About EKOS Politics

We launched this website in order to showcase our election research, and our suite of polling technologies including Probit and IVR. We will be updating this site frequently with new polls, analysis and insight into Canadian politics. EKOS's experience, knowledge and sophisticated research designs have contributed positively to many previous elections.

Other EKOS Products

In addition to current political analysis, EKOS also makes available to the public general research of interest, including research in evaluation, general public domain research, as well as a full history of EKOS press releases.

Media Inquires

For media inquires, please contact: Frank Graves President EKOS Research Associates t: 613.235-7215 [email protected]

And the Winner is… We Don’t Know

EVIDENCE POINTS TO LONG MONDAY NIGHT NAIL-BITER

[Ottawa – October 17, 2015] We have upped our game in trying to discern the winner of what is shaping up to be a historically important federal election. We have run parallel HD-IVR and live interviewer surveys. We have significantly increased the sample size in the home stretch. We have double- and triple-checked the sample diagnostics. We have even done formal experiments with different ballot questions. Despite all of this, we still see the final outcome as very fuzzy.

Nothing is much different in this more richly resourced polling of the last couple of days. At 33.7 points, the Liberals remain statistically tied with the Conservatives who are just a fraction of a point behind at 33.3 points. At 21.9 points, the NDP is not a serious contender for the lead, but is still likely to be a major player in the next Parliament. Our two-day roll looks like our three-day roll and everything is moving within the margin of errors. At this stage, we believe that either the Liberals or the Conservatives will achieve a slender minority.

The demographic and regional variations are modest but worth noting. Quebec seems to be a pretty tight and newly fluid three- (and maybe four-) way race with the NDP holding a modest advantage. Ontario is leaning Liberal, but there are a plethora of virtually tied contests (particularly in the 905 region) that are uncertain at this stage. British Columbia is once again an uninterpretable mix of three and, in some cases, four parties.

We will be offering a seat projection tomorrow based on what we believe to be perhaps the strongest riding prediction model extant at this time.

The winner will hinge on turnout and, while predicting who will show up is a notoriously challenging task, we can offer the following predictions:

  1. Overall turnout will be higher than in 2011 and the higher the overall turnout, the poorer the prospects for Stephen Harper (advantage centre-left).
  2. The Conservatives have a highly engaged voter base and they had a slight lead in the advance polls. Older Canada is leaning strongly Conservative, they are going to vote and there are more of them than in 2011 (advantage Conservatives).
  3. The cellphone-only population looks like they will show up this time. Cellphone-only households make up a much bigger share of the voter population this time out and they continue to be less supportive of the Conservative Party. They have told us that they are more engaged and certain to vote than in 2011 and they appeared to show impressive participation in the advance polls (advantage centre-left).

20151017_slide1

20151017_slide2

20151017_slide3

20151017_slide4

20151017_slide5

20151017_slide6

Methodology:

This study involved a blended sample collected using two separate methodologies: Computer Assisted Live Interviews (CATI) and EKOS’ proprietary High Definition Interactive Voice Response (HD-IVR™) technology, which allows respondents to enter their preferences by punching the keypad on their phone, rather than telling them to an operator. In an effort to reduce the coverage bias of landline only RDD, we created a dual landline/cell phone RDD sampling frame for this research. As a result, we are able to reach those with a landline and cell phone, as well as cell phone only households and landline only households.

The figures in this report are based on a two-day rolling sample. Each day, a new day’s worth of interviewing is added and the oldest day is dropped. The field dates for this survey are October 15-16, 2015. In total, a random sample of 1,537 Canadian adults aged 18 and over responded to the survey (1,105 by HD-IVR, 432 by live interviewer). The margin of error associated with the total sample is +/- 2.5 percentage points, 19 times out of 20.

Please note that the margin of error increases when the results are sub-divided (i.e., error margins for sub-groups such as region, sex, age, education). All the data have been statistically weighted by age, gender, region, and educational attainment to ensure the sample’s composition reflects that of the actual population of Canada according to Census data.

Click here for the full report: Full Report (October 17, 2015)

58 comments to And the Winner is… We Don’t Know

  • Ian

    Any guesses on why EKOS consistently has the liberals lower than Nanos and all the others?

    It’s a risky game for EKOS… If they’re right, they look like geniuses. If they’re wrong, they’re the scapegoat. I hope they’re right though!!!!

