Election polling: Plenty of noise, not much signal

This has also been posted at iPolitics.

Nate Silver became an overnight sensation as a polling guru early on in the 2008 Presidential when he contradicted the conventional media interpretation of the race for the Democratic nomination. The pundits were promoting the inevitability of a Hillary Clinton nomination victory. An early supporter of Barack Obama, Silver shrewdly distinguished between good polls and bad, and early on anticipated the Obama presidency. He went on to write a widely-read book, The Signal and the Noise, about statistical probability and prediction.

What we see in Canada is desperation to figure out where the election race is headed with endless parsing of the smallest nuance in the latest poll. There is an abundance of statistical and media noise, but little in the way of clear signals about where the overall election race is headed.

As the Oct. 19 election date approaches, the outcome will likely become clear enough. Meanwhile the media should be patient and allow the public to weigh their choices. They could assist by emphasizing and clarifying policy choices facing the electorate, as well as asking some tough questions about how the polls are conducted.

Instead as we see a headline across the top of an inside page in in the Globe and Mail, “In Quebec, Trudeau Aims to Connect”, a piece of poll-driven mush (the NDP are slipping/ is Trudeau catching up?), while consigning an Employment Insurance announcement from Thomas Mulcair to a capsule at the bottom of the page under the heading: Election Digest. Unemployment and the impact of EI premiums on the economy, is that an issue?

Meaningless analysis of political psychology compared to a key policy announcement. The editorial choice is clear.

This week we have completely contradictory polls from two of Canada’s most respected polling companies. EKOS Research, which just released a survey giving the Conservatives a clear lead. The daily Nanos Poll releases the past two days suggest a completely different picture, a continuing tight race with NDP second, the Conservatives third. A new Forum poll says no to both of those – the Liberals and Conservatives are tied, the NDP is in third. You get the picture.

Polling is having a hard time these days. The media, with news budgets cut by internet competition, won’t pay for much of the polling that is reported (Ekos/iPolitics and Nanos/CTV are exceptions). However, most pollsters pay for and release their own polls simply as a means of promoting their market research business, so there is no shortage.

The media helps out by reporting every number available, mostly without context, but the polls are suffering from a lack of respondents. Response rates to polls as Nate Silver puts it “are dismal these days”. The growing public resistance to answering poll questions may partly explain the polling fiascos we have seen in recent years.

A good example was the 2013 B.C. election where polls and the media anticipated a significant NDP victory. Instead majority victory went to Premier Christy Clark and her B.C. Liberals. Another part of it may be new polling technologies and methods.

The polls were wrong in this year’s UK election. As Nate Silver noted “polls of the U.K. election — most of them conducted online — projected a photo-finish for Parliament instead of a Conservative majority“.

Polling technologies and methods ought to be placed under the media microscope – instead their results are treated as just another poll. One new technique is a variation on the traditional telephone poll. Interactive voice response (IVR) surveys – essentially computerized robots – dial your number and ask questions in an electronic voice. This technique is used by Forum and EKOS among others.

There are several online pollsters (including Leger Marketing , Ipsos and Angus Reid) that conduct surveys via the internet with samples drawn from previously recruited panels, which can number into the hundreds of thousands. The panel is supposed to represent Canada as a whole and act as a substitute for randomly selecting respondents from the whole population. The media should be asking how big the national panel is, and how big are the panels in the provinces and regions reported in the polls.

Are these smaller panels an adequate substitute for the population of smaller regions and provinces? What matters in determining seat counts is regional support so the issue is important. We should also know exactly how these polls are conducted.

What do their reported margins of error mean? Surely all the margin of error tells us is how confident we can be the sample is representative of the panel, but is this really comparable to a telephone poll that could call any landline or cellphone anywhere in population. We should also get information and analysis from the media on whether polling companies contact both cellphones vs landlines and what that means.

Some polling firms are more transparent about their methods than others, but there is nothing to stop pundits and media from asking pointed questions about this vital source of campaign information. One friend who formerly worked in market research said in her company there were problems with response rates for the online polls, another issue the media could investigate.

