While some caution is always in order, it sure looks as if it'll be a good day for Newt Gingrich. Which would have consequences: Mitt Romney won't be able to act as if he's the nominee for some time now, probably not until after Super Tuesday in March.
But there's a lot of overreaction going on, too. As Nate Silver noted on twitter (sorry, didn't save the link), Romney's InTrade odds have dropped dramatically this week, from over 90% to a current 68%. That's almost certainly far too low, but it's also not unusual at all for conventional wisdom to fall into a panic whenever an almost-certain nominee loses a primary. Most of the time, however, it doesn't mean anything. Here's a little review for you:
In 2008, John McCain lost 19 states.
In 2000, George W. Bush lost 7 states.
In 1996, Bob Dole lost 6 states.
In 1988, George H. W. Bush lost 9 states.
And in 1980, Ronald Reagan lost 6 states.
Of course, none of this means that Romney will win (all of the losers those years lost lots of primaries, too!). It's just that losing a couple of states, in and of itself, doesn't mean that he's not going to get the nomination, doesn't mean that it will be seen after the fact as a particularly difficult struggle for the nomination, and certainly doesn't mean that he'll be a weak nominee if he does win. And, yes, people had just as large overreactions when Reagan and George H. W. Bush lost Iowa and when W. lost New Hampshire (although the 1988 race was in fact probably still very much up in the air at that point). My own sense at the beginning of the week was that a Mitt/Newt race is pretty much a lock for Romney, and I don't see any reason to change my mind about that after a very good week for Gingrich.