    • Cre47

      Given the numbers in the 18-34 group as well as in BC and Quebec, I sense it’s the other pollsters who are right, since their numbers for the most part are more consistent than EKOS. Definitely a risky game for EKOS. Of note in many of the polls, the Liberal support among the 18-34 is even lower than among the 65+ (such as this poll)

    • medtech71

      It will be Nanos sucking wind on Tuesday. If the CPC is ‘overpolled’ I believe that would be the first time ever…they are ALWAYS underpolled.

  • Raunch

    In 2011, Ekos results were among the worst of all the pollsters (6 points off on the Cons, alone)
    All the other pollsters show a much bigger Liberal lead.
    Today’s other polls:
    Leger
    Libs 38
    Cons 30
    NDP 22

    Mainstreet
    Libs 38
    Cons 33
    NDP 21

    Nanos
    Libs 37
    Cons 30.7
    NDP 22.6

    If I was a betting man, I would put my money on anyone over Ekos.

    • i just want to add that the actual results of the mainstreet poll are

      lib 34
      con 30
      ndp 19
      und 10

      and, once again, you see the relationship that shows up with this “dithering progressive” effect over ivr, and it’s subsequent inflation of the conservatives due to a decrease in sample space. and, it again absolutely nails the conservatives at a 30% flatline, which has been true for months and months.

    • Cre47

      It’s also worth noting that EKOS had the tightest margin among major pollsters during the week before polling day in 2011, so EKOS was about 7 points behind the actual results in terms of the winner’s margin. Looks like deja vu all over again in this election

    • Bryan

      EKOS is always wrong in the same way though: they under-report right-of-centre support by at least 5%.

    • medtech71

      Leger was an internet poll…so you can toss that one…and Mainstreet has ridiculous margins of error.

  • Orwell

    There have been consistent differences between Nanos & Ekos. But I really don’t believe any polling company is biased, because all of them depend upon their reputation in order to stay competitive.

  • Michael Crowell

    Well I believe that reality had finally established a foothold. Harper may not be perfect but he offers Canada stability in the world of stormy seas. Any fool can spend billions and offer everybody everything. I see shades of the father coming through.

    • Alain Robert

      You forgot to mention : Paid by the Conservative Party of Canada

    • Leib

      It is a false belief that Stephen Harper is better at managing the economy. He tripled the CSC’s budget over the last ten years costing Canadians 80 billion extra dollars. Yet there are no new prisons built, the inmates are suing because the food does not meet basic nutritional standards, and then another lawsuit for cancelling minority chaplaincy services. The Cons (I mean Harper here) say they have quintupled the budget for prison security, yet there have been more institutional murders and acts of violence against guards since Harper took power (see Mountain institution on wikipedia as an example). Then there was the 2 billion spent on G20 summit security in just 48 hours. These are just two examples of many of the 100s of billions of dollars this man has pissed away while cutting old age security and war vets benefits and other social benefits. The only people trying to convince us he is a better money manager are the super rich like Kevin OLeary who do not want there precious low tax levels to raise to support social systems for the rest of us because he and others like him have no use for them.
      Harper is NOT GOOD for the majority of Canadians. He is good for the ultra rich.period

  • Dave R

    It’s worthless to compare accuracies of various polling firms in past elections since every single firm has changed their polling methodologies and weighting techniques.

  • shob

    So Raunch.

    1)Leger is an online panel, so it doesn’t count at all. its not even a poll and has no margin of error.

    2) Nanos uses only one methodology live interviewer, which we know favours the Liberals aka shy Tory effect.

    3) Angus Reid is showing a 1 point point spread among likely voters essentially the same as Ekos.

    And 3 polling forms all have the Conservatives slightly edging the Liberals among those who have already voted, EKOS, Mainstreet, and Reid. Ekos numbers are in line with all of these.

    Leger and IPSOS are entirely online panels (chuck em in the garbage for horse race numbers) Forum is partially an online panel and river polling, worse than garbage. Nanos is one methodology.

    • angus reid is also an online panel.

      the shy tory effect is bunk. but, we seem to be seeing a “dithering progressive” effect in the ivrs – it is absolutely consistent.

    • Jeremy

      Lol. And if any one of those ‘garbage’ polls showed the Tories in first place they would suddenly become the only legitimate methods of polling. An unashamedly and ironically biased summary of exercises arguably designed to remove bias through their methodologies.