But polls can also be quite accurate. There is randomness to the order of things. It is entirely possible for a poll not to be done well, but still be accurate. Even hamburger polls have been right. I am not aware of any burger polls this year and I lament their passing – I like hamburgers. A burger poll at the Pacific National Exhibition in 1972 correctly anticipated the demise of B.C.’S Social Credit dynasty. In the 2003 Ontario provincial election the Lick’s Hamburger Poll outperformed one of the commercial pollsters that year. Now there is a media polling story I would love to read in 2015.

The Ethnic Vote in 2015

This has been cross-posted at iPolitics

Waves of immigration throughout Canada’s history have made ethnic sub-populations key targets for Canadian election campaigns. Historically this has benefited the federal Liberals; the party supported mass immigration while governing Canada for two thirds of the 20th century, making ethnic voting a staple of Liberal politics. Challenges have come in recent years, notably from the Conservatives — who achieved considerable electoral success in immigrant ridings in 2011 — but also from the NDP.

It’s largely forgotten now but at one time Canadians of British origin were openly suspicious of immigrants’ politics. In 1924 one prominent Winnipeg businessman said of newcomers: “We welcome all good citizens from foreign lands but if they do not believe in the Christian religion, nor intend to keep our laws, they should be asked without delay to return from whence they came.”

Let’s look at some constituencies where there are large concentrations of Canadians from various ethnic backgrounds.

Ukrainians

Many of the immigrants that Winnipeg businessman was talking about came from Eastern Europe, particularly the Ukraine. Most came prior to World War I and settled on margins of the good farmland in the Prairie Provinces. Based on the 2013 redistribution, in 2011 the top five federal ridings with the highest concentration of ethnic Ukrainians would have elected Conservatives, all but one by comfortable margins, all in Manitoba, Saskatchewan or Alberta. Most of this population is made up of Ukrainians whose families migrated to Canada prior to World War I or shortly thereafter and no longer speak the language.

(Data on the ethnic composition of electoral districts comes from the 2011 National Household Survey, which replaced the long form census.)

The federal Liberals were successful at first with this vote, winning strong Ukrainian ridings in the twenties, thirties and forties. But the Liberals were displaced on the Prairies by the Progressive Conservatives under John Diefenbaker in the 1950s.

Harper has made support for the Ukraine in its struggle with Russian-backed secessionists a key symbolic foreign policy priority — no doubt partly for its domestic political benefit, even if many diplomats remain unimpressed. However, in 2015 Conservative support has slipped even in the party’s strongholds — and that includes ridings with significant Ukrainian populations. Four out of five of these ridings would be retained by the Conservatives today, but current polling suggests one (in urban Winnipeg) could go to the NDP. Note that this constituency, Elmwood-Transcona, is about 21 per cent Ukrainian heritage. A majority voters are from other backgrounds. Ethnicity is not the only influence on voting behaviour.

Italians

Large numbers of Italians settled in Ontario and Quebec after the Second World War, mainly in Toronto and Montreal. They reliably supported the Liberals. That may be changing.

The example of Toronto’s designated ‘Little Italy’ neighbourhood is a good illustration. The neighbourhood is located in University-Rosedale, a constituency with the 46th-largest Italian population in Canada (7.6 per cent). But many Italian-Canadians have long since moved to the suburbs. The two constituencies with the highest concentration of Italian voters are relatively prosperous GTA ridings just north of Toronto (both King-Vaughan and Vaughan-Woodbridge rank among the top 25 most affluent constituencies in Canada). Both would have elected Conservatives in 2011. Current polling suggests the Liberals could win back one of the two (and also pick up an NDP seat in Montreal). And the northern Ontario riding of Sault Ste. Marie, which elected a Conservative in 2011, is likely to go NDP.

South Asians

More recent years have seen large-scale immigration from Asia. The Conservatives targeted these ridings in 2011 and achieved significant but not universal success. Again, using the redistributed vote we find that half of the top ten South Asian constituencies would have elected Conservatives in 2011, although the NDP would have won four and the Liberals two. With the considerable improvement in Liberal support during the current election it is likely that the Conservatives would retain just a third of these constituencies; the NDP would drop two and the Liberals would make significant gains.

Chinese

We see a similar pattern among constituencies with substantial Chinese populations: considerable Conservative success in 2011 with likely large-scale losses, mainly to the Liberals but also one to the NDP, anticipated in 2015.

