“The more things change, the more they stay the same” could easily be the title of this latest BC election forecast. Over the past three weeks, the race has seen a deluge of policy announcements from the NDP, Conservatives (BCC), and Greens (BCG), one televised and one radio debate, and the leader of the Conservatives claiming to have seen someone die from a drug overdose on the streets of Vancouver (the veracity of this story remains under question).
But, most importantly to me, the last three weeks have produced ample polling results to sift through. And, despite all the political manoeuvring, the data is telling us that the topline of this race has not changed all that much; this is still an extremely close race, with a slight advantage held by the NDP due to their more efficient vote distribution.
After taking into account the many polls that have been released since our previous forecast was released, our current projection forecasts the same number of seats for the NDP, BCC, and BCG as the previous forecast; 49 for the NDP, 43 for the BCC, and 1 for the BCG.
The greater Vancouver area still goes NDP over BCC by a 2:1 margin, and the BCC have picked up one seat in this region since the previous forecast. Vancouver Island is still dominated by the NDP, while the rest of BC heavily favours the BCC – the model has the BCC winning 26 of 29 possible seats in the region, down by 1 from the previous forecast.
I’m going to repeat a line about models from our previous forecast – election models are far more informative when viewed probabilistically rather than as a singular predicted outcome.
When looking at the range of possible outcomes for the two main parties, there is still a significant overlap between their win ranges. There’s been a modest lengthening of the tails of possible outcomes for both parties, meaning the range of the number of seats each party could win has expanded since the previous forecast. This tends to happen as more polls come out and the variance of vote intent for each party increases.
In our current projection, the floor is 38 seats for the NDP and 36 for the BCC, while the ceilings are 55 and 53, respectively. The NDP win between 44 and 49 seats in 65% of the election simulations, and the BCC win between 42 and 47 in 64% of cases.
When we run 5,000 election simulations, the NDP secure a minority government in 5% of cases (-1% from the previous forecast) and a majority in 54% of cases (+1%). The BC Conservatives win a plurality in 5% of simulations (no change) and a majority in 31% (+2%). In 4% of the simulations (-3%), both parties win 46 seats, with the BC Greens winning one seat and holding the balance of power.
The biggest change since the previous forecast is a decrease in the number of seats considered to be Safe wins. The previous forecast classified 13 of the NDP’s projected 49 wins to be Safe wins, and their Safe seat count is now down to 8. The number of Safe BCC seats is down from 27 to 20. There’s been a large increase in the number of Lean seats for both parties, which are seats that are not as close as Toss-ups, which are essentially a coin toss, but are still going to be competitive and could feasibly be won by more than one party.
If this is the only source of BC election analysis you are consuming, I am honoured. For those of you that are supplementing your election news consumption with this analysis, I’m going to tell you something that you already know – this election is going to come down to the ridings in and around Vancouver. The map below shows the current projection for each seat in this region. You can hover over each riding to see the current projected winner – but, more importantly, you can see how few ridings are categorized as Safe ridings.
The next two maps show Vancouver Island and the rest of the province. You can see the orange wave on the island, and the blue wave that spreads across the rest of the province, Vancouver notwithstanding.
Reach out to us if you’d like to discuss what this election campaign, and the next government, will mean for you and your organization.