I think we can safely rule out the possibility of Bloomberg going third party, since the man himself has already done it on record. Bloomberg despises Trump, and he’s said many times that he’s going to do whatever it takes to defeat him. You’d have to believe Bloomberg is a deep cover Republican asset to think there’s a chance he’ll run as a third party candidate.

As for betting markets, the research suggests they’re pretty accurate, only in the absence of public polling. Our election campaign is already drenched in polls, and it’s going to get worse, so I don’t put a lot of confidence in what the betting markets are telling us.

I don’t think I claimed to be attempting any sort of analysis, but my arguments were fairly uncomplicated. Premise 1: Trump drives big Democratic turnout, and will bring more voters on both sides to the polls than Pence would. That’s hard to argue against, based on the 2018 midterms and more recent polls about which voters intend to vote in November. Premise 2: When turnout is high, Democrats win elections. This has been borne out empirically for decades now, and it’s also supported by the observation that Republicans are always the ones trying to suppress voter turnout. Premise 3: Mitch McConnell needs Republicans to win in order to stay in power. Given that McConnell wants to stay in power, the conclusion is that he made a big mistake by keeping Trump in office.

And the second argument had to do with the impact of negative partisanship on this year’s campaign. Premise 1: In November 2020, voter turnout will be largely determined by opposition to the candidates, rather than positive support for them. (This is a claim that was supported by links to academic research, but it would require a whole lot more analysis to make a prediction as to just how important voter opposition will be.) Premise 2: Democratic voters oppose Trump much more intensely than they oppose Mike Pence. Premise 3: If Trump had been removed from office, Republicans would be likely to vote in larger numbers, because their levels of opposition to Democratic lawmakers would be higher. And Democrats would be likely to vote in lower numbers, because they don’t oppose Pence as much as they oppose Trump. Conclusion 1: Mitch McConnell made a choice that will bring more Democrats and fewer Republicans to the polls in November, compared to another choice that was open to him. Premise 4: McConnell wants more Republicans and fewer Democrats to vote in November. Conclusion 2: Therefore McConnell made a big mistake.