Daily Box Score 10/16: What Do the Playoffs Tell Us?

Playoff baseball is excellent. Does anyone deny that? Even casual fans become glued to the TV come October. For sheer excitement value, you really can't beat the playoffs.

But a question lurks behind the playoffs. The answer to the question seems so obvious that it almost always goes unasked. Because I'm the sort of person who cares about questions no one has bothered to ask, this question has been bothering me.

The question is, what do we learn from the playoffs about the quality of the participating teams?

Table of Contents

Discovery Function
Gauntlet Theory
Positivism
Discussion Question of the Day

Discovery Function

One possible answer to the question above is that the competition of the playoffs tell us something we couldn't learn simply by studying the teams. By watching them in competition, their true abilities and talents come to the fore, and all relevant information about a team's strength is reflected in the outcome of the contest.

This idea is not specific to sports. The Austrian economist Friedrich Hayek proposed a radical explanation for the superiority of free markets. Unlike planned economies (like the Soviet Union of his day), where central planners had to try to guess all the right prices, he argued, free markets were able to combine all available information and properly allocate resources. How could such alchemy proceed?

In his landmark 1945 essay, "The Use of Knowledge in Society," Hayek argued that prices in a free market represented all of the information in an economy on a relevant product. Best of all, they came pre-balanced, maximized, and efficient.

Think of it this way: the price of a gallon of milk contains the information that only a dairy farmer in Wisconsin has about how much it costs to fertilize the land where his cows graze (I'm assuming cows graze; they do graze, don't they?). Because he is only willing to be a dairy farmer so long as it feeds his family, and the same is true of the factory workers at the bottling plant, the final product's price represents the input costs of producing the good.

How is this relevant to sports? Well, it turns out that Hayek thought the discovery function of free markets applied to sports as well. In a throwaway line in that paper, he wrote:

It would be patently absurd to sponsor a contest if we knew in advance who the winner would be...The only reason we use competition at all has as its necessary consequence the fact that the validity of the theory of competition can never be empirically verified for those cases in which it is of interest.

This, you might say, is a corollary to the bar-stool wisdom that unexpected outcomes are why "you don't play the games on paper." (Except, of course, when you do.)

The point is, it is impossible to know the strength of a team until you play the games, and in fact the entire purpose of playing the games is to determine which team is best. It's not simply that predicting success is hard, it's that it simply fails to do the thing we want it to do, which is to gather up all the difficult-to-aggregate information about players' skills and sum them up in the wins and losses column.

If this is true, Hayek seems to be suggesting, then the winner of a championship is not merely given the title of champion. In fact, the only meaningful way of knowing who deserves the championship is to play it and see who wins. Put another way, winning a championship necessarily implies the team deserved it.

But Adrian Vermeule disagrees slightly. In footnote in a recent paper (this is why you should always read footnotes!), he writes:

Games, unlike examinations, may not be a means of uncovering independent information; if the game is played according to its rules, the outcome is necessarily correct.  However, I am unsure of this point.  It seems perfectly coherent to affirm both that the point of the annual Wimbledon tennis tournament is to determine who is the best tennis player at a given time, and also that in a particular year, the winner of the tournament was not the best player. 

This is a serious problem for Hayek's view. It may be the case that the goal of a free market is to aggregate all relevant information even as it fails to succeed in that goal. Similarly, it may be the case that a playoff system is designed to reward the best team even as it awards the championship to an inferior team. 

The Hayekian rejoinder here goes something like this: perfectly functioning free markets do aggregate all the information, and perfectly functioning playoff systems always reward the best team with the championship.

But can we really design a perfectly functioning playoff system, given real world constraints? (I'll bite my tongue on the question of whether we can design a perfectly functioning market, but needless to say, it's an interesting question.)

Gauntlet Theory

Given the length of the playoffs as it is, is it reasonable to think we could extend them? Seven game series take up to two weeks a piece, and even then the very best teams only beat their competition about 80% of the time. So the question is worth asking, could we realistically do better?

One counterintuitive suggestion I'm somewhat fond of is to shorten the playoffs. If we shrunk the playoffs to four teams, the teams who made it would be better. In turn, that would increase the odds that the very best team in fact won the championship. Under the current regime, that rarely happens.

At the risk of outing myself as a political junkie, I'll give an analogy from politics. The justification given most often for the grueling and long process of presidential races, from pre-Iowa fundraising to election night speeches, is that running a long campaign is a good way to see who the best president would be. The better a candidate is able to survive the rigors of a two-year campaign cycle, the better a president she will be. 

It turns out we've already got a grueling and long baseball season. It's 162 (or sometimes 163) games long, features doubleheaders, day games after night games, and plenty of ups and downs. Why not use the regular season, and let only those teams who compile the very best records into the playoffs?

Two objections appear immediately. First, entertainment. As I said above, long playoffs are fun! Even with the risk of cheapening individual games, more playoffs means more excitement. As Dane Cook used to be fond of reminding us, there's only one October. 

A second objection is cold, hard, cash. Everyone makes more money the longer the playoffs go. That's why there are a gajillion (approximately) bowls in college football, why even the Atlanta Hawks make the NBA playoffs, and why there are eight teams in the MLB playoffs.

Positivism

Mao Zedong once said, "Political power comes at the barrel of a gun." 

One possible solution, since it doesn't seem the current playoffs system is going anywhere in the near future, is to assert that the playoffs define who the champion is because MLB has the legitimate authority to say so. This is the Chairman Mao view of the playoffs.

It's perfectly reasonable to say both (1) the strongest team doesn't always win the World Series, and (2) the team that wins the World Series is defined to be the best team that year. 

On a certain level, I think this makes sense. If you were stronger, losing team, then why the heck didn't you win? However much we might like to point to third-order winning percentages or TQI or whatever metric we choose, the team that gets the rings is the team that wins the championship. 

It's also a little tautological. To say that the team that wins the championship is the champion may be boring, but it might be all we've got. But if that's the case, then it takes a lot of pressure off sabermetricians. If we don't have to worry about who the best team is, or which team ought to have won, we sit back and enjoy the action.

Now, you must excuse me. The Phillies are on.