Ontario Election Recap: Very Few Surprises
Well with the 2007 Ontario election (what we have termed the “much-ado-about-nothing-election”) now over, it’s time to recap what happened. The general conclusion is that there were few surprises.
We estimated the Liberals at 42.5% support, they got 42.2% (-0.3%). We had the PCs at 32.2%, they got 31.7% (-0.5%). We had the NDP at 17.3%, they got 16.8% (-0.5%). We have the others (including Greens) at 8.0%, they got 9.3% (+1.3%). The Greens held more of their vote than we estimated, which wasn’t overly surprising since they picked up the protest vote. We estimated the Liberals at 65 seats, they got 71 (+6). We had the PCs at 29 seats, they got 26 (-3). We had the NDP at 13, they got 10 (-3).
We correctly called 92% of the ridings (98 of 107). See our SCORECARD. Of the 9 we got wrong, 7 were in our “too close to call” category (which meant they could have gone either way), the other 2 were within 3-4%. So I was surprised at just 2 ridings — Sarnia-Lambton and Kitchener-Conestoga, but even here only moderately. The PCs ran a very young candidate in the latter and the Liberals were hit because of cost overruns for a new hospital in Sarnia. So nothing too surprising. I knew the Liberals would be hit hard in the North, to the NDP’s gain, which was the case. Although no seats changed hands, I was surprised at how close Thunder Bay-Atikokan (although this wasn’t overly surprising) and Timinscaming-Cochrane were (and how not close Sault Ste Marie was). Otherwise, virtually every riding played out as expected, although in general where the Liberals won, they won by a larger margin that we expected. Clearly, many conservative supporters stay home, or registered a protest vote with the Greens.
The regional results were as expected, within 1 or 2 points of our projections: In the North, we had the Liberals at 41-44%, they got 41.9%. We had the NDP at 31-34%, they got 34.2%. We had the PCs at 15-18%, they got 18.3%. So the Liberals were on the low end, and the PCs and NDP on the end of the range. In Toronto, we had the Liberals at 46-49%, they got 45.3%. We had the PCs at 22-25%, they got 23.5%. We had the NDP at 21-24%, they got 21.6%. And so on across all regions.
Turnout was, as expected, very low — the lowest ever, and likely would have been even lower if not for the referendum, which failed as expected. The result after the election looked pretty much exactly like what it did before. The NDP gained nothing from when the legislature was dissolved, the Liberals gained 4 seats and the PCs gained 1, both due to the expansion of the legislature. As we said, this election was much-ado-about-nothing. And indeed, it was.
Join DemocraticSPACE’s Election Night Coverage
Join us as we complete our DemocraticSPACE scorecard as ridings are called throughout the evening. Greg Morrow will also be commenting on interesting developments and surprises in the comments below (and make no mistake, there are always surprises!). Join the discussion below, or click the riding name to join the individual riding discussions. Our coverage will begin at 9pm, when the polls close. Best of luck to all candidates!
Will Weather Be a Factor?
Here is the current satellite image of Ontario. Will weather be a factor and if so, who does that benefit?
ONTARIO
SOUTHERN ONTARIO
As you can see, it is raining in the Kitchener-Waterloo region, Sault Ste Marie, Sudbury, Orillia, Cornwall, and Toronto. This might make an already light turnout even lighter. And I would guess that this does not benefit the Liberals, who count of masses of people showing up at the polls. We shall see soon…
Which Ridings Could Change Hands Tonight?
What everybody wants to know is which ridings will change hands (i.e. switch from one party to another). Here are the seats that we have changing hands, but please note that each of these is so close that any or none of these could change hands:
We don’t have these changing hands but, again any or none of these could. That is, I would not be surprised to see any of the following change hands:
Note: some of the ridings are new so they don’t have an incumbent; in these cases, the party listed under “from” had the most votes from 2003, as transposed into the new riding.
General Notes About Our Projections
So by now, we hope you’ve all taken the time to look at our projections:
It is important to note that our projections are based on aggregate of the latest polling information from five pollsters: Ipsos-Reid (with whom we had a partnership throughout the election), SES, Harris-Decima, Strategic Counsel, and Environics. And while each shows the Liberals with a comfortable lead, there is disagreement about the magnitude of that lead. Are the Liberals polling 42% or 46%? It makes a big difference. Are the PCs polling 27% or 32%? Again, big difference. Are the NDP at 17% or 20%? Are the Greens at 3% or 11%? As you can see, there is a range. Since our projections are based on weighted averages of all polls, our numbers tend to fall somewhere in the middle of these extremes.
