Quote:
Originally Posted by octothorp
Yeah, the problem with what Brown's saying here is that this pattern of undersampling conservative voters doesn't necessarily hold true in Calgary. For example, every pollster massively underestimated Gondek's support in the municipal election, including Brown, who underestimated by 10%. That still made her closer than most including Mainstreet, but it strongly goes against the 'conservative voters are always undersampled' argument. Granted, she absolutely nailed the Calgary-level numbers in 2019 and so there's a reason to not change her process. But it's also possible that in Calgary there's more likelihood of moderate, centrist voters being a little more shy about left-wing support, in the same way that shy republican voters are a known phenomenon in US elections.
|
Agree with everything your wrote
I think one learning from Brown’s poll would be that in a polling method designed not to undercount conservative voters the UCP is ahead. If they were not ahead then you could say they were in trouble. I’m not sure you glean more than that.
338 polling average putting it as a 70-30 UCP-NDP likelihood seems reasonable. Flip a coin twice two heads, NDP wins.