<div>We've covered the real underlying values. Yet sometimes deciding which system is better by arguing from root values can be like the joke where the physicist tries to improve milk production by starting out with "assume that a cow is a spherical body with radius R". Rather than returning to first principles, sometimes it's easier to work from derived values - heuristics. The good thing is that there are several heuristics which tend to help satisfy several values at once. </div>
<div><br></div><div>In fact, heuristics might in some cases deserve more trust than the underlying values themselves. Sometimes there are paradoxical effects in values - for instance, as I discussed earlier, a more expressive ballot might give more chances for strategy and thus lead to a less expressive system. Also, people are used to being dishonestly marketed to in terms of values. If someone told you that this system is better than that one because it will result in "better" candidates winning - a utility-based argument - you might want to check their underlying motives before trusting them. Similarly if someone (FairVote?) told you they supported a totally different voting system because it would save money.</div>
<div><br></div><div>Heuristics are more resistant to these paradoxes and pitches. For instance, if a system is simpler, well, it's simpler.</div><div><br></div><div>Which brings us to our first heuristic: simplicity. A simpler system will almost certainly be more expressive, legitimate, and cheap. And, though the correlation is weaker, in my experience it's also more likely than not to be of better utility.</div>
<div><br></div><div>It's important to draw a distinction between two different kinds of complexity. There's fundamental complexity - like the rules of (American) football. In nineteen somety-whatsis, people started to do thus-and-so too much, so they had to invent a five or ten or fifteen yard penalty to stop them. And there's emergent complexity - a simple idea which leads inexorably to a complex process.</div>
<div><br></div><div>That's why, when talking outside the group of voting system enthusiasts, I am perfectly happy to talk about "the Condorcet method", meaning the whole set of summable methods which satisfy the Condorcet criterion. The basic idea of Condorcet is simple and powerful. And, as a practical matter, cycles for first place are certainly rare - certainly less than 12%, and probably only 1 or 2%.</div>
<div><br></div><div>Another particular kind of emergent complexity happens when you try to make a system summable. Suddenly, instead of talking about who beats who, you're talking about matrices and linear algebra. I'm facing that myself with SRBV (Summable Reweighted Bucklin Voting), about which I'll have some more posts soon. And I recommend that you should face that problem head-on. "Really, the basic idea is that you want to elect the person who XXXX. But in order to find that person without having to bring the ballots to a central location and trying out each candidate separately, the process gets a little more complicated. Here's the details..."</div>
<div><br></div><div>In particular, for Schulze voting, here's the pitch: "The basic idea is to elect the person who wins against all others. If there's no such person, you try to eliminate the minimum number of ballots until there is. But you don't want to have to bring the ballots to a central location and then try every combination of ballots to eliminate. So there's a process that is designed to almost always give the same answer, but can be done using a local count..." (now, if people ask, you can describe the method.)</div>
<div><br></div><div>I know, the beatpath is not always the same as eliminating ballots, if you have more than 3 candidates in the Smith set and [some other improbable criteria which are too involved to state here]. But for me, personally, I am more likely to support the Schulze method now that I understand it as a summable approximation of minimal-ballot-elimination. And for those who support it more than I do, I think that pitching it as such is honest and useful.</div>
<div><br></div><div>(I also hope that, when I fully expound SRBV, people will not be too turned off by the mathematical complexity, and they'll be able to understand that it's just a summable approximation for the dead-simple RBV.)</div>