      Truly funny.

    • Raunch

      Shob
      Let’s just look at the Mainstreet poll……..Depending on which way you decipher the numbers, the Libs have a 4 – 5 point lead over the Cons. As a matter of fact, the president of Mainstreet says that the Libs are on the razor’s edge of being in majority territory. Now, considering this poll was commissioned by Postmedia (National Post) and they are using it in their political coverage, I would suggest that they take the findings of the Mainstreet poll as a decent template of what is going on with the voting public. Conservative – friendly media pronouncing that the Liberals are surging, in today’s paper( or website), is not a positive harbinger of things to come for Harper & company.

  • Skeptic

    I have to say…can EKOS offer reasoning behind why their polls are so drastically off from the others? Every poll released since before last weekend gives the Liberals a 5-8 point lead over the Conservatives, except for EKOS and the Angus Reid poll today that gives the Liberals a 4 point advantage. They’re also the only polling firm to show moderate Liberal drops when every other poll is showing moderate Liberal gains.

    Looking back, EKOS also showed a tight race in 2011 that wasn’t tight at all. EKOS, what has been done to increase accuracy of your polls? We won’t know for sure until Monday night, but I’m finding it increasingly difficult to trust your numbers when they appear so far off from what everyone else is showing us.

  • these results are only different from the consensus in quebec and bc. in both cases, ekos is polling the conservatives much higher and the liberals much lower. it happens to be that these are also the places that have three or four way races.

    passive consumers need to understand that pollsters don’t just collect opinions and send them out. they modify the results to fit the census, and there’s some interpretation in the process. that could be a part of the difference you’re seeing.

    i’m also not sure that the blended sample is the best idea, because of the way that undecideds are being measured. and, i’d kind of argue against unlike aggregates for the same reason. aggregate live interview with live interview, and ivr with ivr, but don’t aggregate online samples with ivr and live interviews. if the ivr is truly measuring undecideds in such a way that reduces the sample size and inflates the conservatives, and you take more ivr in the sample, than you would expect inflated conservative numbers in the end result.

    it’s not enough to do the “one of these things is not like the others” skit and rule ekos out. but, i think this experiment with ivr and live interviewer is maybe pulling out a bit of a bias, and destroying the ability to adjust for it by mixing the data up. that, i think, is a methodological flaw.

  • Mike Smith

    Among pollsters, EKOS was the farthest off in 2011. Nanos called it in 2011 and was within 1% for every party. Stick to the Nanos numbers.

    • JD

      Like performance evaluations for mutual fund managers, past performance success is not necessarily an indicator for future performance success. It may, but it also may not be….

  • shob

    NO Jeremy. They’d still be garbage. The venerated firm Gallup is no longer polling horse race numbers because they don’t believe it can be done accurately any more. And that’s GALLUP. Polling online is complete and utter garbage for horse race numbers. it has no margin of error and could only be close to accurate by chance and nothing else.

    • Jeremy

      So Gallup says that and it represents what exactly? Are they some sort of gold standard. The reality is that most of the time someone does give the right numbers … some firms more frequently that others with respect to political polls. Some like to overweight attention to the few times pollsters get things wrong but 19 times it off 20 they are right. I don’t know if after sites are even more accurate but they are blending heterogenous samples so who knows (deathtokoalas alluded to this above). You haven’t actually offered a cogent rationale for your opinions in any case.

  • Graham

    To all those crapping on Angus Reid, they were the most accurate in 2011 with a total 5% differential. Nanos was second at 5.3%.

    As for Angus’ online survey, they have made significant changes to their methodology since their terrible performance in the last BC election. IIRC, they’ve hit their last 5 election predictions damn near bang on.

    Mainstreet in Alberta’s election this year – While they were correct in predicting an NDP win their total differential was 13%.

    ALL pollsters under predicted CPC support in 2011

    What does it mean for this election? Haven’t got a clue.

  • Steve

    When I go to the football game
    I watch the game ,not the scoreboard
    The media seems addicted to these polls ,which probably do not accurately reflect the game
    Gallup has said they are not doing it anymore. Just can’t be accurate these days
    Who really knows who is on the others end of a cell phone,or what area they are from ?

