All of this writing and data analysis is a lot of work! So after more than five years of posting, I've finally launched a Patreon to help pay the bills.

2/7/20

Just attack all of Bernie's opponents

When your primary is just contest between two candidates, there isn't much room for strategic innovation: the correct approach is obviously to just run down your opponent's numbers while bringing up your own.

But since 2020's Democratic primaries have been hotly contested by three or four candidates, there have been lots of opportunities for people to come up with some pretty elaborate strategic takes.
For example, consider these reasonably common takes on Mayor Pete:
1. We should hold our fire on Mayor Pete, or perhaps even help him, because he splits off voters that might otherwise rally behind Joe Biden and Elizabeth Warren. 
2. However, we can't promote Buttigieg too much, because he almost won Iowa and is surging in New Hampshire and could, if things go poorly, consolidate the anti-Bernie vote.
To be specific about this: keeping these considerations in balance would probably mean that we should keep Pete at just below 15%. Much more than that and he hits the viability threshold and starts winning delegates; much less than that and he is splitting off an inadequate number of voters from Biden and Warren.

On that note, consider some common takes on Warren:
1. We need Warren to drop out as soon as possible because she takes a disproportionate number of voters away from Bernie. The longer she stays in, the more she does to stifle his momentum, and we also can't rely on delegate pooling to bail Bernie out. 
2. But it might be helpful if Warren stayed in a little longer, at least through New Hampshire, where all of Bernie's opponents are near the viability threshold. If she peels off enough votes from Biden and Buttigieg to keep them below 15%, then she is helpful even if she splits off more votes from Bernie, because in that scenario Bernie would still get all of New Hampshire's delegates. 
3. Oh but if Biden or Buttigieg get above 15%, then we should want Warren out, because in that scenario she peels off a disproportionate number of votes from Bernie without pushing his opponents below the viability threshold. Here, she does more harm than good.
All of this makes perfect sense! But the more rigorous our strategy is, the clearer it is that what is good for Bernie is extremely dependent on all kinds of very specific considerations that change from moment to moment and state to state. If Pete Buttigieg ends up surging, then the strategy of cleverly boosting him early on will in retrospect look extremely stupid. If Warren, Biden, and Buttigieg all get 14% or less in New Hampshire, then we will be happy that Warren stayed in, since her presence almost certainly split votes to keep them below the viability threshold. But if Biden and/or Buttigieg get just one point more, then we would have been better off with Warren gone. And of course, it's entirely possible to imagine scenarios where Warren with high numbers and Pete with low numbers helps Bernie in the short-term but end up screwing him over in the long term - and vice versa.

Here, the point I want to make is that while this kind of strategic analysis is fun to toss around in online bullshit sessions, it does not actually work as a plausible guide for action. This is simply because we cannot control the polls precisely enough, from moment-to-moment, to engineer optimal state-by-state outcomes. There is no way for us to keep everyone else's polls exactly at 14%, which would clearly be optimal; often, we don't even know if particular candidates are above or below the viability threshold, and if we should try to push their numbers up or down. This may often end up being something we only notice in retrospect.

Moreover, even if we did know exactly where candidates were polling at any given moment, we only have very limited control over what happens to their numbers. If Warren is polling at 18% and all of the other candidates are at 13%, we know that we should push her numbers down a bit - but what if we unintentionally push them down to 10%, giving Pete and Buttigieg so many defectors that they both hit 15% and take away delegates from Bernie? That could totally happen if some criticism of Warren proves too effective - but since we can't really control how effective our criticism is, we can't avoid this kind of outcome.

As a rule, the upshot of this kind of speculation is that we should either hold our fire on some opponent of Bernie's or ramp up our criticism - and then if someone fails to fall in line with this strategy, we can accuse them of working against him, and insist that the only true Sanders supporters are the people who've bought into our analysis.

But since we can't actually control the polls with the kind of precision you would need to make this sort of strategy work, this kind of eleven-dimensional chess has no real relevance to whether Bernie Sanders wins or loses. 

Ultimately, there is only one strategic approach that can work regardless of what the polls happen to say at any given moment, and that's the same one you use when Bernie only has one opponent: try to bring his numbers up, and try to bring everyone else's numbers down as much as possible. If you bring everyone else's numbers down low enough then you don't have to rely on vote-splitting or other risk tactics to win. You don't have to boost Pete and hope that he doesn't built momentum; you don't have to push Warren up to 14% in New Hampshire and hope that she doesn't hit 15%. And you never have to worry about criticizing someone too much.

The other great thing about this strategy is that it lets you tell the truth about everyone. Bernie Sanders is the best candidate in this race and all of his opponents are deeply flawed. Just be candid about this! Take every opportunity you can get to talk about how great Bernie Sanders, his movement, and his platform are, and take every opportunity you can to highlight the problems with his opponents. Do this enough and you win. I don't think that any of this strategy is very controversial - it's plainly the strategy that Sanders himself has adopted, which is why his campaign has routinely criticizes all of the other candidates while insisting that his candidacy is superior.

The simple strategy is the most effective, it's honest, and it's on-message with what Sanders himself is doing. Why do anything else?