If I’ve learned one thing while working with data, it’s that people are going to interpret it however they see fit. In the business world data can be manipulated to tell a story that benefits a company’s goals. In baseball, things like the PECOTA projections are hotly debated and fans either fall on the side of “these projections are gospel” or “these nerds don’t know what they’re talking about.”
Whichever side you fall on, please stop. The PECOTA projections are simply a collection of information splattered onto a page. They take past data and put it through an algorithm that attempts to predict future performance. Sometimes they’re pretty damn close and other times they predict the Royals are going to suck and then they win the World Series.
Old timers love when projection systems are off because they think every single new-age analyst relies on these numbers to do their jobs. What they don’t realize is the good analysts not only take what the numbers say into consideration but also the beloved eye tests of gray-haired baseball men. It’s a balancing act. Not to say there aren’t people who only take numbers into consideration, but you usually don’t hear much about them because they don’t have important jobs.
PECOTA is useful because it puts together tons of data and spits out what would happen if we all lived in a non-emotional robot world. It levels the playing field and gives fans a grander view of how their favorite teams or players stack up against the rest of the league. They’re not meant to be taken as the word of the lord, and they’re not meant to influence world leaders. Past data + prediction models = PECOTA. That’s it. Beyond that, people are going to draw conclusions based on their own opinions.
And those opinions usually suck.