GraphicMatt.com Archived MetroNews.ca columns and blog posts by Matt Elliott

# Toronto City Council Electability Index: can math predict council elections?

This probably sounds like a ridiculous thing to ask, but bear with me a second. Can math help predict Toronto's 2014 council elections?

I’ve spent some time this week considering that question. Mostly because I’m a gigantic nerd. But also because one of the things I asked for in my Christmas column was a bigger focus on the 44 council races that’ll take place across the city this year. After all, this election is about more than the mayoral horse race with its exciting transit announcements and nightclub appearances. But before anyone can really delve into individual council races, it’s important to figure out which councillors are most vulnerable to electoral challenges. That’s where the math comes in, and that’s what led to the creation of what I’ve called the 2014 Toronto City Council Electability Index.

This wasn’t at all an exact process. It’s more of an experiment. Nobody does approval polling on individual councillor popularity, so I’m limited to a pretty narrow range of more-or-less objective data when trying to figure out which sitting councillors might be ditched by voters later this year. But the results are interesting, and may just provide a good starting point for thinking about the direction council will take post-2014.

### The 2014 Toronto City Council Electability Index

With no opinion polling available, I used five data points to determine an electability score for all 44 members of city council:

• Margin: The number of percentage points over which a winning candidate beat their second place opponent in the 2010 municipal election.
• Share: The percentage of the total vote won by the winning candidate in 2010.
• Ford Margin: Did Mayor Rob Ford win this ward, and by how much?
• Ford Nation Percentage: From my City Council Scorecard, how often a councillor voted with Ford on major issues brought to council this term. Percentage is current as of December 31, 2013.
• Turnout: Of all eligible voters, what percentage of them came out to vote in the 2010 council election in their ward?

All those data points get fed into a complicated formula that weights all of the factors. If a councillor beat their closest opponent by more than 40 percentage points in 2010, they’re awarded a maximum of 20 points. They also get points if their share of the popular vote was over 35%. But, conversely, they lose points if their share was below 35%.

I also used the Ford Nation Percentage as a comparator with the mayoral result in the ward. So if the residents voted strong for Ford, but then their councillor went and voted against him on most issues, they lose a few points off their electability score.

Finally, there’s turnout, which is a simple calculation. Across the city, the average turnout in all wards was 50.27 per cent. Councillors gain electability points if their ward turnout was higher than that. They lose points if their turnout was lower. The idea being that low turnout indicates that someone could successfully mount a challenge against an incumbent by engaging the disengaged.

The theoretical maximum Electability Score is 116. Coun. Doug Ford — who says he isn’t running but sure could win if he tried — scored the best at 48.55. Coun. James Pasternak scored the lowest at -8.61, due to a very low percentage of the vote share in his ward (19.16 per cent) and low voter turnout.

For visual reference, I colour-coded the last column. Councillors with an electability score higher than 15 are green because they’re basically safe. Councillors scoring between 5 and 15 are orange because they should be a bit worried. While councillors below 5 are red because they’re in the danger zone.

The algorithm does tend to be biased against first-term councillors who emerged from crowded fields — keep that in mind when considering, for example, Coun. Kristyn Wong-Tam's score — but there are a number of names toward the bottom of the list that I do think are vulnerable to challenges. (We’ll ignore Coun. Peter Leon, who was appointed and really shouldn’t run this year.)

Coun. Pasternak is a wild card because he won with less than a fifth of the vote in 2010. The man is a walking advertisement for ranked ballots. Just above him — somewhat surprisingly — is Coun. Frank Di Giorgio, who serves as Ford’s budget chief. He’s a veteran councillor, but the algorithm bets against him due to low turnout and a weak showing in 2010.

Long-serving North York reps Coun. Anthony Perruzza and Coun. Maria Augimeri also face risks, owing to the tough challenges they faced in 2010 and because they’re both left-leaning councillors in areas that went hard for Ford.

Also vulnerable? Rookies Coun. Josh Colle, Coun. Gary Crawford and Coun. Vince Crisanti. They all lose points due to turnout. It should be noted that Coun. Crisanti won with the lowest voter turnout in any ward in 2010, at just 43.37%.

Rounding out the danger zone is Coun. Cesar Palacio, who has been around for a while but faced a tough challenge in 2010 in a race that saw voter turnout at 44.99%. The math shows he could be defeated by a dedicated opponent.

But, like with all council incumbents, the math doesn’t mean anything if no one steps up and makes the challenge.

This post was originally published at http://www.metronews.ca/views/toronto/ford-for-toronto-matt-elliott/2014/01/15/toronto-city-council-electability-index-can-math-predict-council-elections.html on 2014-01-15T00:00:00.000Z