But another word of caution is in order: polls tell us only the recent past, they do not predict the future, and do not account for intangibles. They do not account for the particular candidates in your riding, or particular issues that might sway voters one way or the other. As the government is formed from the results of one-on-one local battles, we don’t always know how that will translate into seats. That’s why DemocraticSPACE exists — to provide the “best guess” of what we think will happen in each riding. But, it’s important to note that even our aggregate of all available polling data still has a 3% range. So, in very close ridings — and there are 17 ridings where our projections are within that 3% range — the polling information can only tell us that it will be close. These battles will be won by boots on the ground. We’ve gathered as much information as we can from objective sources on the ground in making our “best guesses” for these close ridings, but invariably some will be wrong. That’s the nature of projections.
But more importantly, polls tell us party support free of any constraints (”who do you support?” or “who would you vote for?”). But the real world has constraints. If your preferred party has no chance of winning in your riding, you might vote for someone else who does. This impacts the smallest parties the most. So, you might expect the Greens to lose support at the ballot box (unless they receive the lion’s share of protest votes), since their support is widespread, not concentrated enough in specific ridings to win seats (unless Shane Jolley pulls off a miracle in Bruce-Grey-Owen Sound, which frankly, I hope they do, then at least every person who voted Green has at least one voice in Queen’s Park). Likewise, since the NDP is competitive in only about 1/4 of the ridings, they often lose support in the other 3/4. So the incidence of “strategic” or “tactical” voting is real, and does impact party votes, if however small.
Another factor is turnout. Which party’s supporters will actually get out and vote? We know that 55 years-old + voters tend to have much higher turnout than does the 18-25 age bracket. Since the 55+ crowd supports the PC Party more, we might expect their vote to register higher than what the polls are telling us. Likewise, the 18-25 crowd is more likely to support the NDP and Greens (and to some extent the Liberals), so it’s likely some of that support will not be registered on election day. Yet another factor is the referendum. Will the chance to change the system help get out the vote for the NDP and the Greens? Will opposition to change get more Liberal and Tory elites out? The question whether opposition to change is a stronger motivation to vote when you otherwise wouldn’t, more so than the hope for change? I’m guessing the pro-MMP folks are more motivated, but the question is how many of them are there? I’m guessing there aren’t that many who are that passionate so make a huge difference.
Then there are the issues themselves. The Tory proposal to fund religious schools was rejected by many Conservatives. There is some reason to believe that some PC supporters will choose to sit this one out. How Tory’s reversal on this 9 days before the election plays out is uncertain — some might have come back into the fold, but others might sit out because Tory undermined his leadership campaign. We also know that social conservatives are not exactly in love with John Tory, as they see him as too socially liberal. They too might sit out (or vote for the Family Coalition Party). Some might simply want to force Tory out, so will not go out and support his party. Likewise, there are warning signs for the Liberals. Despite the Liberals polling in the 42-46% range, Dalton McGuinty’s personal support levels trail his party support by a wide margin, typically in the 30-35% range. The Liberal faithful may also not be too excited about this election. Even some NDPers are not entirely on board with their party. Some are disappointed with the party’s decision to support the status quo in education funding, preferring instead a position similar to the Greens (to create a single public secular education system). Will some of their support bleed to the Greens? So there are many variables that impact the difference between stated support and actual registered votes.
Another thing to consider is the votes for independents and small parties. These are typically not picked up in the polls. But we know that these accounted for about 1.35% of votes in 2003. This will likely increase this time since the Family Coalition Party is running 83 candidates instead of the 51 they ran in 2003. The Libertarians are running more candidates too. And all others and independents are about that same. So it’s likely this will rise to 1.7 or 1.8%.