Over time, new Canadians become more integrated into Canadian society. As they do, their ethnic identities become less important in determining how they might vote. As second, third and fourth generations replace the original immigrants they develop political views they share with others outside of their ethnic sub-groups. Whether they are environmentalists or social justice advocates, free traders or anti-tax conservatives, their ethnic identities have progressively less influence on how they vote and view politics.

Although the Liberals continue to do well among ethnic minority voters, political support from Canada’s minorities has diversified. The efforts made by the Conservatives in 2011 met with considerable success and the NDP has made its own gains. The days of monopolizing the immigrant vote are over, and the political importance of ethnic identity clearly fades over time.

Would the Senate tangle with a Prime Minister Mulcair?

This has been cross-posted at iPolitics.

There was a time when the Senate would sleep from one election to the next. Although it has the same legal authority as the House of Commons (except that it can’t introduce money bills), its non-elected status means it has little moral authority to interfere with the elected House.

Or so you might think.

In its early years, the Senate was still active in amending bills; that role faded when the Liberals came to dominate the chamber in the mid-20th century. The Senate briefly woke from its long slumber after Brian Mulroney became prime minister in September 1984. In early 1985, during Mulroney’s political honeymoon period, the Senate’s Liberal majority delayed a borrowing bill (it was a parliamentary issue — governments are supposed to table spending estimates before raising money).

This action was widely viewed as impudent meddling by the remnant of a discredited ancien régime, and it drew hostile reactions from government representatives, pundits and members of the public. An angry Prime Minister Mulroney threatened to launch a constitutional amendment to strip the Senate of almost all its powers (he backed off, eventually).

The Liberal Senate House leader who provoked Mulroney’s ire was former Trudeau minister Allan MacEachen. At the time I was putting together a documentary on the Senate for the CBC centred on this episode, and I followed the issue closely. MacEachen was a legendary political character — a wily House of Commons tactician and a key cabinet minister in the governments of Lester Pearson and Pierre Trudeau. When we interviewed MacEachen for the documentary I got a strong impression of someone who had been publicly humiliated — and who was unlikely to forget the experience.

MacEachen did not have long to wait for an opening; the Mulroney honeymoon was over by early 1986. The Liberal-controlled Senate began to hold up bills on a wide variety of topics: transportation, drug patents, refugees, copyright and energy incentives. My impression was that MacEachen applied pressure on measures where there was public support for the Liberal position, and where Liberal MPs gave their consent for the Senate caucus to act.

This amounted to a non-constitutional — but very real — change in the Senate’s role. A Senate controlled by partisan opponents of the government can bide its time for a while after an election, read the polls and pick the right moment to oppose the government on issues where it has the public is on its side. It’s a recipe for periodic deadlock.

The most notable development in the mid-1980s came when the Liberal Senate refused to proceed with legislation authorizing the U.S.-Canada Free Trade Agreement until after a federal election. The Progressive Conservatives did go on to win the 1988 election and the Senate passed the deal — but confrontation between the chambers didn’t end there. In autumn 1990, the Senate started holding up the legislation to create the GST.

The Liberal grip on the Senate was suddenly broken when Brian Mulroney used an arcane constitutional provision that let him appoint eight extra senators, two for each region, bringing the PCs a majority.

The new Progressive Conservative majority in the Senate did not forget their party’s treatment at the hands of the Liberals. When the Chrétien Liberal majority swept into office in late 1993, Conservative senators didn’t hesitate to exercise their authority.

Early on, the Conservative Senate blocked Liberal legislation that would have cancelled the Mulroney government’s initiative to privatize the Pearson airport. They held up the bill for several years; the issue only died when the government reached a compensation agreement with the Pearson airport developers. The Tory majority also blocked a Liberal bill to suspend and change a redistribution of seats in the House of Commons then underway, and delayed several other pieces of legislation.

The Harper period has been somewhat different. The Conservatives had a majority in the Senate by the end of 2010, before they won a majority in the House of Commons. Prior to gaining full control of the chamber the Senate Conservatives, using a procedural manoeuvre, killed an NDP climate change bill supported by the Liberals. The bill had been pushed through the House of Commons over the objections of the minority Conservative government.

