Democrat 2020

[Montmorency]

I dont disagree with you, if future polling in those states change I will happily change my opinion.

My point is that you're trying to extract a certain predictive value that probably isn't there beneath the surface.

Look at the polling for Clinton vs. Trump and Sanders vs. Trump in 2016. In March-April 2016, the period in which Sanders had any hope of securing a lead over Clinton, both candidates had some of their best-ever leads over Trump. Sanders' lead was bigger than Clinton's. That doesn't mean he was more electable at the time, since he was not in fact nominated. There would have been no reason to believe that Clinton's polling would narrow again but Sanders' wouldn't. You should also note that against both candidates throughout primary season, Trump had the same floor in the high 30s, which is probably more suggestive than the spread on any given day.
https://www.realclearpolitics.com/ep...nton-5491.html
https://www.realclearpolitics.com/ep...ders-5565.html

Speaking of spread, let's look at the actual polling for the 2020 general. If you refer to the post-Iowa polls (there are three of them) then the spread between Democratic candidates against Trump is 47-51 vs. 43-45.

On the basis of even this polling, the candidates are indistinguishable from each other against Trump. If you think these polls can tell you something about Election Day performance, they are telling you that the Democratic candidates are tied with one another. The tautologies of electability are just that, axiomatic circular reasoning with no independent validity.

If you still want to lean on these matchups, you can't do more than project Trump to have a ceiling of 45% in the general. Disaggregating that tentative projection to the state level will be entirely unattainable for a while. Trends in Trump's state-level approval may be more instructive than his head-to-head polling against hypothetical opponents.

Polling in one context (i.e. candidate vs. candidate) cannot responsibly be extrapolated to a different context (nominee vs. nominee). This is even more the case with state-level polls that are fewer in number and have higher margins of error. If you were presenting some kind of well-developed theory backed by data, that would be one thing, but to reason along the lines of 'Klobuchar is ipso facto more electable than Buttigieg because she polls 1% higher in matchups as of now' is fallacious. You just can't do that.

Bloomberg's money has achieved some results -- and not just in poaching staff.

His poaching staff probably cost Democrats a special election in Vermont Connecticut a month ago. A political newcomer was running for a state assembly seat, but Bloomberg hired her campaign manager out from under her less than 3 weeks before the election and she lost by 79 votes (1.6%). I mean, it doesn't affect the legislative math - Dems hold 3/5 of the seats - but it's got to sting the contender who got screwed by a distant, indifferent, behemoth. It's a veritably-Lovecraftian tale.

Bloomberg qualifies for next Dem debate

The qualifications have differed for each debate, and the party has consistently erred on the side of having (too) many candidates on stage. With his current polling it's not unreasonable that Bloomberg should have an appearance. I expect he gets savaged by a united front.