Tuesday 28 October 2014

Ottawa Election Evaluation


As I mentioned below, I was underconfident in many of my predictions for winners, and overrated the chances for the minor candidates.  In particular, here are the actual probabilities for each of my forecasted classes:

  • All 9 candidates who I said had a 95-100% chance of winning won.
  • All 6 candidates who I said had a 70-80% chance of winning won.
  • 4 out of 5 (80%) of the candidates who I said had a 55-60% chance of winning won.  Katherine Hobbs was the exception.
  • 3 out of 7 (43%) of the candidates who I said had a 30-40% chance of winning won.
  • 2 out of 13 (15%) of the candidates who I said had a 20-25% chance of winning won.  
  • None of the 13 candidates who I said had a 10-15% chance of winning won.
  • None of the 22 candidates who I said had a 1-5% chance of winning won.
The calibration curve from this data is shown below.  The line represents a properly calibrated forecast, where the forecasted probability is the same as the actual probability.

I underestimated the probability that the leading candidates would win, as the dots are above the line for forecasted probabilities greater than 40% or so.  I similarly overestimated the chances that trailing candidates would win, as the dots are below the line for probabilities smaller than 40%.  To be fair, this analysis assumes the races are uncorrelated, which I never claimed and I don't think is true.  I think this may have been an unusually good year for incumbents, especially in comparison to, for example, 2010.
The chart below shows the margin of victory for forecasted winners, as a function of the forecasted win probability:

Obviously the margin of victory is increasing in forecasted win probability, which is good to see.  The biggest outlier is again Katherine Hobbs, the (60%, -25%) point.
In terms of a single number, the average Brier Score of my forecasts is 0.209.

The Power of Incumbency

Overall, the biggest lesson of this election was the power of incumbency; incumbents won in all but two races where they were running.  Candidates endorsed by/associated with previous incumbents also did well. In particular, Catherine McKenney in Somerset, Riley Brockington in River, Jean Cloutier in Alta Vista, George Darouze in Osgoode, and Michael Qaqish in Gloucester-South Nepean all won in crowded races after being endorsed (or associated with, in the case of Qaqish), previous incumbents.

Ottawa Election Reflections

My predictions for city councillors seem to have been reasonably accurate.  In total, of the candidates who I said had the highest chance of winning, only 2.5 did not win.  In particular, Jeff Leiper won in Kitchissippi when I expected Katherine Hobbs to win.  Michael Qaqish won in Gloucester-South Nepean instead of Susan Sherring.  Jody Mitic won in Innes, when I was unsure if he or Laura Dudas was more likely to win (so I was half-wrong).  Overall, I think I was a bit underconfident in my predictions; I could have given higher percentages for the candidates I thought would win.
Going through ward-by-ward

  • Mayoral Race: Jim Watson won solidly, as expected.
  • Ward 1 Orléans: Bob Monette solidly as expected.
  • Ward 2 Innes: Jody Mitic won by a larger margin then I might have expected.  I thought this race might be closer.  Also, I may have overweighted Laura Dudas and some of the minor candidates.
  • Ward 3 Barrhaven: Jan Harder won solidly, as expected.
  • Ward 4 Kanata North: Marianne Wilkinson won by a fair margin, about as expected.
  • Ward 5 West Carleton-March: Eli El-Chantiry won by a fair margin, about as expected. Jonathan Mark did a bit better than I might have thought
  • Ward 6 Stittsville: Shad Qadri won by a smaller margin than I would have guessed
  • Ward 7 Bay: Mark Taylor won a fairly close race, about as expected
  • Ward 8 College: Rick Chiarelli won solidly, as expected
  • Ward 9 Knoxdale-Merivale: Keith Egli won solidly, as expected
  • Ward 10 Gloucester-Southgate: Diane Deans won by an even larger margin than last time. I may have been underconfident here.
  • Ward 11 Beacon Hill-Cyrville: Tim Tierney won solidly, as expected. I overestimated Francesca D’Ambrosio, who didn't even finish second.
  • Ward 12 Rideau-Vanier: Mathieu Fleury won by a larger margin than I expected.
  • Ward 13 Rideau-Rockcliffe: Tobi Nussbaum won by a larger margin than I expected (especially over Peter Clark). I may have been underconfident here.
  • Ward 14 Somerset: Catherine McKenney won by about the margin expected. Martin Canning, who didn't put up any signs, placed third.
  • Ward 15 Kitchissippi: Jeff Leiper won by a large margin. I am not so much surprised by the fact that he won as I am by the margin he won by.
  • Ward 16 River: Riley Brockington won by a fair margin. I may have been underconfident here, and given too much weight to the minor candidates.
  • Ward 17 Capital: David Chernushenko won solidly, as expected.
  • Ward 18 Alta Vista: Jean Cloutier won by a fair margin, about as expected.
  • Ward 19 Cumberland: Stephen Blais won solidly, as expected
  • Ward 20 Osgoode: George Darouze won by a fairly small margin. My biggest mistake was underestimating the chances of George Wright, who finished second, despite not even being on my list of candidates with an appreciable chance of winning.
  • Ward 21 Rideau-Goulbourn: Scott Moffatt won by a fair margin, as expected.
  • Ward 22 Gloucester-South Nepean: Michael Qaqish won by a fair margin. I may have miscalled this race; Susan Sherring, who I though would win, finished third.
  • Ward 23 Kanata South: Allan Hubley won by a huge margin, exactly as expected.

