Sports are a terrific lens for focusing in on the process of decision-making — they’re almost laboratories for incentives. Consider the problem of instant replay in Major League Baseball.

Instant replay lengthens the games, and longer games are arguably detrimental to baseball’s efforts as a whole. All else equal, a season of games with instant replay will take longer than a season without by the amount of time needed to review the plays.

However, the likely larger time issue is the change in incentives that have resulted from the replay system itself.

Delays now exist at the end of any sufficiently close call. Typically, the “losing” team’s manager ascends to the top step of the dugout and signals the umpire that he is considering requesting a replay of the previous call. The manager is then allowed a brief window to contact their replay advisor. Should he get the go-ahead from his colleague, the official replay process then commences.

Insofar as viewers prefer less downtime during a game, these recurring delays represent a significant cost to them. For managers, however, there is essentially no cost of considering a replay, and the only cost to the manager of losing a replay is the inability to go through the process later in the game, because challenges are limited — and don’t forget that this cost must be weighed against the possible gain of the call being reversed in the manager’s favor.

In other words, managers capture much of the benefits of replay and transfer most of the costs onto the viewers.

In economics, we often speak of superior social outcomes when individuals are able to make decisions that capture all of the costs and benefits of their actions. To that end, what might be some solutions to the instant replay/cost transferring matter?

  1. Eliminate instant replay. By definition, you can’t have costs related to instant replay if you don’t have instant replay.
  2. Institute a limited time window for challenging a call. As it currently sits, there is an ill-defined window by which to issue a challenge to a call on the field. If the instant replay/time problem is rooted in managers being able to pre-review each play before officially issuing a challenge, simply remove their ability to do that, time-wise.

    If a manager only had, say, three to five seconds after the end of a play to issue a challenge, only plays that appeared particularly egregious to the naked eye would likely be challenged — as opposed to plays that appeared right in real time but were arguably incorrect upon a frame-by-frame replay. In fact it is exactly these latter-type replays that seem to cause the most consternation amongst opponents of replay.

  1. Penalize a team for losing a challenge. As it sits right now, replay is a win-only proposition — either the call is overturned in your favor or the status quo prevails. Increasing the costs of an action means people do it less; making a lost replay costly would reduce the incidence of replays.

    It wouldn’t need to be draconian; maybe lose a strike or ball on the next batter. (I would expect fines here to be of minimal impact, though admittedly the idea of having to pay, say, $30,000 if a challenge is lost is mildly amusing.)

  1. Create a market for challenging calls. As of right now, only one side can institute a challenge. What if the other side more strongly prefers that the play not be challenged? The currency here is the interesting part: what if the Cubs were willing to pay one strike on the next batter to challenge a call—only to be outbid by Cleveland’s offer of two strikes to avoid the challenge?

    This is not dissimilar to #3; connect costs back to the decision makers. The equilibrium prices for challenges in different scenarios within games and throughout the season would be fascinating.

In my mind, #1 isn’t going to happen any time soon; replay is far more likely to expand than go away. #2 may well be on the horizon. #3 and #4 are at least incentive compatible, if exceedingly unlikely, and #4 especially would be entertaining to watch — especially for economists.