Who Can Stop Doug Ford?
Today’s Ipsos poll confirms last week’s polls. Across Ontario the PCs are at 40%, NDP 35%, Liberal 22% and others 3.5%. In the past two weeks the PC are steady or stalled, depending on your perspective, but numbers at these levels can produce a majority. The NDP is now firmly and decisively established in second place as the chief, and perhaps only, rival to Ford.
The strategic considerations for the parties and the voters had changed dramatically by the end of the first week. Liberals now face a tough decision. To stop Ford, do they move their votes to the NDP? Or, do they still have a chance of defeating PCs? For NDP supporters, how to appeal to Liberals is straightforward. But how do they appeal to soft PC votes? To which ridings should they shift resources? Constituencies like Durham or Peterborough or all of the Brampton ridings come into play on these numbers. They will likely make the difference between a Ford or Horwath government.
How do you assess the state of play in your riding? Today’s numbers are so different from the 2014 election that they can’t provide any definitive answers. Working with those results, however, is still the best place to start.
In Britain since the end of the Second World War, election experts have developed the “swing-o-meter”. In its most primitive form, it takes the national swings from the previous election and applies them to each constituency. For example, if Labour was up 5% nationally, the Conservatives were down 5%, and the Liberal Democrats and others posted no change, we could conclude a five percentage point national swing to Labour. You can then apply that swing to to the previous election results in your constituency to produce a prediction.
Using this method in the Ontario election, the PCs are up eight points, the NDP 12 and the Liberals down 17 in today’s Ipsos poll. There’s a useful website named Too Close to Call that is set up to do a sophisticated computer-based version of the British swing-o-meter. It has adjusted local riding results from 2014 to take into account regional and other differences such as whether an incumbent MPP is running again.
Go to the site and punch in the gross province-wide numbers, such as PC 40, NDP 35, Liberal 22 and others 3. The site then generates an outcome for each riding. By the way, those percentages would give the PCs 69 (six seats clear of a majority of 63,) the NDP 50 and the Liberals 5 if the election were held today – if the sites adjustments were correct, if the swing followed their model and, of course, if the polling percentages were precisely accurate. The site, therefore, can help size up the situation in a particular riding but it’s important to treat these results with some skepticism.
A similar site, without the detail or the ability to test run different total numbers, is the Laurier Institute for the Study of Public Opinion and Policy. Another site, Election Prediction Ontario project, offers more information, mostly anecdotal. That site has a group of experts study each riding and predict the outcome. It has a poor track record for overall predictions, mostly because it usually fails to pick up the dramatic up or down drafts that occur in modern elections. However, it has a feature that often generates interesting information. Political junkies can post their thoughts in brief comment form. This can provide local polls, information on local candidates and campaigns, and occasionally useful insights. The riding-by-riding analyses need to be read with caution and a sense of humour. It is especially vulnerable to partisans stacking the comment section.
A few days of door knocking cures one of any sense that precise predictions at this stage of an election are possible. Each voter is different from the one next door. Each is either certain or not about how they will vote, even if they will vote. Ford’s 40% is, as are the others, made up of the “leaners” and those who are certain – in his case perhaps 25% of the voters won’t likely change their minds and will almost certainly vote Ford. The leaners may be peeled off by any number of factors or events. Or, he may attract new supporters during the rest of the campaign. The same applies to the other party leaders.
Andrea Horwath goes into the next three weeks with momentum, and relatively positive and high approval ratings. She’s scoring well in crucial perceptions such as trustworthiness and suitability for the office of Premier. She also has the largest number of second place and potential voters presently lodged in other camps or who are undecided. It looks now as if she will have to catch Ford, and even surpass him in total popular vote to stop him. The Liberal collapse, which may gather negative momentum just like the Conservatives did in the 1993 federal election, is turning this into a two-party contest. The first-past-the-post parliamentary system invariably creates this dynamic.
A large majority has decided that it’s time for a change of government. Which way voters will swing to stop Ford will come down to a complex set of perceptions about the past, present and future of this province.
The strategic considerations for the parties and the voters had changed dramatically by the end of the first week. Liberals now face a tough decision. To stop Ford, do they move their votes to the NDP? Or, do they still have a chance of defeating PCs? For NDP supporters, how to appeal to Liberals is straightforward. But how do they appeal to soft PC votes? To which ridings should they shift resources? Constituencies like Durham or Peterborough or all of the Brampton ridings come into play on these numbers. They will likely make the difference between a Ford or Horwath government.
How do you assess the state of play in your riding? Today’s numbers are so different from the 2014 election that they can’t provide any definitive answers. Working with those results, however, is still the best place to start.
In Britain since the end of the Second World War, election experts have developed the “swing-o-meter”. In its most primitive form, it takes the national swings from the previous election and applies them to each constituency. For example, if Labour was up 5% nationally, the Conservatives were down 5%, and the Liberal Democrats and others posted no change, we could conclude a five percentage point national swing to Labour. You can then apply that swing to to the previous election results in your constituency to produce a prediction.
Using this method in the Ontario election, the PCs are up eight points, the NDP 12 and the Liberals down 17 in today’s Ipsos poll. There’s a useful website named Too Close to Call that is set up to do a sophisticated computer-based version of the British swing-o-meter. It has adjusted local riding results from 2014 to take into account regional and other differences such as whether an incumbent MPP is running again.
Go to the site and punch in the gross province-wide numbers, such as PC 40, NDP 35, Liberal 22 and others 3. The site then generates an outcome for each riding. By the way, those percentages would give the PCs 69 (six seats clear of a majority of 63,) the NDP 50 and the Liberals 5 if the election were held today – if the sites adjustments were correct, if the swing followed their model and, of course, if the polling percentages were precisely accurate. The site, therefore, can help size up the situation in a particular riding but it’s important to treat these results with some skepticism.
A similar site, without the detail or the ability to test run different total numbers, is the Laurier Institute for the Study of Public Opinion and Policy. Another site, Election Prediction Ontario project, offers more information, mostly anecdotal. That site has a group of experts study each riding and predict the outcome. It has a poor track record for overall predictions, mostly because it usually fails to pick up the dramatic up or down drafts that occur in modern elections. However, it has a feature that often generates interesting information. Political junkies can post their thoughts in brief comment form. This can provide local polls, information on local candidates and campaigns, and occasionally useful insights. The riding-by-riding analyses need to be read with caution and a sense of humour. It is especially vulnerable to partisans stacking the comment section.
A few days of door knocking cures one of any sense that precise predictions at this stage of an election are possible. Each voter is different from the one next door. Each is either certain or not about how they will vote, even if they will vote. Ford’s 40% is, as are the others, made up of the “leaners” and those who are certain – in his case perhaps 25% of the voters won’t likely change their minds and will almost certainly vote Ford. The leaners may be peeled off by any number of factors or events. Or, he may attract new supporters during the rest of the campaign. The same applies to the other party leaders.
Andrea Horwath goes into the next three weeks with momentum, and relatively positive and high approval ratings. She’s scoring well in crucial perceptions such as trustworthiness and suitability for the office of Premier. She also has the largest number of second place and potential voters presently lodged in other camps or who are undecided. It looks now as if she will have to catch Ford, and even surpass him in total popular vote to stop him. The Liberal collapse, which may gather negative momentum just like the Conservatives did in the 1993 federal election, is turning this into a two-party contest. The first-past-the-post parliamentary system invariably creates this dynamic.
A large majority has decided that it’s time for a change of government. Which way voters will swing to stop Ford will come down to a complex set of perceptions about the past, present and future of this province.
Comments
Post a Comment