Saturday, August 28, 2021

Election 2021: On the Cusp of Week Three

Week two of the campaign is coming to an end and what can we make of it?

Well, you cannot talk about week two without talking about the polls. Although, as Jackie, someone who visits this blog from time-to-time, has pointed out it is just three polling companies that are driving the polls right now. They are the three that are doing rolling polls every night and publishing their estimates the next day.

These polls have not been good news for the Liberals. Two of them are rather gruesome while the third shows the Liberals basically where they were at the end of the 2019 election. Are these polls a reflection of reality? (Probably not) Two of these polling companies have estimates that, if they were true, would result in a Conservative victory but the polling companies are still saying that their seat models are showing a Liberal minority government. Say what? I know that the Liberals have the best voter efficiency of all of the parties but what you are describing is out of this world voter efficiency. Your commentary on your own polls is out of whack and the question then becomes, why. Do you not trust your own data? (They don't) It is interesting that neither of the two polling companies that have published gruesome results for the Liberals did rolling polls in the previous two elections. However, the one that has done them since I cannot remember is showing very different results, results that would truly indicate a Liberal victory with their voter efficiency. 

So how are the parties reacting to these polls? The PM took the day off. According to the polls the proverbial wheels are flying off of the campaign bus and he took the day off. This was a planned day off, by the way, because if this would have been a last minute decision, where planned events had to be canceled, our balanced, non-biased media would have discovered that fact by now and provided their usual balanced and un-biased take on that decision. The polls have been troubling long enough to allow for planning events for today without making them look like they were thrown together at the last minute, which would also be noticed and yet he still went ahead with his day off? Over the past two weeks I have not seen any indication that the Liberals have been rattled by what the public polls are saying. They are getting on with getting on.

How about the Conservatives? Erin O'Toole is still warning Canadians about a Trudeau majority. Say what? Dude, the polls say that such a majority if off of the table and a couple of polls are saying you might actually win this thing. What are you talking about? 

As for the media and the "professional" political commentators I would just point out that a certain former Liberal operative, who will never miss an opportunity to bash Mr. Trudeau, has also said the Liberal majority if off of the table. He has not gone to the next level despite the  gruesomeness of a couple of the rolling polls. 

I know that I have been saying this for weeks if not months but modern election campaigns are driven by data. Gone are the days when some "Rainman's" political instincts were the only factors impacting political decisions. While those political instincts are still important, data and its analysis has become its equal in making political decision.

Both of the major parties have data analytics teams working for them. They are collecting and analyzing a tremendous amount of data each day, using the latest data analytics techniques and providing the two leaders with accurate snap shots on a daily basis. As an example, the polling companies doing the rolling polls are interviewing 400-500 respondents each day, from across Canada. The data analytics teams of the two parties are doing that in hundreds of ridings each day and gathering data from other sources besides. What they are telling the two parties is probably much different from what the public polls are telling us. And I think the pollsters and pundits know this which is why they are hedging on what their polls are actually saying.

Enough about polls. What how has week two gone otherwise?

The Liberals book ended their week with a couple of health care announcements. The first one was to promise to hire more healthcare workers and reduce wait times. That announcement was made in Halifax, which just had an election that revolved around those very issues. Most of rural Nova Scotia voted to change their provincial government and the Liberals went to Nova Scotia and essentially said that a Liberal government would be a willing partner with the new provincial government in achieving its key election promise. That is not flashy but it is brilliant. What was funny was that announcement was preceded by a video of Erin O'Toole agreeing with the idea of more privatization of Canada's health care system. I believe it was just a way to show the contrast between the two party positions but then Twitter put a warning label on the video, the media noticed and made it an issue. That lead to the video being viewed hundreds of thousands of times and having Erin O'Toole dodge questions about his view on privatization until he could no longer dodge them. He finally had to admit that what was said in the video is accurate, he would allow more privatization. I wonder, knowing that the Liberals realize the media is hostile towards them, if they did not plan it that way. Did they pawn the media and create what could be THE wedge issue for this campaign in the process? If they did bravo.

One final note. I noticed that the DPM has been going around the country campaigning with Liberals. She is not just sticking to her own riding but campaigning on a level just below the PM. As well, I have noticed the the Liberal Party has been allowing individual election candidates make some of the more mundane election announcements. With a hostile media that is the kind of campaign you need to have. These efforts will not be noticed by the national media but they will be noticed by the local media, which are usually much less hostile towards any of the political parties.

