The first place to look is the 2005 race. Sure, the circumstances are different. That was an open seat race for example, but Jon Corzine will have competed as the Democratic candidate in both races. Just looking at the raw data, Corzine led throughout in the polling of the race. That lead fluctuated but was consistently in the upper single digit/lower double digit range for the duration. And once November hit, the current governor won by a little more than ten points.
So 2005 was not indicative of the small margin early/big margin late polling phenomenon witnessed in the 2004 presidential race. The striking thing as you look back beyond 2005 -- in terms of gubernatorial races -- is that there has not been a Democratic incumbent reelected governor of New Jersey since Brendan Byrne in 1977. The last Democratic incumbent that stood for reelection was James Florio, who lost to Christine Todd Whitman in 1993. Governor's races in New Jersey, then, may not be where the aforementioned pattern is witnessed.
There are, however, some interesting lessons to be learned from those particular races. It is fortunate that the Eagleton Center at Rutgers has a robust archive of the polls conducted on the gubernatorial races since 1973. Now granted, this is just one poll, but the series of them gives us some indication of what conditions were like during these election years. Ideally, we'd be able to get a sense of both what general polling looked like in the race as well as the approval level of the incumbent. In 2009, Corzine's approval -- or lack thereof -- will play a large role in determining the outcome of the race in November. And polling, relatedly of course, has favored the Republicans, specifically Chris Christie, throughout.
The Byrne example from 1977 is actually an interesting parallel. In May of that year, prior to the primary, only 27% of New Jersey voters supported Governor Byrne for reelection and 42% preferred the generic Republican. When Byrne was removed and it was a question pitting a generic Democrat against a generic Republican candidate, the Democrats had a 45%-20% advantage. Later on, in July, once Raymond Bateman became the GOP nominee, Byrne continued to trail in the Eagleton poll. By November, though, Byrne had completely reversed the trend and won by nearly 15 points. [You can read a much more thorough treatment of the dynamics of the race over at Blue Jersey from back in April.]
The 1993 race also offers some insight. In Florio's case, you have an incumbent Democrat who was in much better shape in the polls than Brendan Byrne was a decade and a half earlier. In the Eagleton polls conducted, Florio hovered around the 50% mark across all three polls with Christine Todd Whitman lagging behind, closer to the 40% level. The measure to take note of here, however, is the change in the undecideds. Contrary to what might otherwise be expected in a race involving an incumbent, the number of undecideds in the series of Eagleton polls rose as the race came to a close. That typically isn't the greatest indication of incumbent success heading down the stretch of a reelection effort. After all, incumbents are more a known quantity than their challengers are. And in this race, those undecideds seemed to have broken for Whitman in the end, or at least enough to tip the balance in her favor in a one percentage point victory.
Still, the Byrne example seems a better match in this case if only because the number of undecideds is likely to be rather small in a race with an unpopular incumbent. Either you still like him or you don't. A Byrne-like comeback is possible in 2009 as well, but it will be dependent upon Jon Corzine performing a balancing act between "defining" Chris Christie and smearing Chris Christie. The governor obviously has plenty of cash and is already on the air trying to do the former, but maintaining the perception that he is defining Christie and not trying to drag the former ambassador down to his (approval) level will be the true test; one made all the more difficult when the prevailing sentiment regarding the governor currently is negative. Christie is in the driver's seat, but he (or his campaign) has to be the one that is defining the race. "Not Corzine" may be enough, but we'll have to see how the race develops. Regardless, this one will be fun to track.
Cross-posted at Frontloading HQ.