Saturday 25 October 2014

Ward 23 Kanata South

Allan Hubley 100%
Allan Hubley is facing no serious challengers in this race.  He will win.

Ward 22 Gloucester-South Nepean

Susan Sherring 40%
Michael Qaqish 25%
Jason Kelly 20%
Scott Hodge 15%

The race in Gloucester-South Nepean is a complicated one with four major candidates and no incumbent.  Susan Sherring has high name recognition, and has been endorsed by Bob Chiarelli, which is a sign of seriousness.  Michael Qaqish worked in the office of the outgoing councillor, which serves as a quasi-endorsement.  The other two major candidates, Jason Kelly and Scott Hodge, could well win; it is hard to tell with such a crowded race.
UPDATE: downgraded Qaqish and upgraded Kelly

Ward 21 Rideau-Goulbourn

Scott Moffatt 80%
Daniel Scharf 20%

Scott Moffatt, who won in the last election with 52% of the vote to solidly defeat incumbent Glen Brooks with 26% of the vote, is running against Daniel Scharf.  Scharf does not seem to be a very strong candidate, and Mofatt will probably win.  The wild result in the last election makes me less confident of this outcome.

Ward 20 Osgoode

George Darouze 30%
Liam Maguire 20%
Mark Scharfe 10%
Jean Johnston-McKitterick 10%
Davis Jermacans 10%
Tom Dawson 10%
Kim Sheldrick 10%

Osgoode is a crowded race with at least 7 candidates with a reasonable chance of winning and no incumbent.     George Darouze is probably the frontrunner, as he has been endorsed by the previous incumbent, Doug Thompson.  Liam Maguire has name recognition, which may help him in a crowded field.  Mark Scharfe finished second in the last election with 23% of the vote, and is the leader of a slate.  The other candidates all seem the same, but one of them could win.

Ward 19 Cumberland

Stephen Blais 80%
Marc Belisle 20%

This ward is a bit of a “suburban vs. rural” battleground.  In the last election, Rob Jellet, the “rural” candidate and the incumbent, was defeated by Stephen Blais, the “suburban” candidate (52% to 43%).  In this election, the new rural candidate is Marc Belisle.  Given the growth of South Orleans in this ward, Stephen Blais’s base has increased, and he now has the advantage of incumbency.  E will likely win this time as well.

Ward 18 Alta Vista

Jean Cloutier 55%
Clinton Cowan 35%
Hussein Mahmoud 5%
Perry Marleau 5%

Jean Cloutier has been endorsed by the previous councillor, Peter Hume, which makes him the clear favorite.  Clinton Cowan got 20% of the vote in the last election, and so is a reasonable contender.  Hussein Mahmoud and Perry Marleau also seem to have credible campaigns, but they likely won’t win.
UPDATE: Mahmoud has a lot of signs, but few are on front lawns.  Cloutier and Cowan seem fairly evenly matched in terms of signs on front lawns.  I am slightly less confident Cloutier will win.