Even if the NDP was willing to fill all 22 vacancies in the Senate (it says it won’t but, eventually, it might be forced to do so by the Supreme Court), it would be many years before the party could achieve a Senate majority. The same problem would not face the Liberals. Just by filling existing vacancies a Liberal government would regain effective control of the Senate. In the late 1980s and early 1990s, Senates controlled by the opposition showed how it could thumb its nose from time to time at the elected House of Commons. It could happen again.

The current Senate is likely to feel some restraint in the short run, given its tarnished reputation. But that won’t last. Simple hostility on the part of a chamber controlled by Conservatives and Senate Liberals could create legislative gridlock and an unprecedented legitimacy crisis for Parliament.

That might fuel the NDP’s push to abolish the Senate — but unanimous provincial approval isn’t likely any time soon. If there is a Thomas Mulcair government, expect to see at least a new version of the selective delays and vetoes that bedevilled Brian Mulroney and Jean Chrétien in the last century.

Campaign 2015: the story so far, poll by poll

This post has already been published at iPolitics. 

Most elections experience a moment where the campaign gels — a point where political ideas and forces come into focus, framing the likely outcome. It can happen when a party or candidate departs from the pack on a key issue: think of Tim Hudak's pledge to eliminate 100,000 public sector jobs early in the 2014 Ontario provincial election. It could be a major stumble by a leader — such as John Turner's "I had no option" exchange with Brian Mulroney in 1984, or Jim Prentice's unfortunate "math is difficult" crack in the recent Alberta election.

When the moment happens, you know it; twists and turns may follow, but the campaign itself is suddenly on a track going in one direction.

The issue of Europe's refugee crisis may turn out to be 2015's moment. Soon after that shocking photo of a drowned 3-year old Syrian boy made its way around the world, the campaign narrative instantly turned to foreign policy, with Stephen Harper's Conservatives largely on the defensive. The Nanos Poll released September 6 was conducted the weekend after the issue arose, and it reported a dip in Conservative popularity that continued into the next day, dropping the incumbent party to third place.

It's still too early to tell how the refugee issue will affect the remainder of the campaign. We've been at this for 40 days now, but polling on national public opinion has changed little since the campaign began. Regional numbers tell a different story.

If we compare the average of the national polls released since August 2 to the July numbers, we find just a slight change: the Conservatives are down just over two points, the Liberals are up just over two and the NDP is up less than one, holding a narrow lead.

Monthly National Polling Averages

For all this apparent precision, there is wide variation in reported poll results. For example, in Ontario different polls have reported support for the Conservatives ranging between 40 per cent and 21 per cent since August 2.

Campaign Polls in Ontario since August 2

Ontario has the most seats, so this variability is something to watch.

The campaigns at the level of individual provinces and regions differ greatly; every party holds a lead somewhere. When it comes to individual ridings, it's the vote share in the provinces that determines wins and losses. Let's break them down.

In Atlantic Canada the Conservatives are down about four points since the campaign started, and we see a corresponding increase for the Liberals. This appears to represent an early election-related shift.

Atlantic Polls

In Quebec, NDP support has grown significantly since the campaign began at the expense of the Conservatives and the Bloc, while the Liberals have held steady but far behind the New Democrats. Decisive polling shifts are always unmistakeable; the dramatic shift during the Alberta election was a striking example. Compared to the first three months of 2015, Quebec now looks dramatically different. The NDP has gained over 14 percentage points in support since then from all of the other parties — including six points in just the first month of the campaign. This implies a repeat of 2011's 'Orange Wave'. Has the campaign in Quebec already gelled?

Quebec Polls

In Ontario we've seen a small apparent shift away from the Conservatives, more to the Liberals' benefit than that of the NDP. However, given what we know about variability, it is not clear that a shift this small is significant. Along with its size, Ontario's apparent indecisiveness will make it the key target for leaders touring the country.

Ontario Polls

When we move west to the Manitoba/Saskatchewan region we see a clear pattern of movement away from the NDP, which appears to be a partial fading of the NDP surge following the Alberta election. It peaked in July here and in B.C. The 2015 campaign appears to have had not much of an impact to date.

Manitoba and Saskatchewan Region Polls

The Alberta pattern is similar to that of Manitoba-Saskatchewan, with a slightly greater drop happening sooner here for the NDP. There has been a gradual improvement in Conservative fortunes here.

