Sunday, October 22, 2006

My thoughts on the final (?) Liberal numbers

Barring an appeal in BC, and allowing for campaigns to appoint people to fill vacancies, we now have the final delegate totals for the Liberal leadership:

Michael Ignatieff 1309 (30.2%)
Bob Rae 881 (20.3%)
Gerard Kennedy 751 (17.3%)
St├ęphane Dion 694 (16.0%)
Joe Volpe 211 (4.9%)
Ken Dryden 187 (4.3%)
Scott Brison 153 (3.5%)
Martha Hall Findlay 38 (0.9%)
Undeclared 109 (2.5%)

In my earlier post on this, I already pointed out how factoring out Kennedy's poor showing in Quebec shows that he dominates in a very strong second in the rest of Canada. I won't rehash that here.

There some good news in this for Ignatieff and Rae - in that they both broke through the psychological barriers of 30 and 20 percent respectively and for Kennedy who in widening his lead of Dion makes it more likely he'll survive the second ballot making him the come-from-behind contender and Dion the king maker.

Also in my earlier post, I pointed out how in the U.S. - where their primary/caucus system bares a resemblence to the Liberal proportional delegate system - the media and momentum does not depend on number of delegates but instead on states won. I was thinking through why this is and I believe it is due to electability. In the U.S. system it doesn't matter how many votes you win in the general election, it matters how many states you carry to raise your electoral vote total.

Similarly, in Canada it doesn't matter how many votes you win, but how many seats. As we saw just a few weeks ago in New Brunswick, you can win less votes and more seats. So while delegates will decide the next leader, the main reason why I like the delegate system as opposd to a one-member-one-vote system is because it gives the leadership candidates a trial run at an election campaign. They have to do well in many ridings across the country to get enough delegates to win.

So, as a result, I thought like the U.S. counts states won to measure success not delegates, I would count ridings won to measure success as everyone else is covering counting delegates.

In 27 ridings there were ties (mostly two way, but there were three three-way ties and a four-way tie). In these cases I have allotted 0.5, 0.33 or 0.25 ridings to a candidate as appropriate.

The results

Ignatieff won 100 ridings
Kennedy won 54 ridings
Rae won 49 ridings
Dion won 43 ridings
Volpe won 11 ridings
Dryden won 9 ridings
Brison won 8 ridings
Hall Findlay won 1 riding
And the "undeclareds" carried 5 ridings

When I factor in the ties and round them off, we end up was a 309 seat Liberal House of Commons if this were an election and we had the Ignatieff Party, Kennedy Party, etc.

Ignatieff 100 + 5.92 ~= 106 seats
Kennedy 54 + 7.92 ~= 62 seats
Rae 49 + 5.25 ~= 54 seats
Dion 43 + 4.08 ~= 47 seats
Volpe 11 + 3 = 14 seats
Dryden 9 + 0.83 ~= 10 seats
Brison 8 + 0.5 ~= 9 seats
Hall Findlay 1 + 0.5 ~= 2 seats
Undeclared = 5 seats

Here is what the Liberal House of Commons would look like. As the winner I gave Ignatieff Liberal red and as the longest running Liberal in the field I gave Kennedy the Wikipedia Liberal colour. I couldn't resist giving Rae NDP orange and Brison Tory blue. Dion got the Bloc's colour as the only Quebec candidate. Finally I rounded it off by giving Dryden purple, Volpe green and Hall Findlay yellow. The undeclareds get independent grey.

So although the Ignatieff Party is the clear winner, his minority is far smaller than Harper's and barely bigger than the current Liberal opposition. He is far ahead of his nearest competitor but the Kennedy Party or Rae Party could both form coalitions with smaller parties to make a larger minority, while the Kennedy Party could form a majority in coalition with smaller parties. A Kennedy-led coalition of all parties in opposition would form a massive majority of 202 seats. A Kennedy Party led coalition without the Dion Party would control a bare majority of 155 seats, while a Kennedy-led coalition without the Rae Party would form a large minority of 148 seats. An Ignatieff-Kennedy or Ignatieff-Rae coalition would form a majority, while any other majority coalition combination for the Ignatieff Party would require 3 or more parties.

Obviously there is not going to be a scenario where all other parties fold and the Liberals split into 8 factions behind each of the leadership candidates. However, considering that, at the end of the day, the main purpose of parties and leaders are to win elections under our first past the post electoral system to implement the policies we believe in, the above analysis gives us the best measure to see which leadership candidate (or coalition of candidates) have the best shot of doing so. Moroever, the Americans who have been using a system like this for far longer than us use the "winning in a real election" method to handicap the race so there is no reason why we shouldn't.


JL said...

Mr. Kennedy would be wise to run in the by-election for the seat in London.

He has good support there (from what I've seen) and if he were to win a seat in Parliament on the eve of the Liberal convention..whoa, talk about serious momentum.

Harrap said...

Interesting idea JL -- that could give Kennedy momentum. However, such a by-election could occupy alot of his time and take that away from his leadership campaign.

Great analysis NBPolitico, I never thought of it from a mock Parliament perspective -- you're right, the leadership convention does serve well as a trial general election and helps leadership candidates familiarize themselves with the dynamics in the different ridings.

petroom said...

If you can't vote for Kennedy in London, go have some fun here:

Anyone-But-Iggy Web Poll

YOu might Like It!

Anonymous said...

Am I the only person who would've changed my delegate support based on the fact that Ignatieff said in the debate that he'd recognize Quebec as a "nation"? I'm a big-L and a small-l liberal, but that would be a deal-breaker for me. Nobody who would drag us closer to separation, or even back into the debate, by whatever means, would ever get my vote, as a delegate or regular voter. Will this have no impact on his lead at all?

nbpolitico said...

anon - I would certainly hope so. If the separtists are to win in Quebec in the next election and move towards a referendum another disappointment over what amounts to a meaningless constitutional change would do it.

Anonymous said...

Bob Rae will win the Liberal leadership race and he will be Canada's next Prime Minister. He will send George W. Harper packing.

Anonymous said...

Both Rae and Iggy will be gone after the second ballot. It will be Kennedy or Dion after that.

Anonymous said...

Kennedy blinked by not taking a shot at the London by election seat. Shades of Allison Brewer. Kennedy is toast.

nbpolitico said...

anon at 6:36

I could not disagree more.

Had Kennedy run in that by-election he would have dramatically reduced his chances of winning the leadership.

Had he run, he risked ending up in a nomination showdown with local candidates which would be bloody and given him bad press. He may even have lost the nomination which would have destroyed his campaign.

Had he won the nomination, this is going to be a close race. He would have had to work full time at the by-election to win. He would have gotten good press, however this is not the part of the race where press will carry you through.

Had he lost the by-election, he would have been toast, had he won he would have had some superficial momentum but he would not have had the opportunity, like the rest of the leadership candidates, to work the delegates across the country and reach out for their support.

Harrap said...

As usual, I fully agree with you NBPolitico!