Saturday, August 17, 2013

Instant Replay

I'm pretty sure I've ranted on this before. But it's time to rant again. Major League Baseball announced a plan for expanded instant replay. They aim to have it in place for next season. Now baseball people are all up in arms. Some defend tradition and the "human element." Others want things right. They're both wrong.

If a team of skilled observers cannot reliably distinguish outcomes as they happen in a game, then the game itself is the problem. In other words, a game which requires repeatedly viewing several angles of super slow motion high definition replays is a dumb game. Baseball is a great game because it produces very few situations which warrant such meticulous inspection. And on those that do, video evidence is often inconclusive anyway. In which case, we're back to the original solution: A tie goes to the runner.

Close scrutiny also forces us to think about really inane technicalities. In geometry, a line is infinitesimally small. In baseball, the foul line is a few inches wide of chalk or painted grass. What if one of those painted blades extends over the geometric line? And what if a seam of the ball grazes just that one blade of grass? To the naked eye in real time, it's fair if your team is hitting and foul if they're pitching. For players like this, I'd be perfectly fine with implementing some sort of laser or camera system to make a ruling on the fly. That would be far more effective than repeatedly watching the replay to decide if the ball actually struck the blade of grass or it was just the air moved by the ball.

What about the obviously blown calls? There are calls that are definitely beyond an acceptable margin for error that are missed. Those calls should still stand, but Major League Baseball should be completely transparent in acknowledging them. MLB should review every call in every game. Then, they should post whether replay confirmed the call, showed it was wrong, or was inconclusive. I'd also like to see each call categorized as trivial (say a fly ball), indeterminate (something as close as my blade of grass example), or somewhere in between. The in-between ones are most important. Then you could say how many total calls were blown. You could see how many of those calls favored or opposed your team. You could see which umpires were the most chronic abusers. In the YouTube era, this would prevent the handful of terrible calls from overshadowing the majority that are correct.

With all of this data publicly available, MLB would face pressure to stay ahead of the Bill James analytic types. What patterns are there? What plays are most commonly missed? Are there calls that would have been correct with one more umpire? One less umpire? What if an umpire is noticeably better at calling bang-bang plays at first but calls a lousy strike zone? Would he be better off permanently stationed at first? Or would umping the same position ultimately make him less effective? I'm sure they do this to some extent now, but by giving fans access to the data, they'll look pretty dumb if they don't figure it out themselves.

I'm all for getting calls right. But in baseball, you really only have one shot at it. Let's get it right as it happens.

No comments:

Post a Comment