Alberta Polls

In British Columbia the NDP's upward trend preceded the Alberta election, but the Alberta outcome did give it an extra bump. The NDP's losses have gone more to the Liberals than the Conservatives. Polls here are notoriously volatile and variable, reflecting B.C.'s many political microclimates.

British Columbia Polls

The race may be a dead heat nationally; at the local level it's a different story. Given that polling is intrinsically variable, the post-Labour Day campaign is moving into a political climate of high uncertainty.

The Wynne-Harper Feud

A version of this item has been crossposted at iPolitics.

One warm spring evening in 1957 the Premier of Ontario, Leslie Frost, mounted the stage of Toronto's Massey Hall at a Progressive Conservative rally marking the official launch of the federal party's campaign - one that would end 22 years of national Liberal rule.

Premier Frost's role was to introduce the main speaker of the evening, federal Tory Leader John Diefenbaker. He did that and something else besides: his introduction to launch a blistering attack on a recent federal tax-sharing deal offered to the provinces, declaring the issue wasn't "...the Federal Government giving Ontario or the provinces anything. That is the patronizing attitude of Ottawa. All we ask is a reasonable part of our own, a part which is commensurate with the size of the job we have to do...."

Ontario Premier Leslie Frost
 and John Diefenbaker

Frost wanted more resources from the federal government to fund investment in infrastructure, education and economic development among other things. Diefenbaker agreed that Dominion-provincial fiscal relationships "were in a mess." However, he also pledged tax cuts as "one of the major items of business at the next session of Parliament."

La plus ça change.

A dozen years earlier Prime Minister Mackenzie King jousted with Ontario Premier George Drew over the timing of federal and provincial elections as World War II came to an end. Drew, having lost a vote of confidence, scheduled an election for June 11, 1945.  King's five year mandate was up and he planned a federal vote for June 25. Fearing the impact of an Ontario Conservative victory on the federal campaign, King (on the legendary C.D. Howe's advice) called the federal election to coincide with the Ontario election. It was not to be. While King attended FDR's funeral in Hyde Park, Drew outmaneuvered King by moving the date up to June 4. It was too late for King to react. Both would win their electoral contests, but clearly established that federal-provincial feuding would be the new normal - even during elections.

In 2015 the premier is Kathleen Wynne and the prime minister is Stephen Harper. He is refusing to offer federal administrative support to Wynne's proposed Ontario Retirement Pension Plan. She, in response, is doing whatever she can to support Justin Trudeau.

It is hard to overstate just how deeply irresponsible Harper's position is. Federal-provincial cooperation is an essential part of how the federation functions. For example, all provinces except Quebec have their own personal and corporate income taxes collected by Revenue Canada.

Federal-provincial cooperation failed during the SARS crisis of 2003. A subsequent report stated that without fixing this problem Canada would "be at greater risk from infectious disease and will look like fools in the international community."

Various Harper ministers have issued press releases and tweets touting the virtues of federal-provincial cooperation. Free trade agreements require provincial cooperation to be implemented and respected. To take one (rare) example of provincial non-cooperation - when Newfoundland under Premier Danny Williams broke the usual protocols by expropriating the assets of AbitibiBowater, the cost of the action brought by the company under NAFTA had to be paid by the federal government. What Williams did was far from the norm. The same is true of Mr. Harper's refusal of cooperation with Premier Wynne. If a province wants this administrative assistance for a new pension program, Harper is obliged to extend the normal courtesy essential to the efficient management of Canadian federalism - whether he likes it or not.

Harper's position appears to be little more than ideological zeal. His opposition to public pensions has extended in the past even to the Canada Pension Plan.  Toronto Star Columnist Thomas Walkom has noted that "at various times in his career, Harper has dismissed the Canada Pension Plan as a boondoggle and tax grab that should be taken apart and privatized."

Premier Wynne herself has involved herself more deeply in the federal campaign appearing at a rally on August 17 with Justin Trudeau in downtown Toronto where she not only criticized Stephen Harper but also attacked NDP leader Thomas Mulcair's ideas as "incomplete or ... unworkable or ... impossible". If Harper's response to the Ontario pension proposal was intemperate so was Wynne's partisan broadside against Mulcair. She may need a relationship with a Prime Minister Mulcair after October 19. The rules of federal-provincial diplomacy suggest she should have adopted a more measured tone, especially given that both Mulcair and Wynne agree on the need for CPP expansion.