  • Dan

    The numbers in BC don’t make sense. This is what EKOS showed 6 weeks ago:

    September 4th
    Ontario- LIBS 34 – CONS 32
    BC- LIBS 25 – CONS 31

    Today
    Ontario- LIBS 43 – CONS 34
    BC- LIBS 24 – CONS 37

    So the liberals are now doing much better in Ontario but worse in BC? Can’t be.

    • JD

      I can understand the BC numbers, but what baffles me are the Ontario numbers. Admittedly it is federal and not provincial but with Wynne’s close association with Trudeau, are Ontarians really so content with the Wynne government that they would vote 43% for the Liberals???? It boggles the mind!

      If it is true, then all I can only say to Canadians, welcome to DEFICITLAND and uncontrolled government spending!

      • regan flint

        Maybe it was HARPER showing up with ROB and DOUG FORD that hurt HARPER in Ontario?
        People in Ontario remember what MIKE HARRIS did to Ontario. They also remember ROB FORD.
        If you want to talk Provincial politics ask the ALBERTA CONS why they left the NDP a $$$ 6.9 BILLION deficit to go a long with HARPER’S 2 recessions and 8 deficits. It does work both ways.

        • John Smith

          Two recessions. I seem to recall Canada did rather well during the Global Finacial Crisis.

          • Bruce M

            Yes we were fortunate the Liberals had put banking regulations in place. In the US, conservatives repealed Glass Steagal, and that didn’t work out so well.

        • I am fed up with the LIES check out stats can form 380-0066 it will show that in the second quarter Canada’s gdp grew by 3.4 billion so technically no second recession! Ever wonder why they lied? How about with regulation pot legalization will result in less consumption… in Alberta we have a saying ” I CALL BULLSHIT’…. Both Colorado and Washington show consumption more than doubled… Can only wonder about the” creditable sources” Trudeau claims to have , must be his good dope token buddies. Does it not bother you that the Liberals are going to “run deficits” is double speak for ” we going to spend a lot of money really inefficiently on all are friends”. Did you know that since the Great Recession no country has gained more jobs on a per capita basis in the G7… It is called the jobs index … Look it up , I guess the liberal platform will be “Lets try for number 2 I am from Alberta that NDP government gets run out town on a rail next election… We have serious buyer’s remorse Just look at the raw deal BCE and the CTV have given the PC’s, clearly Harpers enemies are out for his hide. I guess that after 10 years of making the tough calls you gain enemies, To his credit he never kissed ass but kicked a few , not so sure we should lose him..ljr

      • Dan

        I don’t pretend to know which ones are true. It is just the conflicting trends that don’t make sense.

  • JD

    One of the biggest factors determining the ‘accuracy’ of a particular polling result is the likelihood that a particular demographic will actually go out and vote. It means nothing if I say that I support party A, but do not actually cast my vote.

    On Monday evening, the 3rd game of the ALCS take place, and it will be interesting to see what impact that will have on the voter turnout among twenty somethings particularly in Toronto or in Ontario.

  • Raunch

    Nanos poll this morning
    Libs 37.3
    Cons 30.5
    NDP 22.1

  • Bcando

    The only true number is the result tomorrow and we all know that people want change So we will see. I still think it’s going to be a close race because the Tories may leave it to the ballot box to make their decision if they want to jump ship or not. There are still a lot of staunch a conservatives that may not stay home.
    I’ll be so glad when it’s all over.

  • John Smith

    Im still calling it for the Tories. As someone has already pointed out the right-of-centre vote is always underestimated by EKOS. And remember the collapse in Lib support last election, it look like the decline has set in but nothing like what happened in 2011.

  • Cre47

    I’m still sticking to the 7-point advantage for the Liberals as well as obtaining between 140 and 150 seats. There are so many wild cards. A big one is how will the Bloc will fair in Quebec as the projections from the polls are giving anywhere between 1 seat to almost 20. There are many three-way races between Liberals, NPD and Bloc and some 4-way races with the CPC, that a movement of a couple points could have a 10-15 seat swing. Some polls show the Liberals gaining 25 seats approx. Although if you drop the NDP vote by just a few points and raise the Liberals by the same, than the Liberals could win the plurality of seats in Quebec. Liberals can win in some of the Montreal’s closest suburbs, as well as in the Outaouais and Eastern Townships, but are also in contention in Eastern Quebec, Quebec City and the Mauricie region. I definitely see them win at least 20 seats, but I don’t rule out as many as 30 to 35 seats

