First polls for the BC election show the NDP ahead but less than previously

We finally got some new polls for the recently started BC election. Overall they continue to show what we knew: the NDP is ahead and in majority territory. However, the lead of Johan Horgan over the BC Liberals of Wilkinson is smaller than what we could have expected given previous polls.

Research Co. published a poll, conducted between September 21 and 23 among 750 people. It shows the following:

Then Insight West published their own poll, conducted between September 22 and 23rd among 1000 residents (a high number for a BC poll) with the following numbers:


The NDP is ahead in both polls but the margins are very different. Maybe not surprisingly, the poll with the BC Conservatives at only 4% -- a level that is a lot more realistic -- also has the BC Liberals much higher.

So I'd say that I believe the Research Co. poll much more here. Polls in BC always have the issue of overestimating the BC Conservatives. It's most likely because respondents don't realize it's not a 'real' party like the federal Conservatives. To give you an idea, in 2017, the BC Cons only ran 7 candidates. To be fair to them, they have been more organized in the past and have had relative success. Still, at this point, I don't think there is any chance this party is at 12%. So a lot of the 12% in the Insight West numbers will likely end up being added to the Liberals on election day. The two polls are therefore not as different as you might think. The Insight West one also has a lot more undecided (20% versus 10%) and once I allocate those undecided not proportionally (I currently give 50% to each of the Liberals and NDP), I get smaller differences between the two polls.

What is more interesting is the NDP at 'only' 42-44%. This is quite lower than most polls published during the last few months (where they'd be closer to 50% in some). This should worry John Horgan. If he's below 45% at the beginning of this election, after benefitting from a covid-boost for the last few months, he could end up regretting this snap election.

Regionally, they also have similar results although I'll need to ask them to tell me precisely how their regions work (what exactly is 'Vancouver' and how far does the region go). They have the NDP well ahead in Metro Vancouver as well as on the Island, while the rest of BC is a close race. This region is the most surprising to me and I'll need other pollsters to confirm the NDP is that competitive outside of the Lower Mainland and the Island. The polling average has the NDP at 39% in the rest of BC, same as the BC Liberals. In 2017 those numbers were, respectively, 32% and 49.8%!

So, anyway, here are my projections for now. They are still very basic and not based on much data, but it's a start. The model will be updated as the campaign progresses, for now it's a very basic one. But such basic models usually perform not too bad anyway. As usual, feel free to use the simulators to make your own projections.

It's a NDP majority but things are closer than we might have thought. If the Liberals can climb closer to 40%, we might even get a race that will ultimately depend on a few seats (heeelllo Surrey...).