Here are two reasons from Arizona history that should worry us about Brewer v. Goddard.
1) We re-elect incumbent governors, even if they were never elected to the job in the first place. Since the switch to four-year terms, five governors have run for reelection. All have been reelected. Assuming office only 8 months before his first election, Bruce Babbitt beat Evan Mecham 52%-44% and Leo Corbet 62%-32% in 1978 and 1982, respectively. Fife Symington beat Eddie Basha 53%-44% in 1994. A little more than a year after Symington's resignation, Jane Dee Hull crushed Paul Johnson 61%-36%. Janet Napolitano's reelection over Len Munsil was an even bigger blowout, 63%-35%.
2) We don't elect governors of the president's party. Now this is less of a hard-and-fast rule than 1) but it's still a clear trend. The president's party has lost six of our nine elections in the four-year term era (Castro/Ford, Babbitt/Reagan, Symington/Clinton, Hull/Clinton, Napolitano/Bush twice). The three exceptions are somewhat explanable. Babbitt's 1978 (during the Carter administration) falls under the reelection rule. Mecham's 1986 win was a fluke, as Carolyn Warner (D) and Bill Schulz, who had dropped out of the Democratic primary to run as an independent, split 60% of the vote. And Symington's win over Terry Goddard in 1990 was one of the closest elections in Arizona history, going to an eventual run-off (although his loss in that election is perhaps another reason to worry).
History is obviously not determinate of future elections. Our first 55 presidential elections resulted in white presidents, after all. But given that we'll be running against an incumbent governor, two years into a Democratic presidency, we should be very worried.
Another bit of history, Goddard has out-polled Brewer in both percentage and raw votes in both their 2002 and 2006 races. That's a good sign. So's this 2007 Rocky Mountain Poll matching Terry against Andy Thomas.
At this point, I call it a toss-up. We'll see how the new governor does.