Harper's ongoing refusal to meet with the Premiers outside of one-on-one chats and vapid photo ops - his indifference to working with provinces - has not been good for Canadian federalism. It is time to change the tone.

The campaign so far

The 2015 election started off close and competitive as the fall campaign actually got underway in the summer. After the Alberta provincial election in early May that elected New Democrat Rachel Notley, the national NDP surged to a narrow lead in what is now three-way race. But the national picture is always misleading. Canada-wide numbers are a composite of regional and provincial electoral contests.

All three parties currently have leads in different parts of the country. A tight race means that dramatic change could happen in the blink of an eye. Third could become first; first could become third. The final outcome is far from determined.

Current headlines emphasize the weak performance of the Trudeau Liberals compared to their chart-topping rise in the wake of Justin Trudeau's assumption of the party leadership in April 2013. Nonetheless the Liberals remain substantially stronger than in 2011.

There have been three polls released since the campaign commenced.  One was the Nanos poll, which was taken over a four week period from July 10 to August 7 so it captured more of the pre-election period than the election itself. The others were a Forum poll and an Ipsos poll taken this week.  What was notable about the Ipsos survey is not only was it the first poll after there had been some campaigning, but it used a different methodology than we are accustomed to seeing from Ipsos. Their monthly polls in the past few years have been online. As their news release states "a sample of 2,022 Canadians eligible to vote was interviewed: 1,022 were interviewed online via the Ipsos I-Say Panel, and 1,000 interviews were conducted by live-interviewer telephone (including 40% of interviews conducted on cellphone)". 

It would appear that Ipsos has greater confidence in their polling if it is not done exclusively online.

The individual poll results with my seat estimates can be found below. Early campaign polls and all seat estimates need to be taken with a grain of salt.

The Debate
The Ipsos and Forum polls were conducted following the first major event of the campaign: the television debate. However, the audience size was a fraction of previous debates, so it almost certain that most of the audience consisted of the politically engaged and knowledgeable.

Most debate commentary placed emphasis on the relatively even nature of the performances and the absence of a defining moment, but suggested Justin Trudeau may have done better than expected. My guess is that due to its small audience and uncertain outcome, it had zero impact on the distribution of political preferences in the polls. A couple of polls tried to gauge the debate's impact but they should be distrusted. Pollsters face high refusal rates generally, and must make numerous calls before even one survey is completed. The numbers of calls needed to find those who actually watched would have been astronomical.

I asked people I bumped into the week before about the debate. Only one person even knew it was coming. The debate itself had amateurish production values and a host, journalist Paul Wells, whose lack of television hosting skill was painfully obvious.

The debate situation in Canada this year is a fiasco for which Stephen Harper is squarely to blame. There ought to be a strictly limited number of debates, widely available on broadcast media including television and radio. They should be managed by an independent commission and participation should be based on objective criteria, not the whims of the debate sponsors. The reality is that most citizens have little political knowledge and broadcast debates are an excellent medium for citizen participation in the electoral process. The English debates ought to include the Green Party and the French debates should include the Bloc Québecois. Both parties have demonstrated sufficient levels of support and participation among the electorate to deserve inclusion.

The absence of a widely accessible debate means the electorate is more dependent on news clips and ads, information received passively and often inadvertently. The Conservatives no doubt hope that remains the case. Their behaviour suggests they are fearful that a big widely viewed debate will harm their prospects.

All of this said, debates frequently have no impact on outcomes despite media assumptions that they will. In the Ontario 2014 election Kathleen Wynne was clobbered in the TV debate and then went on to win a majority. Nonetheless, voters should at least have the opportunity to glimpse the offerings of the the parties and leaders without the filter of the news media.

As well, scandals often have less impact than anticipated. Headlines this week are focused on Nigel Wright's testimony at the Duffy trial, and it is true that support for the Harper government was at its lowest in the fall of 2013 at the height of the scandal. If nothing else the news coverage will serve as a reminder of those days. However, I suspect that the impact of the scandal has already been felt, and is priced into the over 20 per cent drop in Conservative support since the last election. It will be difficult and likely impossible for the Conservatives to win back the confidence and trust of the electorate lost in part because of the Senate scandal.

