Frank Viola was a very good MLB pitcher and was always a favorite of mine. There was his World Series MVP winning season in 1987, CY Award season in 1988, his 20-win season for the Mets in 1990 and two good seasons for mediocre Boston teams in 1992 and 1993 when I had Red Sox season tickets. Then after his playing career was over he became a minor league pitching coach, one who seemingly had good success working with pitchers on the farm for the Mets.
Yet somehow, Viola never made the majors as a coach. Yeah, there was the heart condition that required surgery back in 2014. But he came back later that year and continued on with the Mets through the 2018 season. He had some interviews with other organizations but never landed a gig. And then this past season, not only was he not in the majors, he wasn’t even in affiliated ball.
Viola spent the 2019 season as a pitching coach in the Atlantic League.
Now, the Atlantic League may not be the last stop in professional ball, but you can likely see it from there. If you’ve heard of the Atlantic League, it’s likely because MLB signed an agreement with the independent league, which essentially was going to act as a proving ground for some initiatives that MLB was interested seeing introduced on a competitive basis.
By far the most interesting of these rule changes instituted by the Atlantic League in 2019 was the adoption of an automated strike zone, something the league used the second half of its season. The league utilized the TrackMan system, which uses Doppler radar, best known for its work in weather forecasting, to determine if a pitch was inside the strike zone.
The joke almost writes itself. Only MLB would use a system best known for unreliable weather forecasts to try to get balls and strikes called on a consistent basis. And whaddya know – the people in the Atlantic League didn’t care for the TrackMan system. John Harper’s piece at SNY yesterday had quotes from numerous sources, primarily by old pal Viola. The piece closed with this quote from the pitching coach:
So I’ll put it this way: I’m not the biggest fan of umpires, but after getting to see what the robotic system looked like, I’d love to be able to argue with an umpire again.
It’s never fun when you find out that a guy you like has opinions that don’t line up with yours. It’s my opinion that the biggest challenge that MLB faces is not time of play or the shift but rather that umpires miss too many ball and strike calls. It’s not necessarily the fault of the umpires. We’re asking humans to do a job 300+ times a night that they’re not necessarily great at doing.
MLB needs an automated strike zone. But it was never realistic to think that the first time that a computer model was put in place that it would be foolproof and better than the human system that’s been around for roughly 150 years.
It’s not a direct comparison but let’s look at Watson, the question-answering computer system developed by IBM to respond to questions on the quiz show Jeopardy. In 2004, the idea was hatched to create Watson. According to Wikipedia, in initial test runs in 2006, “Watson’s first pass could get only about 15% correct.”
By 2011, Watson was able to beat Ken Jennings and Brad Rutter, two of the most successful Jeopardy contestants at that time, in game action.
IBM had already created Deep Blue, which was an AI system that played chess and eventually was able to beat Grand Master Garry Kasparov. And despite all of their programming success and the amount of manpower and materials they were able to dedicate to developing Watson – it still took five years after the initial deployment before it was able to perform at the highest levels and better than the best humans, as it defeated Jennings and Rutter.
Maybe TrackMan isn’t the right system to use to call balls and strikes. Harper’s piece indicated MLB may move to the Hawk-Eye system, which makes in-or-out calls in tennis. And that’s okay. You keep trying, looking to improve however and wherever you can.
The disappointing thing, to me at least, is that MLB Commissioner Rob Manfred, a man seemingly intent on creating a legacy, doesn’t consider a computerized strike zone to be a top priority. Maybe that’s for the best. Perhaps this will allow the designers of the system to work out the kinks at a reasonable rate and not have the computerized model forced into duty at the MLB level until it’s completely ready.
At the end of the day, my preference is for the reason my team wins or loses a game to be completely dependent on the level of their skill on the field and in the dugout. Too many times it feels like an umpire’s blown calls has too much sway over the final outcome. No one wants to see the Mets lose a game because the umpire can’t get strikes called correctly in the 18th inning. Or any inning.