Robo-ump Short Circuits, Starts Calling Games Like Angel Hernandez

by Alex Poletti

Humanity’s rapid technological advancements have turned against their creator once again, as an electronic umpire in the independent Atlantic League short circuited and began to call games like MLB umpire Angel Hernandez.

“The robo-ump was working great until the fourth inning,” Atlantic League president Rick White says. “Then the pitcher threw a curveball in the other batter’s box. The robo-ump called it a strike. After a while, it became very inconsistent and it wouldn’t let the home plate umpire override the calls. It had a mind of its own.”

The Atlantic League has garnered a reputation as one of MLB’s leading testing sites for new
rules and technology, the robo-ump chief among them. Implemented during the league’s All-Star Game earlier this month, the robo-ump is a function of the TrackMan Computer System, which features a radar utilizing the Doppler effect, similar to the one used by StatCast for gathering advanced metric data.

According to programmer Sean Gupta, the program also includes an artificial intelligence system. Though not complex and rather minor, Gupta finds it possible that the AI may have taken over the machine altogether.

“It’s a minute part of the system, but it’s technically possible that the robo-ump could develop a mind of its own,” the programmer explains. “Based on the inconsistency of the strike zone and a very stubborn demeanor, we don’t think it developed a mind of its own; rather, we think it redeveloped the mind of Angel Hernandez.”

Hernandez has developed a reputation for being one of the worst umpires in baseball, calling strikes right down the middle as balls. While The Second String acknowledges that a Major League umpire has a hard job, it’s not as hard as Angel Hernandez makes it out to be.

So how could such a mistake happen? Gupta hypothesizes that data inputs from MLB might have included Hernandez’s pitch-calling; it’s probable that the AI latched onto his specific information and developed a schema for the strike zone that was entirely inaccurate.

“We input every at-bat from every umpire in Major League Baseball over the past five years into the AI system to give it a general sense of how the strike zone should be formed. We thought because there is a bit of variation among umpires, it would help settle a general sense of the zone. It looks like it was an overload of information, though. It’s similar to how Microsoft released an AI on twitter and it became a terrible neo-nazi racist; AI systems seem to have a tendancy to settle for the lowest common denominator.”

The strike zone wasn’t the only problem with the robo-ump, unfortunately. When coaches began to approach home plate umpire Brian deBrauwere about the technological deficiency, they were ejected by the computer system.

“Coach made contact to TrackMan,” the AI message read, despite the TrackMan not having a physical presence. “Warrants ejection. You’re out of here.”

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s