I never thought I’d vote for Hillary Clinton, and I really never thought I’d be excited to do so, but here I am, voting for her today: It turns out that she’s by far the sanest choice in an insane landscape. Most political commentary is really about signalling (including people who say, “most political commentary is really about signalling,” since they’re making a point and trying to signal their intelligence). Still, to understand why I vote for sanity, consider Jonathan Rauch’s argument in “Political Realism: How Hacks, Machines, Big Money, and Back-Room Deals Can Strengthen American Democracy” and “Left-Leaning Economists Question Cost of Bernie Sanders’s Plans.” The former explains that it turns out some level of cash handouts actually make politics function much better.

Many of you may remember the fight over “earmarks” from the 2000s—that is, whether Congresspeople should be able to allocate cash for specific purposes in their districts. I thought eliminating earmarks would be a good idea. Turns out I was wrong: Eliminating earmarks means that it is much harder for party leaders to keep their members in line.

As a result, we get more and more moves towards ideological purity, at the expense of, you know, making the country run. Congress has broken down in the last decade or so in part because party leaders can’t discipline their followers by taking away money that should go to members’ districts. The Tea Party, and, on the left, Bernie Sanders, can become more prominent because of that issue.

I’m not the first person to notice this—”How to fix what ails Congress: bring back earmarks” is one good account—but it is a serious problem that has caused particular dysfunction among Republicans, who appear ready to nominate people who are manifestly unqualified for being a big-city mayor, let alone president.

Ideological purity turns out to be very bad. Jonathan Haidt’s “The top 10 reasons American politics are so broken” (and “The Ten Causes of America’s Political Dysfunction“), along with the paper he links to, “Why the Center Does Not Hold: The Causes of Hyperpolarized Democracy in America,” explains why.

The papers I’ve been citing also explain why it turns out that Obama has been a much better president than most people, including me, realized, or thought in advance. He’s an incrementalist, a negotiator, a thinker, and a realist—all traits that are not selected for amid political polarization. Clinton is too.

Trump and Sanders espouse opposing policies (to the extent Trump espouses any policies), but both are alike in that they are “outsiders” who want to tear down existing systems; they are both temperamentally similar in that they don’t want to work within existing systems. Both are poor traits in leaders and figureheads. We want evolutionaries, not revolutionaries.

It may simply be that, as Matt Yglesias argues, “American democracy is doomed.” I hope not, but the bout of insanity on right and left does not auger well.

I haven’t dealt much with the specifics of Sanders policies apart from the second link in this post because the short version of the critique is, “There’s no way to pay for all this stuff.” Or even a small amount of this stuff. Alvin Chang observes, probably correctly, that “Most Bernie Sanders supporters aren’t willing to pay for his revolution.” If you ask most people if they want more services, handouts, and stuff, they say yes. If you ask most people if they want lower taxes, they say yes. Stated in those terms, you can see the problem.

If you want to understand that people don’t want to pay higher taxes, look at where they’re moving. The major population growth metros are in Texas. Phoenix and Atlanta do really well too. People are moving to lower-cost states, and that should tell us something important about revealed preferences. Hell, I just voted in the New York primary, and I’d like to move to Austin or Nashville, chiefly for cost-of-living reasons.

As for Sanders and banks, the bigger issue than “big banks” is the “shadow” banking system, which I don’t fully understand, but I do understand well enough to know that Sanders is wildly focused on the wrong things.

Still, the last two paragraphs probably don’t matter because the vast majority of Sanders voters aren’t looking at policies; they’re looking at mood and feelings, and I doubt that 1 in 20 people who start this post will have gotten this far, because it’s wonkish, detailed, and not heavily mood affiliated. Out of the who, what, where, when, why, and how of politics, the “How” is often most important and least discussed.

“First, do no harm” is a good political rule, but it’s also kinda boring. I rarely write about politics because most of the time most politics in the U.S. are about incremental changes, about which I have some opinions, but those opinions aren’t important, and they’re as poorly thought out as the political opinions you see on Facebook. In the last couple presidential cycles, I’ve had opinions and voted accordingly, but the major party candidates have mostly been kinda okay. In 2012, Romney would’ve done some things differently from Obama, but I don’t think he would’ve been a total disaster.

This cycle is scary because at least two candidates would likely be total disasters. Yet people keep voting for those candidates and posting mood affiliated comments on them and so forth.