Fairness

“That’s not fair!”

If I had a dollar for every time I’ve heard that at a tournament (including many times that I said it myself), I’d buy something rather nice. When someone sees something happening at a tournament that they don’t like, there’s a good chance that the complaint, whatever its origin and merit, will be couched in terms of fairness.

Fairness is, indeed, crucially important. If your participants see the tournament as unfair, it’s likely in for trouble, not matter what its other virtues. So it’s worth spending some thought, early on, to unpack the notion of fairness.

I’ll begin with my least favorite definition of fairness:

(A)  Fairness means doing it the way we’ve always done it. 

Here’s another, possibly more palatable, definition that gets at the same thing:

(A1)  Fairness means meeting the settled expectations of the participants.

Either way it’s stated, I find such a definition unappealing. If you really think that the right way to do things is the way they were done in the past, what’s the point of a site like tourneygeek? – the implicit message of the blog as a whole is that it ought to be possible to do things better than they’ve been done in the past.

But an appeal to the way we’ve always done it standard is common, and it often deserves to be treated with a certain amount of respect, especially if it can be recast as settled expectations. If players alter their performance in response to a reasonable expectation that the current tournament will be run as in the past, these settled expectations have become, in effect, part of the game, and changing the rules after the game has started is unfair.

So, at a minimum, this first fairness principal should mandate that the tournament organizers communicate clearly about how the tournament is going to be run so that everyone is, or at least could be, playing the same game.

Let me offer two other possible definitions of fairness:

(B)  Fairness means giving everyone an equal chance.

(C)  Fairness means rewarding superior skill.

Neither of these can stand on its own without some consideration of the other. If equal chance were the only criterion, the tournament organizers could award the trophy the way they would a door prize – to someone whose name is drawn out of a hat. Or if superior skill where the only criterion, they could simply hand the trophy to whomever, on the basis of past performance, is generally acknowledged to be the best player. Either way they could dispense with the messy business of actually playing games. A proper notion of fairness needs to include both equal chance and the superior skill elements.

And it’s important to acknowledge that these may be in conflict. Consider, for example, the issue of seeding. There are several aspects of seeding, and I’ll share some thoughts about them in a future post. But at its most basic, seeding involves manipulating the tournament to give an advantage to the better players. The better players may get byes, and advantages like home court or first move, and they’re likely to be protected against early loss because they won’t have to play the other good players in the early rounds. All of this makes it more likely that a superior player will win, and thus increases fairness under the superior skill criterion.

If equal chance is paramount, seeding should be avoided. Advantages like byes, home court, or first move should be allocated randomly, and the pairings should be made in a blind draw. Or, pushing things a little further, the organizers could use a kind of inverse seeding, where seeding advantages that usually go to the better players are given to the weaker players instead.

There is no one right way to strike a good balance between equal chances and superior skill  – it’s a matter of judgment. But I hope to show, in subsequent posts, that there are some reasonably unambiguous ways to improve on particular aspects of one or the other. In particular, I’ll show how reducing a tournament to an abstract model and running repeated trials of the model can show that some tournament designs are more conducive to rewarding superior skill than others. There are often plausible arguments in favor of either of two alternatives, and in these cases, such experimentation can show the right path.

One final thought. Though fairness is a very important consideration in tournament design, it is not the only consideration, and it will often have to be compromised in favor of some other legitimate goal. More on this to come.