dibo
Well-Known Member
Yeah, I don't think that's right. There's some pretty fanciful looking seat-by-seat polling.
If the polling is right, we're going to have a national result around 51-52 plays 48-49 and seat results more consistent with a national result of 55-60 plays 40-45, which would basically mean each side's seats would have to swing *hard* to the extremes and there won't be a marginal seat left in the country.
More likely, the MOE on marginal seat polling is way bigger than reported, mostly because we're new to large scale robopolling of small constituencies and haven't worked out the sampling and methodology kinks yet.
National robopolling seems fine (not least because you can compare against a bunch of other national polls taken by various methods), but the local stuff seems pretty whiffy.
If the polling is right, we're going to have a national result around 51-52 plays 48-49 and seat results more consistent with a national result of 55-60 plays 40-45, which would basically mean each side's seats would have to swing *hard* to the extremes and there won't be a marginal seat left in the country.
More likely, the MOE on marginal seat polling is way bigger than reported, mostly because we're new to large scale robopolling of small constituencies and haven't worked out the sampling and methodology kinks yet.
National robopolling seems fine (not least because you can compare against a bunch of other national polls taken by various methods), but the local stuff seems pretty whiffy.