One other thing: our projections are built largely from the “bottom up” not from the “top down”, which means that we look beyond the province-wide support levels and place more weight on the regional polling data. It’s fine to say the Liberals are at 42%, but if they get 60% in Toronto, it overestimates their support relative to the actual seats they can win. So the distribution of support is what’s important. Unfortunately, regional polling data is less reliable because the samples are small. Again we aggregate samples and do weighted averages. Another problem is that pollsters use different boundaries for their regions, so we have to make adjustments, which are necessarily imperfect. For example, different pollsters have different definitions for what is included in “Central Ontario”. And in some cases, in Hamilton-Niagara, for example, we have scant polling data because it is typically merged with other regions (either “Central Ontario” or “905″). The North is always a challenge because with a small population, the samples are usually very low. So accordingly we have less confidence in our North and Hamilton-Niagara projections. Fortunately, we have good data for the Southwestern, Eastern, 905 and 416 regions.
We’ve made our “best guesses” to account for what happens at the ballot box in our final projections. For comparison, it is useful to look at the difference between the final polling averages in 2003 and the actual votes.
COMPARISON OF 2003 ELECTION: POLLS VS ACTUAL
|
PARTY |
FINAL POLL AVG |
ACTUAL RESULT |
DIFFERENCE |
 |
LIBERAL |
48.8% |
46.5% |
-2.3% |
 |
PC |
31.0% |
34.6% |
+3.6% |
 |
NDP |
16.2% |
14.7% |
-1.5% |
|
OTHERS |
3.8% |
4.2% |
+0.4% |
As you can see, the Liberals and NDP dropped while the PCs gained. Hidden in the “other” is the likelihood that the Greens dropped but the small parties, since they were not picked up in the polls, gained.
We think we’ll see a similar trend this year, but we don’t think the magnitude will be as great. Here is a comparison of the actual observed final poll averages and our adjustments to them.
|
PARTY |
FINAL POLL AVG |
ADJUSTMENT |
PROJECTED |
|
LIBERAL |
43.2% |
-0.7% |
42.5% |
|
PC |
30.5% |
+1.7% |
32.2% |
|
NDP |
18.3% |
-1.0% |
17.3% |
 |
GREEN |
8.0% |
-1.7% |
6.3% |
|
OTHERS |
- |
+1.7% |
1.7% |
So, as you can see, we’ve made adjustments to the final polling figures. The Liberals drop by 0.7%, but as you can see from the 2003 results, the potential for a larger drop is high, which would lead to a weaker majority. The PCs gain by 1.7%, about half of its 2003 ballot box gain; again, this if we see a gain similar to 2003, then Liberals could barely win a majority. The NDP drops 1.0%, smaller than their 2003 drop. And, of course, it’s also possible that the ballot box adjustments could swing the other way. All of this is to say that even when there appears to be a consensus in the polls — in this case, showing a strong Liberal majority — there is the potential that it could break differently.
That’s why we provide ranges. Here are the best and worst case scenarios for each party as we see it:
|
PARTY |
SEATS |
SUPPORT |
|
LIBERAL |
56-72 |
41.0-44.0% |
|
PC |
25-38 |
31.1-33.3% |
|
NDP |
9-15 |
16.7-17.9% |
 |
GREEN |
0 |
5.9-6.7% |
|
OTHERS |
0 |
1.6-1.8% |
So, when you sit back and watch the results come in tonight, keep in mind that nothing in ever certain. We’ve done our best to anticipate the range of possible outcomes, but in the end, as always, it all depends on you, the voters.
Referendum: If You Don’t Know, Leave It Blank
We have a fundamental problem with the Ontario referendum: most people don’t know what the heck mixed-member proportional (MMP) is. If you don’t know, don’t vote for the status quo (first-past-the-post). DemocraticSPACE encourages you to simply leave it blank. How can you vote against something that you don’t know is better or worse?
If you like the idea of electoral reform, but don’t like this particular model, leave it blank. If you think we just need more time to decide, leave it blank. Everyday citizens like you recommended this change, but if you’re not sure about it, don’t endorse politics-as-usual (what the party bosses want you to do), just leave it blank. Every vote for first-post-the-post is a vote for politics-as-usual.
MMP: The Ultimate Protest Vote
Voters today have the ultimate protest vote. A vote for MMP sends a clear signal that you are not happy with politics as usual. It’s hard to believe that people would not seize this opportunity, what with trust in politicians running at about 10%. So when you vote today, remember that a vote for mixed-member proportional (MMP) as recommended by your fellow citizens, is a vote against the establishment, against the backroom deal-makers, and against the kind of bickering politics that is only getting worse. A vote for MMP will send a clear signal to the politicians that that politics as usual is not good enough, that we can and we must do better.
