Though he fell just short of a majority, Dalton McGuinty's somewhat surprising come-from-behind election win in Ontario caps a pretty good electoral year for incumbents.

Beginning with Stephen Harper's federal victory in May, his first majority in three tries, incumbent governments were also returned earlier this week in Prince Edward Island (Liberal) and Manitoba (NDP).

Throw in Monday's election in the Northwest Territories, where only one of the 15 incumbents running was defeated (just five of the 19-member legislature will be new faces), and you might be forgiven for thinking incumbency is the magic key to staying in office these days.

The next test in this hyperactive election season is Newfoundland and Labrador on Tuesday followed by Saskatchewan on Nov. 7.

In both cases, the incumbent Conservatives and Saskatchewan Party, respectively, were enjoying healthy leads, at least according to the pre-writ polling.

So, what's going on? Are election-weary Canadians (four federal elections in the past seven years) just sticking with the devils they know? Or are we perhaps looking for a little stability in the face of a gathering economic storm?

Underlying turbulence

Harper certainly played the economic stability card in the federal campaign (an almost monotonous drip-drip-drip of past commitments to show that he wasn't about to play with the nation's piggy bank).

So did McGuinty, running predominantly on his record, eschewing costly giveaways. As did Manitoba's Greg Selinger in his first win as leader, but the fourth in a row for the provincial NDP.

With Manitoba's unemployment rate at 5.4 per cent, well below the national average, and the Winnipeg Jets NHL team back in town, there was not a lot of economic egg throwing to clutter up that campaign.

Still, each of these wins had their own political (and regional) flavour as well as an underlying turbulence that belies the importance of incumbency, particularly in these days of fixed election dates.

In the spring federal election, for example, Harper started with a modest lead, if the polls can be believed.

But he wasn't in majority territory by any means and his victory was anything but predictable. Who could have foreseen the wild ride of the last week or so when the NDP would surge in Quebec (58 seats!) and destroy the Michael Ignatieff Liberals as a force to be reckoned with.

Meanwhile, in both Ontario and Manitoba, the governing parties were looking more than a little shopworn and trailing in the polls in the months before their respective elections were called.

Incumbency didn’t look like much of an advantage at that point. In March, nearly half of Manitoba voters were saying they wanted a change of government, while a Toronto Star poll had Dalton McGuinty's approval rating at a meagre 16 per cent.

In both cases, however, some aggressive campaigning, coupled with mistakes by their main opponents — and, in Manitoba anyway, the collapse of the third-party Liberals — changed the dynamic of the races.

Manitoba's was a classic two-party urban-rural confrontation, with the NDP claiming 46 per cent of the vote (but proportionally more seats) to 44 per cent for the opposition Conservatives.

Ontario's was very much a three-party race with an only slightly surging NDP allowing the Liberals to squeak through in a number of key ridings and take 53 seats (one short of a majority) with only 38 per cent of the vote, barely three percentage points more than the Tim Hudak’s Conservatives, which won only 37 seats.

The hedging of bets

The re-election of Robert Ghiz's Liberal government in P.E.I. was also a pretty aggressive affair, observers pointed out, particularly when Conservative Ottawa weighed in on an alleged scandal involving immigrant investors.

Another factor, quite apart from incumbency, is the time-honoured Canadian practice of hedging our political bets.

But as political scientist Peter McKenna of the University of P.E.I. has observed, the result could also be attributed to "the ineluctable forces of electoral history." Over the past 75 years, he noted, "virtually every P.E.I. government that secured a first mandate was given a second (and even a third) in the succeeding provincial elections."

Another factor, quite apart from incumbency, is the time-honoured Canadian practice of hedging our political bets.

A Conservative government settles in for what could be the long haul in Ottawa, and provincial voters take a hard look at whom they want their provincial champions to be.

Look at the political map when Liberal prime ministers Pierre Trudeau and Jean Chrétien were in power: not many Liberal partners on the provincial scene. The reverse was true in the Brian Mulroney years.

Liberal Dalton McGuinty certainly mined this argument during the closing days of the Ontario election ("Who do you want standing up to Stephen Harper over health care?").  And it was a prominent theme in P.E.I. as well.

Not to put too fine a point on it, but even Alberta's new Conservative premier is viewed as a red Tory, with what is said to be no overabundance of affection for her Ottawa cousins.

So, if you are an incumbent premier with no direct ties to Conservative Ottawa, now would be the perfect time to pull the plug and get a new mandate, right?

Well, maybe. But with the harmonized sales tax mess in her lap, B.C.'s Christy Clark has apparently decided against an election this fall, according to recent reports, something she had been openly contemplating. Quebec's beleaguered Jean Charest, despite the election taunts of his foes, seems to be looking the other way, as well.

Incumbency, in the right hands, is a fine sword. But much depends on how it's grasped.

CBC producer Robert Sheppard is a former political writer and columnist for the Globe and Mail and Maclean's magazine.