Professor emeritus of political science at the University of Calgary and a former campaign manager for conservative parties.
The Conservative Party leadership race has a dozen candidates but no front-runner. Kellie Leitch's lead in public opinion polls means little, because the leader will be elected by party members, not the general public.
Election rules will exercise great influence on the outcome of the race. Each of the 338 constituency associations gets 100 points, divided among the candidates in proportion to their vote shares in that riding. The constituency associations differ greatly in number of members, but that doesn't appear to be a major factor in this race. More significantly, voting will be carried out by preferential ballot, in which members are invited to rank all candidates from top to bottom. The ballots will be recounted while trailing candidates will be dropped and their points transferred until someone receives 50 per cent + 1 of the 33,800 points.
This is similar in principle to the sequential voting in a traditional convention except that all the information is gathered on a single ballot. Any deals between candidates must be made ahead of time rather than on the convention floor.
One way to win is to go for broke, to aim for a first-count victory, which is how Stephen Harper and Justin Trudeau won. Ms. Leitch seems to be going this way with her demand for screening immigrants for Canadian values. Her position is getting a lot of criticism, which also means a lot of publicity, so maybe her strategy will work. But going for broke in a crowded field carries the risk that opposition within the party may coalesce around some other alternative, as happened to Bob Rae and Michael Ignatieff in recent Liberal leadership races.
The other main strategy for winning in a crowded field is to build alliances during the campaign. With so many candidates, some are bound to drop out for lack of support or because of inability to raise money. The offer of a critic's position after the campaign is over, or promising to help raise money to pay campaign debts, may bring a support-building endorsement from a failed candidate.
Candidates may also want to send out signals to supporters of other camps that they would be worthy of their second or third preferences. Such appeals might be implicit, as by speaking favourably of certain other candidates' policies; or they might be explicit, as in a final mailing request for second or third preferences.
Candidates might also consider striking formal alliances by "exchanging preferences," as is done in Australian elections, which are carried out with ranked ballots. Two candidates could agree to recommend to their supporters to give their second preferences to each other. Stéphane Dion and Gerard Kennedy did something like this in the 2006 Liberal leadership race, in a delegated convention with sequential voting. Each reportedly agreed to throw his influence to the other if he was dropped from the balloting. Mr. Kennedy finished behind Mr. Dion on the second ballot, and Mr. Dion went on to win.
The mechanics would differ in the Conservative race because all voting takes place in advance. Alison Redford gave an elegant example of how to do it in the 2011 Progressive Conservative leadership race in Alberta. Shortly before the final voting, she announced that Doug Horner would be her second choice. He reciprocated, and she won on second-count transfers.
Preferential voting helps to build consensus around the winning candidate. Unfortunately, however, the qualities needed to build internal consensus do not always lead to victory over other parties. Mr. Dion lost big. Ms. Redford won an Alberta election but lasted only two years as premier. In contrast, Mr. Harper and Mr. Trudeau proved to have the "right stuff." The Conservatives have to hope that a consensus winner in an internal party race can also become a winner in the larger arena of national elections.