As for the Conservatives they have not really had a spectacular campaign so far. If the media was not pulling for them they would be nowhere. Erin O'Toole is a one man band. As well, so far he has come down on the wrong side of around a half-a-dozen issues, including admitting that he would allow more privatization of Canada's health care system. That provides plenty of material for the Liberals to use over the next three weeks and there are probably a few people in the Conservative camp wondering when the next shoe will drop.

The preliminary round of the election is over. All of the campaigns will switch gears and begin the real election this week. It should be an interesting three weeks.

4 comments:

Jackie Blue said...

What do you make of this? I found it buried in a Reddit thread after succumbing to despair from Ipsos' Darrell Bricker's commentary. This person appears to be a volunteer or staffer working on the campaign in one Liberal-held riding, replying to one of the seat modeller people who had some snark to throw at the Liberals (and is a more right-wing fellow who has a tendency to do so):

https://twitter.com/OttawaEngineer/status/1431421314556891139

What I found curious wasn't just the numbers he said they had, but the rationale given for why the rolling polls were so bad for the Liberals. This is coming from the party itself and it was the same as Evan Scrimshaw's (who used to be a Liberal staffer himself) from the other day: their core voters are away on holiday right now and not answering telephone polls. The ones who are, are mostly Conservative or NDP, so you're going to end up with an oversampling effect.

He also pointed out that their individual riding polls are more bullish because they are similar to Mainstreet's, which have a discrepancy with their national picture. He also said that the party expected to take a hit for the early election call that they felt should fade by early September. Which also happens to be when the cottage dwellers come home, when the debates are held, and when their platform book is due out. This cottage theory sounds so incredibly far fetched yet is fascinating.

There was also a Toronto Star article where some Liberals speaking on background said that they conceded the "runaway majority" had faded. If this is true, those candidates were playing semantics, because that's a majority, albeit not a runaway. Also, the Liberals' own internals beat the prospectives of all the major seat aggregators in 2015 and 2019. I have not seen JT campaign in safe Liberal ridings that the CBC and 338 indicate are in danger of falling to the Conservatives or NDP, so going by what you've said about watching the parties, that seems to line up.

I'm grabbing for anything I can call "copium" at this point, but putting those puzzle pieces together did come out to something that at least is interesting.

What's your take?

ottlib said...

I have a blog post swimming around in my head about this that I will "put to paper" once it gels and I can confirm a couple of things.

However, my take is what I have been saying for quite some time. The Liberal Party data analyst team have been crunching the numbers and seeing that, yes they have seen some slippage but not enough to become very concerned about. All that being said, I would caution anybody from putting too much stock in what some random person writes on Twitter or Reddit.

As I stated in this post the pollsters saying the the Liberals are way down while the Conservatives are way up are still saying the Liberals would win the most seats. That is BS, considering the actual polling estimates. That tells me that the pollsters may be questioning their own data. And as you say Mainstreet is showing better riding polls for the Liberals than their national numbers would indicate, which makes me think I am not out to lunch on this.

The Liberal Party and the PM have not looked uncomfortable since the election started despite the bad polls and a hostile media. As well, I watched all of the footage from yesterday and if that Justin Trudeau is the one that shows up at Liberal Party events for the remainder of the campaign Erin O'Toole and Jagmeet Singh are not going to know what hit them. He was passionate and he seemed to be much more engaged than he has been for the last two weeks.

Finally, the whole cottage theory is plausible but what is also plausible is the fact that conducting a rolling poll means you are conducting between 400-500 interviews per night, across the country. That means about 160 in Ontario and much lower for all of the other provinces. With such low sample sizes the risk of oversampling/undersampling is very high. However, here is the kicker, it can go both ways. Right now it seems it is going to the Conservatives' advantage but the Law of Averages states that eventually the Liberals will enjoy the benefits of it. Personally, I have no problem with the Conservatives benefiting from it during the first two weeks of the campaign if the Liberals can benefit from it during the final two weeks.

Jackie Blue said...

Well, Quito Maggi just tweeted that it's possible or even likely that the CPC are peaking too early. 🤷‍♀️ Could be he's referring not so much to the law of averages, but the law of gravity. Fingers crossed. 🤞 🙏

ottlib said...

LOL, translation: I do not trust my own data.