A Little A/B Testing on American Politics, Anyone?

There’s a lot that’s great about the U.S. political system. So much that I wont bother to go into it here. But there’s also a LOT that’s bad about it, and that bad part isn’t getting any better. That’s not controversial to say – it’s actually the one thing that everyone in the country agrees on, whether they’re left, right, center, or off on some other axis entirely.

And yet though there is a lot that is bad, at the very highest, most abstract level, much of this I think boils down to one central problem: our political system isn’t designed to produce compromise.

It’s not designed for politicians to find common, win-win answers to the open questions facing us. Instead it’s designed (unintentionally) to produce just the opposite – polarization, playing to political bases, and actually penalizing anyone who seeks compromise.

A “win” in American politics isn’t moving the ball forward, it’s taking the other team down.

I don’t think that’s what the people who put the system in place had in mind, but it’s where the system has evolved to. And it’s why people often refer to the government as “broken”.

This is clearly a design problem. Nothing less, nothing more.

The good part is, design problems can be fixed, right?

Well, they could if we were talking about the web – just try some variations, do some A/B testing to find out what works better than our current design, and move forward. But of course we’re not talking about the web, we’re talking about a Constitutional Republic. The idea for those was “perfected” over 200 years ago, and hasn’t been iterated on a whole lot since.

But why not do A/B testing on politics in America?

Why not do some isolated tests on small parts of the system, aiming for better results than we’re getting now? If the tests pan out, roll them out more broadly. If they don’t, scrap them and revert to the current system. Try lots of things. See what works. Base decisions on solid evidence. Just the way we do when we try to make the web work better.

Of course to do this we’d need to agree on a common set of guidelines as to what “better results” would be. It couldn’t be “my side gets its way more often”. It would have to be something along the lines of moving the ball forward faster and finding more win-win solutions to disagreements.

I’m not knocking the people who created our fine political system, but I bet that we could A/B test our way to an even better version of it in a very short period of time with the right resources.

It’s a big idea. An explosive idea. One that people might reject out of hand, just out of pure nervousness over breaking something that isn’t totally broken now. But if we don’t do that, how will we ever make the system better?

Or is that it? Are we stuck with what we’ve got?

Original post