Are the Harper Conservatives really the best bet to win?

Recently a west coast political commentator with ties to the BC Socreds and B.C. Conservative Party named Will McMartin published a commentary in the Tyee arguing that Harper was still the best bet to win this year.

He acknowledges there are polls putting the NDP in first place but states: "In fact, Harper is near-certain to be our next prime minister unless the NDP makes significant further breakthroughs in key parts of Canada."

He goes on to say that those "predicting" Harper's defeat are wrong, further asserting that he is "just the number crunching messenger here".

The numbers he crunches, however, appear to be entirely based on the 2011 election results. He may be basing some of his observations on polls but does not say so. While polls can be subject to error on the whole they get things right.  The polls, for example, correctly anticipated the result in Alberta this year.

The problem with basing his analysis only on 2011 is that the Conservatives are currently averaging a lot less support in 2015 than in 2011. Their average support in July polls is 30.1 per cent compared to Harper's 39.4 per cent of the votes in 2011, about 24% less, all of which has gone to other parties. So it is minus nine percentage points for the Conservatives and plus a like amount for the opposition, an eighteen point swing.

He argues that the new expanded House of Commons "has made the Tories' task much, much easier. This is because when votes from the last election are transposed onto the newly drawn electoral districts, Harper's Tories pick up an extra 22 seats, compared to the NDP and the Liberals adding just six and two respectively."

He makes a logical error previously made by others (it is also a feature of the 2013 book The Big Shift, pollster Darrell Bricker and Globe and Mail columnist John Ibbitson). The Harper "gains" from redistribution only occur if he receives the same number of votes.  With far fewer votes there would be far fewer Conservative seats not the same.

Ontario is getting 15 of the new seats many in the suburban belt of constituencies surrounding Toronto. In the 2011 federal election the Conservative Party did indeed sweep almost all of those suburban ridings. In Ontario – except for the north – federal and provincial constituency boundaries are identical. Provincially most of the ridings swept by Harper have now voted twice for the provincial Liberals. In fact, 40 constituencies that elected a member of Stephen Harper’s 2011 caucus sent a Liberal to Queen’s Park on June 12, 2014. Another seven 2011 federal Conservative ridings went NDP provincially. Voter preferences here are not frozen with the colour blue. They vary over time.

Current polls and my seat estimation model make it clear that most of these same suburbs would now elect Liberal or NDP members not Conservatives. McMartin seems to assume the 2011 results somehow apply automatically to 2015. They do not.

His regional commentary analyzes what to expect in the various regions across the country but it suffers from the same inability to acknowledge that to date the Conservatives have suffered losses of support. The Conservatives have lost ground everywhere except Quebec where they had their weakest showing in 2011 and now average about the same level of support. His seat estimates are all therefore faulty for the same reason. He sees only five or six seats lost in Atlantic while I make it ten. He thinks the Conservatives will hold 18 of 22 seats in Manitoba and Saskatchewan while I think they will only keep 14. In Alberta he sees no losses while I estimate they will lose six.  The federal Tories are really getting hit hard in B.C. where I see them currently dropping to 7 seats while McMartin thinks they will get 20.

His estimate in Ontario is less precise but he thinks they will get well over 60 while I have them currently at 52 (this would represent a loss of 31 seats from the 2011 transposed results in this province alone).

It is perhaps better to look at the analysis as representing an optimistic view of what might be possible for the Conservatives in 2015 if somehow all goes well (headlines about economic and fiscal troubles will clearly not help).

The Conservatives do have a realistic chance of having the most seats, but if they do they will almost certainly be far, far short of a majority, and very much on life support post-election.

I found one part of the commentary I do agree with: "A riding to watch is the newly created Hamilton West-Ancaster-Dundas. (my emphasis) Transposed results from four years ago show the Conservatives with 42.4 per cent of the vote, closely followed by the New Democrats and Liberals with 28.2 per cent and 24.9 per cent respectively."

Right now I have the NDP ahead in this constituency just ahead of the Conservatives and the Liberals, all parties with around 30/31%. Not clear who will win but it does illustrate that having a big lead coming out 2011 is no guarantee of success in 2015.