Ward 17 Capital

David Chernushenko 95%
Scott Blurton 5%

Chernushenko won this ward in 2010 with 41% of the vote against a number of challengers.  This time, he is the incumbent, and he is facing only one serious challenger, Scott Blurton.   I expect he will win fairly easily.

Friday 24 October 2014

Ward 16 River

Riley Brockington 55%
Mike Patton 15%
Vanessa Sutton 10%
Antonio Giannetti 5%
Barbara Carroll 5%
Michael Kostiuk 5%
Jeff Koscik 5%


Maria McRae, the previous incumbent, has endorsed Riley Brockington, which is a strong positive for his campaign.  Brockington’s experience as OCDSB trustee should also be an asset.  Patton and Sutton are the other two major candidates.  Both could win.  Patton seems to have more signs than Brockington.
UPDATE: I am adding a bit of weight to Antonio Giannetti and some of the other minor candidates.  They all have signs up, and Giannetti in particular seems to have a lot of signs up, including on front lawns.  On the other hand, his website is not that impressive.
UPDATE: Downgrading Giannetti given that he does not seem to be able to speak English.

Ward 15 Kitchissippi

Katherine Hobbs 60%
Jeff Leiper 35%
Michelle Reimer 5%

Katherine Hobbs is in a close race with Jeff Leiper, with Michelle Reimer as the third major candidate.  The situation is not dissimilar to the last election, where Katherine Hobbs won with 45% of the vote versus Christine Leadman’s 40%.  The incumbency advantage should help Hobbs in this election though.  Also, vote-splitting between Michelle Reimer and Jeff Leiper should help Hobbs.

Ward 14 Somerset

Catherine McKenney 75%
Martin Canning 20%
Jeff Morrison 5%

Although this race does not have an incumbent, Catherine McKenney has been endorsed by the previous incumbent, Diane Holmes, and seems to have inherited her campaign apparatus.  Her main challenger, Martin Canning, does not have an signs, which is a big negative.  Jeff Morrison seems to be the third-place candidate, and might win if there is enough vote-splitting among the top two.

Ward 13 Rideau-Rockcliffe

Tobi Nussbaum 40%
Peter Clark 25%
Sheila Perry 20%
Jevone Nicholas 5%
Penny Thompson 5%
Cam Holmstrom 5%


Peter Clark won this race in 2010 with only 25% of the vote against a very crowded field.  He has not been a particularly impressive councillor.  This time, there are again 6 serious candidates, so there will be a lot of vote-splitting.   This probably helps Clark. Nussbaum seems to be the main challenger, and he has the endorsement of the councillor before Clark, Jacques Legendre.  Many of the other candidates could also win.
UPDATE: Nussbaum has the most signs in the ward, especially on front lawns.  Shelia Perry has some signs.  Peter Clark has a fair number of signs, but none of them seem to be on front lawns.  The other candidates have a very limited sign presence.  I am adjusting accordingly.

Ward 12 Rideau-Vanier

Mathieu Fleury 60%
Marc Aubin 30%
Catherine Fortin LeFaivre 10%

Fleury upset the previous incumbent, Georges Bédard, by a tiny margin.  In this season, he is facing two serious challengers.  Marc Aubin is his most serious rival.  Fleury will probably benefit from vote-splitting between the other two candidates.

Ward 11 Beacon Hill-Cyrville

Tim Tierney 95%
Francesca D’Ambrosio 5%
Tim Tierney won this race by a small margin over the previous incumbent, Michel Bellemare.  His only serious contender in this election seems to be Francesca D’Ambrosio, and she doesn’t seem to be an strong challenger (although she has a fair sign presence).

Ward 10 Gloucester-Southgate

Diane Deans 80%
Lilly Obina 15%
Other Candidates 5%

Diane Deans won in this ward in the last election with 48% of the vote against Obina’s 32%.  This election has a few more third candidates, which will probably help Deans due to vote-splitting.

Ward 9 Knoxdale-Merivale

Keith Egli 95%
Cristian Lambiri 5%

Keith Egli won the last election by more than 10% against a crowded field.  This time, he is facing only a single challenger, Cristian Lambiri, who, although serious, is not likely able to overcome the incumbency advantage.

