set of circumstances doesn't really lend itself to statistics like the one you presented.