Well, That Was Much Ado About Nothing!
Well, Ontarians head to the polls today after what can only be described as much ado about nothing. Truly, this has to be one of the most content-free campaigns ever conducted in Ontario. The entire campaign was dominated by an issue that wasn’t even an issue until John Tory raised it — public funding for private religious schools.
Issues that were on Ontarians minds were barely discussed — how to curb greenhouse gases, how to deal with traffic congestion in the GTA, how to get more family doctors, how to shift to a more sustainable energy production, how to fix the education funding formula, how to fix municipal downloading, how to address the growing problem of affordable housing and homeless, how to tackle our embarrassingly high incidence of child poverty, how to ensure all regions have jobs and opportunities, and so on.
The referendum? What referendum? Elections Ontario flubbed the public education campaign so badly that the majority of people voting today either have no idea there is a referendum or have no idea what the alternative is. That information void was filled by party backers who stood to lose under MMP, spreading blatant disinformation about MMP through email and their major media friends happy to assist. MMP never had a chance and so I suspect the party elites have dodged a bullet this time. Their power is safe for now.
It has been said that campaigns are not the right time to deal with serious issues. Hogwash. The proposed solutions should not be cooked up on the campaign bus, they should be studied and prepared in advance, but absolutely campaigns should be the time to debate different solutions to pressing problems. Unfortunately, this campaign wasn’t about that. It was, as is so often the case these days, a product of carefully scripted nothingness. Say nothing of substance, do nothing out of the ordinary. Just get by.
Consequently, the electorate is in a precarious position. People are not that excited by the incumbent Liberals (the usual partisan folks, of course, being of course an exception). But people also don’t particularly like the alternatives. John Tory hasn’t offered much that’s different from the Liberals — indeed his whole campaign was about him (”leadership matters”), which more or less imploded when he caved in under pressure about the religious schools issue (which of course was so badly out of step with the electorate it was doomed to fail). The NDP did its thing for “working families”, a tired euphemism for the working poor. It’s narrow focus (6 commitments, all sensible) were effectively the terms for a minority government, but if the Liberals win a majority (and we think they will), what then? Out of this ideas vacuum may emerge the Greens, who will surely gain in support. They have probably forwarded the most new ideas this campaign, even if most people didn’t pay attention, but unless Shane Jolley pulls off a miracle in Bruce-Grey-Owen Sound, they will likely be shut out.
With such widespread apathy towards the parties, anything is possible, as either the NDP or the Greens could gain a healthy dose of protest votes. But we’re betting that the only thing people remember about this forgetful campaign is that the Tories want to fund religious schools and the Liberals don’t. So, we expect that by the end of the night, we’ll have a Liberal majority government, almost by default. Turnout will be at an all-time low, which is ironic and somewhat sad, since this is the one election, with a referendum on the system itself, that gave voters the ultimate protest vote. And even that, sadly, has been much ado about nothing. Just how the parties like it.
Why Should Liberals Vote for MMP?
Why Should Liberals Vote for MMP (mixed-member proportional) in the referendum? Because it is in their long-term interest to do so. With the Conservatives circulating an email urging its supporters to reject MMP, Liberals should be asking themselves “if the Conservatives don’t like it because they stand to lose their inherent seat advantage in Ontario, shouldn’t we, as Liberals, be supporting MMP?” And if you look at Ontario’s election history, the answer is a clear and unequivocal “yes, MMP is good for the Ontario Liberal Party”. In fact, since 1975, the Liberals would have won more seats under MMP than under the current system in 7 of 9 elections (see analysis). That’s right, the Liberals would have gained under MMP almost 80% of the time. Likewise, the NDP would have won more seats under MMP the same 7 of 9 times. The Conservatives would have won fewer seats 6 of 9 times, so you can see why they are circulating an email urging its supporters to vote against (so much for the parties not trying to influence to outcome of the referendum). Indeed, at 37% each, the Conservatives would win 8 more seats than the Liberals. So what do you say, Liberals, do you want to eliminate the PC party’s inherent seat advantage under our current system? Then vote for MMP.
Stay Tuned for Revisions
In light of a couple of new polls out today, I will make be making some revisions to our final projections. Stay tuned. -GM