Ward 8 College

Rick Chiarelli 99%
Guy Annable 1%

Rick Chiarelli won the last election with 66% of the vote against a more crowded set of candidates.  Despite his late start in terms of putting up signs, he is very likely to win again.  Guy Annable has a serious campaign (with signs), but it is hard to believe he can beat Chiarelli.

Ward 7 Bay

Mark Taylor 70%
Alex Cullen 30%

Mark Taylor won the last election with 37% of the vote in comparison to Cullen’s 30%.  The race in the last election was more chaotic, with Terry Kilrea and Shawn Little also receiving 8% and 6% of the vote respectively.  Given that most of the Kilrea and Little voters will gravitate towards Taylor rather than Cullen, and the significant incumbency advantage, I expect that Taylor will win in this election as well.  On the other hand, in the last election, voters may have been angry at Cullen for his last-minute decision to run for the ward as a “consolation prize” instead of the mayoralty.  That would no longer be true this year.

Ward 6 Stittsville

Shad Qadri 95%
David Lee 5%

Shad Qadri won this race last time with 91% of the vote.  Although David Lee is a serious candidate, it is hard to imagine he can overcome the incumbency advantage in this ward.

Ward 5 West Carleton-March

Eli El-Chantiry 80%
Jonathan Mark 20%
Eli El-Chantiry won this race in the last election with 77% of the vote.  This time, his main challenger is Jonathan Mark, who seems to be trying to represent the more rural parts of the ward.
Note: this prediction has been edited from the original version.

Ward 4 Kanata North

Marianne Wilkinson 60%
Matt Muirhead 25%
Jeff Seeton 15%

Marianne Wilkinson, the incumbent, is facing two serious challengers.  She won with 50% versus Seeton’s 45% in the last election.   The vote splitting that might happen between the two challengers in this election will probably be in her favour.

Ward 3 Barrhaven

Jan Harder 99%
Ian Bursey 1%

Jan Harder is almost certain to win this ward.  She won with 66% of the vote last time against stronger candidates.

Ward 2 Innes

Laura Dudas 25%
Jody Mitic 25%
François Trépanier 20%
Fred Sherwin 10%
Teresa Whitmore 10%
Chantal Lecours 5%
Chris Fraser 5%

Innes is one of the most chaotic races in the city.  There are 8 candidates with signs, no incumbent, and no frontrunner.  The previous incumbent, Rainer Bloess, has not endorsed any of the candidates.  The Citizen lists Laura Dudas, Jody Mitic, François Trépanier and Fred Sherwin as frontrunners, so I will give most of my weight to them.  Laura Dudas is probably one of the more likely candidates to win, due to her experience on the Blackburn Community Association.   Jody Mitic is an Afghanistan war veteran who is running a credible campaign.  François Trépanier seems to have more managerial experience than some of the other candidates.  Fred Sherwin finished second in Orleans ward in the last election, and thus should be taken seriously.   Teresa Whitmore was a former councillor in Belleville.  Chantal Lecours may have some name recognition from being a local French school board trustee.  Finally, Chris Fraser ran in the last election and finished third; if he can maintain or increase his vote share, he may squeak through.

Ward 1 Orléans

Bob Monette 95%
Jennifer Robitaille 5%

Bob Monette won last time with 60% of the vote against two serious candidates.  His main challenger this time is Jennifer Robitaille, who only captured 14% of the vote in the last election and placed in third.

Mayoral Race

Jim Watson 99%
Mike Maguire 1%

Jim Watson has this race locked up.   He got 48.7% of the vote in the last election against a stronger array of candidates.

Purpose of this Blog

I can’t find any forecasts of who will win in Ottawa’s municipal elections.  Thus, I am going to post my predictions for the mayoral and city council races.   I will give them in terms of probabilities.  For instance, my first prediction will be
Jim Watson 99%
Mike Maguire 1%
which means that I think that Jim Watson has a 99% chance of winning, and Mike Maguire has a 1% chance of winning.  It does not mean that I think that Jim Watson will get 99% of the vote.  Also, the politician I think will win is not necessarily the politician I want to win.  Note: These predictions are reposted from my Google+ account.
I might also use this blog to make predictions about other topics, from time to time (I promise not to talk about really boring topics like the Second Thomas Shoal).