Wildcards
Last night's Red Sox victory in the ALCS has gotten me thinking about, of all things, the academic job market. What, you ask, is the connection between the two? Not surprisingly, the answer for me has to do with how the two systems organize and process competition. As I try to think through the issue of what an agonistic approach to knowledge would actually look like in practice, it occurs to me that what bothers me about academia is similar to what bothers me about baseball. It's not the game/metagame thing I was talking about in the previous entry, or not precisely that. Instead, it's the fact that both the academic job (and publishing) market and the present major league baseball system are structured to allow too much
randomness and not enough
contingency. That's what bothers me about both domains of action.
The Red Sox made it into the playoffs this year as the "wildcard" team, which is the team with the best record that didn't win its division. This change was made in the early 1990s in order to promote a more "exciting" end to the baseball season, modeled on the successful "March Madness" playoff system in college basketball. ("Exciting" here means "larger television ratings and more ad revenue," not necessarily "better on-field play.") As a result, it is now quite possible for a team to make it to the World Series and win it without having amassed the best record in its league over the course of the year, and this now happens regularly. The last two World Series were won by wildcard teams (the Florida Marlins in 2003, and the Anaheim Angles in 2002), and this may be the second time in three years that the World Series is played between two wildcard teams if Houston defeats St. Louis tonight.
What's the problem here? The wildcard system -- the playoffs in general, but especially with the introduction of the wildcard -- introduces a great deal of
randomness into a season of baseball. By "randomness" I mean that the outcome of the competitive interaction between teams is reduced to a roll of the dice; there is very little that one can do to affect that outcome in any meaningful sense. The genius of baseball as a sport is precisely that, although the outcome of any particular game is unpredictable due to random fluctuations, the regular season as a whole is designed (whether by accident or on purpose is something that I am still trying to determine) to eliminate those random fluctuations, by having teams play enough games over the six-month extent of the season that good teams rise to the top of the standings and weaker ones fall to the bottom. Baseball statistics are
meaningful, precisely because there are enough measurable situations that one can draw conclusions that stand up. A shorter season, or a design that emphasizes particular games, magnifies the importance of random fluctuations, and the outcome of any such situation is effectively a crapshoot.
Any team in baseball can beat any other team in any
particular game, but over the long haul,
not any team can amass an excellent overall record.
What this design does is to make winning and losing
contingent on factors like team design and balance, and on the way that particular players are deployed during the whole campaign. By "contingent" I mean that the outcome of the competitive interaction between teams depends on things that those teams
do rather than on factors completely beyond their or anyone's control. The regular season system differentiates between good and bad teams by making the outcome of the campaign contingent on what those teams do; this involves both on-field game and off-field metagame, and as I argued before I have no problem with this as long as the metagame is related to and in a sense subordinate to the on-field game itself. One bargains hard in the off-season free agent market in order to enhance one's on-field performance; one trades during the season for the same reason. And all of this activity comes to a head on the field over the long haul: teams streak and slump, but in the end the team with the better players and better strategy prevails.
The playoff system, especially with the wildcard, is random rather than contingent. Any team can prevail in any finite number of games; any team can win a five-game series, and any good team can win a best-of-seven series. Which one actually wins is
not contingent on the fundamental soundness of the team or the ability of its players, but is instead basically random: even the best pitchers have bad starts, and even the greatest sluggers have poor-hitting slumps. So we get the absurd notion of "clutch players": people who supposedly perform better under pressure, when everything is on the line, which is basically is all the time in a short series that will decide whether or not one advances to the next round. Sabermetric studies have confirmed for years that there ain't no such thing as a clutch player; no one in the major leagues performs significantly better in the "post-season" than they do in the regular season, once one takes into account the difference between playing a handful of games and playing 162 games. Short spans of time distort a player's record and mis-state his ability; short series, or a set of series stacked so that each one is do or die, are nothing but an arena for pure random chance to take over and determine who goes to the next round. Fortuna -- Machiavelli's evil bitch goddess -- is given free reign.
The academic job and publishing market is like this too, in that whether one gets hired or published is basically a matter of random chance instead of being contingent on one's abilities and long-term ability to compete and perform. An article sent to three reviewers might have gotten a firm reject had it been sent to three other reviewers. A candidate might have made the short list for interviews for a position if her or his advocates on the committee had not been out-maneuvered by the supporters of another candidate, or if they hadn't cut a deal to bring in or hire a compromise candidate. My book is about to go to contract, after a multi-year struggle to get it reviewed; the reason for this is that a colleague with whom I was having a conversation at a conference that I do not normally attend happened to mention the book to an editor that he knew at a press that I had not previously approached; the editor liked the concept, read the manuscript, and arranged to have it reviewed by people not fundamentally hostile to the approach that I take in the work. Random. The book being published has zippo to do with its quality, or importance, or logical soundness; it has to do with a chance meeting and a chance conversation. Ditto my job here, which owes more to the presence on the hiring committee of someone I knew from before, who could vouch for me and keep my file near the center of the committee deliberations…
What bothers me here is that the outcomes in each case are a matter of chance instead of being the sort of thing that you can
do something about. Yes, there's a networking game that you can play to try to get articles reviewed favorably or place students in jobs (or in graduate school: same basic principles apply here). But this is a much trickier game, without firm rules and lacking a clear strategy for success. Baseball -- regular-season baseball -- has a simple recipe for success: hitters with high on-base percentage and slugging percentage, and pitchers who induce outs by not walking people and by striking hitters out (and perhaps also by getting more groundouts than flyouts, although the debate rages on about the importance of this factor). Score runs; don't permit your opponent to score runs. Easy to articulate; hard to achieve, but at least
feasible and
knowable, as well as being within the sphere of a team's control (again, to the extent that the metagame is "fair," which at the present point in time in baseball probably requires a more aggressive luxury tax on team payrolls and a better distribution of advertising revenue).
The academic job and publishing market is even more random than baseball with the wildcard system. I write an article and send it off to a journal, and I might as well be making it into a paper airplane and launching it from my office window. There are some steps I can take to help maximize its chances of seeing the light of day in print, but in the end whether or not it happens is a matter of sheer random happenstance. It's like winning a particular game: whether you do well on a given day or not is largely unpredictable. A system based on contingency is also unpredictable, but in a different fashion: a contingency-system makes the outcome dependent on something within the actor's sphere of control, while a randomness-system makes the outcome dependent on things quite beyond the actor's control.
Then there's always a third option: the "rigged" system, in which someone always win and someone always loses because the outcome is determined in advance. No agency in this system. Note that bargaining situations in which people have fixed preferences are "rigged" in this sense, as no outcome other than the one pre-determined by people's objectively-given interests can possibly come to pass. Agency, however, is contingency, not randomness; a randomness-system is just as problematic for a meaningful conception of social action as a system in which the outcome is determined in advance.
Thank God that we still have the regular season. I wonder what a regular season equivalent in academia would look like.
[Posted with ecto]
