As I indicated in this post, I'm not convinced by the arguments that a computer system ought to replace human umpires for calling balls and strikes. Rather, I think human umpires are the best way to determine balls and strikes.

Before I launch into the full argument, though, it is important to distance myself from a certain type of justification. Arguments from tradition often fall victim to the is-ought problem, and so even though I do think baseball is particularly (and appropriately) conservative in its traditions, I will not attempt to justify the use of human umpires based on the fact that they have traditionally been the method of calling balls and strikes. 

Rather, I think the argument for human umpires must ultimately derive from baseball's social purpose. (I think this is true of all questions about baseball's rule structure, including the All-Star game, interleague play, the designated hitter, and so on.  But we can save those for another day.) Consider, for a moment, the question of baseball's ultimate utility. Forgive me if this seems obvious, but I don't think it is arguable that the goal of baseball is entertainment. By entertainment, I do not mean the base satisfaction of wants, but rather the rich spectrum of pleasurable experiences, ranging from drunken yelling to a sunny day at the ballpark to sublime literature. If baseball did not satisfy the goal of entertainment, it would be useless.  In any event, the answer to the question "why baseball?" simply cannot be "to create a maximally efficient and internally consistent set of rules."

Proponents of computerized umpiring base their arguments on fairness. Certainly, fairness and entertainment are not mutually exclusive. Quite the contrary, if we could plot the fairness of a system of rules for various games against their respective entertainment values, I'm confident we would find a positive correlation. However, that is not to say that each move in the direction of fairness necessarily entails an increase in entertainment value.

All this leads to the (I hope) uncontroversial conclusion that arguments from fairness alone will not suffice to do away with human umpires. However, it may still be the case that increasing fairness in this case would increase the entertainment value of baseball.

So let us compare the two alternatives on the metric of entertainment value.

Computers

1. Certainly, computer systems would create fewer infuriating moments, and thus perhaps a more satisfying experience. This could make the game more entertaining.

2. There is also a longer-term benefit (particularly to the sort of fan who frequents sites like this one) of knowing that the game is ultimately designed to reward skill in a disinterested manner. This could also lead to greater overall entertainment.

These two benefits are not small, but I believe they are the only arguments in favor of computerized umpiring given that we are evaluating options for their entertainment value.  It is noteworthy that most of the benefits are gained from eliminating problems with human umpires, and not virtues in and of themselves.

Human Umpires

Human umpires, on the other hand, have obvious flaws. Human psychology, particularly selective memory (the tendency to remember outlier events and thus overestimate how common they are), reinforces these flaws. But umpires have benefits as well:

1. One of the best moments in a baseball game is when an angry manager ambles up out of the dugout to argue with an umpire--and this is especially true on a balls-and-strikes call. Because players and managers are explicitly forbidden from arguing balls and strikes, when someone does, it is particularly exciting because the chance of ejection is increased. 

2. Different umpires have slightly different strike zones, and so there is a skill to be learned in how to pitch to a particular umpire. It is perhaps arguable that this reduces the entertainment value, but I believe that certain players, most notably Tom Glavine, are Hall of Famers not because of their stuff or their control (though those are necessary components) but rather because they understood how to pitch to a particular umpire's strike zone. I don't think that's a deficiency; I think that's a real virtue, because it requires players to be smart. As I re-read Dan's excellent 2008 post on close calls, I was struck that perhaps the correlation between, for example, age and frequency of favorable calls, was in fact causation triggered by some unmeasured skill. And watching a really smart pitcher expand the zone is very entertaining.

(Perhaps computer systems could have some variable, chosen at random before each game, that would vary the size and shape of the zone. However, I doubt this would actually happen.)

3. I am convinced that home plate umpires practice their punch-out motions (perhaps even in front of the mirror). And they are great. If computer systems called the balls and strikes, no one would watch the umpire after a third strike, and so the practice would all but disappear. This would be a tremendous shame, and would certainly make baseball less entertaining.

4. Finally, I think the very anger that many of us feel at seeing a bad call may contribute to baseball's entertainment value. Granted, this is probably my weakest argument, but there is something enjoyable about righteous indignation at the bloody unfairness of it all. Without any gray area, what in God's name would we argue about? And aren't these arguments fun to have?