  • I am going by the vast majority of polls that have the LIBS in the lead. Some of them by more than 6 points.
    The NANOS poll has the largest poll sampling and the least room for error. It is sponsored by the GLOBE & MAIL [who have endorsed the CPC but not HARPER] and CTV. Those are hardly left wing institutions. Even a major poll sponsored by POST MEDIA [who have endorsed HARPER] have the LIBS a head. If HARPER is losing in right leaning polls. Harper is in big trouble.
    What isn’t showed in any poll is how the 71,000+ students at schools will vote? I don’t think that they are big HARPER fans. It also doesn’t include how the FN Peoples in their Ridings will vote. I would guess that in those 51 Ridings not many votes are going for HARPER.
    I think that it will be a LIBERAL minority and the NDP will do a little better than most people think.

  • I think that you point 2. in who have advantage / disadvantage is nullified by the fact that the article before this one says that the CONS numbers were exaggerated in the advance polls numbers.

  • BC

    What matters is how these numbers translate into actual seats won, and I will put my money on the Conservatives.

    In the end I simply cannot believe that Canadians would want a pot-head as Prime Minister.

    The sensible vote will take it.

  • Andrés

    The biggest difference between these results and those predicted by Nanos is the BC results. I doubt that the Conservatives have the support Esko predicts for them in BC, given them more than 10 points over the Liberals, and I hope they are wrong. ABH!

  • Darren M

    I actually think the strong numbers for the Liberals may work out to the CPC advantage. I think there could be many partially engaged voters wanting change but may look at the Nanos poll numbers that the media is fixated on and decide not to go vote thinking its in the bag. I would assume the majority of those votes would be Liberal and some NDP.
    The Jays game is interesting cause it isn’t so much who’s in the stands its who’s sitting in the bar watching the game, I would argue the higher percentage of those watchers will be the 18-35 year old demographic, which is showing to be mostly NDP & Lib suppoerters in ONT and Que. It could actually effect the outcome of a hand full of seats that are running tight.
    At the end of the day, polls are at best a wide lens picture of what could happen, but not necessarily will, and you wonder how non-partisan some of the pollsters can be when they get hired by media corps who clearly have agendas, that are usually centre-left…… Actually always centre-left.

    • Cre47

      The game is at 8:00 PM, so it won’t make much a big impact as the polls close at 9:00 PM. They have several hours available between end of work and the start of the game and I don’t think line-ups will be an hour long as there will be many more polling stations available/open instead of the one or two boxes during advanced voting. Plus the Jays are trailing 0-2 in the series, so there maybe less enthusiasm – I know they did came back from 0-2 deficit but against a much weaker team then they are facing. Okay enough of my baseball analysis now.

      • Nick

        yea, when this game is a must win that pretty much guarantees a hot show of support. Not sure if you follow sports at all because this is pretty basic. If we came off of a win or two wins you would see the relaxed attitude. This is a must win that will be watched by everyone. Last game had 11m Canadian viewers. I would expect more this game.

  • John Smith

    If Turdeau loses whats the bet the BlueJays will get the blame.

  • 1. I think that people thinking that the people hoping that the Jays game will make a difference in voting are ” swinging for the fences ” and are having some wishful thinking.

  • IPSOS REID gives the LIBS and 7 point lead over the CPC. Just another poll saying that the LIBS are ahead.

  • Dean

    Regardless of what Nanos or Ekos has to say, Harper is clearly going to win this election. Trudeau’s pride disqualifies him to be Canada’s prime minister, and many Canadians that want to vote for him because of his modern style personality will get cold feet when they cast their vote because deep down in the bottom of their heart they know he does not have the depth that is needed, nor does he have the depth he says he does. Harper’s humility will be regarded by God who is over government anyway. Refer to James 4: 6 ‘God sets himself against the proud’.

  • Gord Mac kay

    To all liberal voters.Fool me once shame on you. fool me twice …

  • Wayne D

    To me its clear that the Liberals will win a majority government today. The wave for change and anti-Harper sentiment (especially in the media) is simply too strong. I’m not sure why more are not expecting this result. God help us…

  • Mark

    Deficit spending is STEALING FROM YOUR CHILDREN TO BUY VOTES TODAY………..JUSTIN HAS PROMISED TO DO SO.
    Anybody casting a ballot for this must be so proud!

Leave a Reply to Orwell Cancel reply