This is a sneaky trick question.
I think one of the things that can be hard for people to understand at first is that in most districts, whoever is the most popular party and whoever is the second most popular party will both have their representatives elected in that district, which is great, because it means that many voters are fulfilled. If the winning candidate got 40% of the vote and the second-place candidate got 30% of the vote, then you have 70% of the voters in that district who have the representatives that they voted for, which is a very nice feature.
In dual-member proportional, sometimes it happens—and this is in a minority of districts—that it is the first-place and the third-place candidates, or the first-place and even the fourth-place candidates, who can be elected. You can imagine that that might confuse and upset people in that district, but I think it's understandable because of the provincial balance. If people vote for the second-place candidate in a district, and if that second-place candidate didn't win, then it would be because the party of that second-place candidate was awarded their seats in places where their party did better than in that particular riding, and so the third-place and fourth-place candidates would be the strongest showing for that party across the province.
One of the nice things about dual-member proportional is that the district where that happens changes every election. You wouldn't get one geographic district that is consistently disadvantaged over time. It might be that one riding gets their first-place and third-place candidates this election, but the next election it'll be their first-place and second-place candidates.
In learning about DMP and really looking at the models and studying it, I think that it is by far the strongest system that I've seen. I was spending a number of years working on climate policy in different countries, and I thought to myself, “What would my ideal electoral system look like?” I had a back of the envelope idea. When I was considering my submission to the provincial process, I was thinking that I'd sketch out this idea, but then I read through the other submissions and I found Sean's report. I thought this guy had written exactly what I would have written had I had a two-year period to research, report, develop the system, and test the model mathematically.
I think it's a very robust system. The weaknesses of DMP are very slight compared to the weaknesses of other proportional systems. Every proportional system, every system, has its pluses and minuses, and I think that DMP has the fewest weaknesses.