Artificial Intelligence: Meet the potential scouting director of the future

One of the most important ways AI can help baseball teams in the next decade is by improving team drafting and player evaluation models.

And there may be few tasks more difficult in baseball than improving a draft model, because the very nature of the baseball draft makes it difficult to know for years whether a pick was good or bad.

In other sports, it doesn’t take long to know how a draft should have lined up. And even in baseball, the best picks quickly become apparent. The Angels didn’t have to wait long after the 2009 draft to know that picking Mike Trout 25th overall was a great choice.

But too often the success or failure of a pick hasn’t been known for years. Consider the top of the 2017 draft. The Twins took high school shortstop Royce Lewis #3. 1 ever. Even six seasons later, it’s hard to say how astute that pick was, especially as Lewis has battled injuries.

Now, it appears the Reds are picking the No. 2 wisely by picking high school righty Hunter Greene a spot ahead of the Padres pick of prep southpaw MacKenzie Gore. Going into 2020, however, Gore was arguably the best pitching prospect in baseball, while Greene was struggling with injuries and getting back on track.

Braves selection of Vanderbilt right-hander Kyle Wright at No. 5 in general has increased and decreased equally. He struggled for several years but broke out in 2022, his sixth season as a pro.

Six years later, we were just clarifying the 2017 draft. To properly evaluate the 2021 and 2022 drafts, we will have to wait many more years.

If teams are looking to see how well their draft model lines up with a draft board, they need to look at drafts from a number of years ago. So they’re looking at drafts from the 2010s to best design who they want to draw to be in their lineup in the 2030s.

The challenge of this task is obvious, especially in a sport where the conditions of the game change. The liveliness of baseball seems to have changed several times over the past decade, which affects which players are successful. Stolen bases were extremely rare in the 2010s, but have returned due to rule changes in 2023.

So whenever a team decides to let the AI try to model their draft model based on potentially millions of iterations that are tested over and over again to find the right weighting for potentially hundreds of thousands of different data points, a team willing to use an AI-powered draft model will take a leap of faith, knowing that they won’t know how well the model performed until five or more years into the future.

WHAT IS A DRAFT TEMPLATE?

Long before there were draft and player rating templates, there were scouting directors. And in their heads, scouting directors had to have a model shape, long before the first team ever bought a computer.

To align a board of directors, a scouting director had to evaluate how strongly different scouts aligned their preference lists. The opinion of a veteran scout who always seemed to have pitchers pegged needed to be weighed more strongly than a first year scout who was still establishing himself.

Then, the playmaker must decide how to field a less athletic, but more productive college player versus a more promising and more accomplished college player who hasn’t been as successful in his career.

Oh, and the best player is looking for a $2 million more signing bonus than the player the scouting director thinks is nearly equivalent in terms of talent. That extra $2 million in bonus pool money could lead to signing a couple of late picks.

When a scouting director was lining up a team’s board of directors, he was building a draft model in his head. Figuring out how to weigh a multitude of variables has long been part of the job of scouting directors.

But we humans struggle to hold too much different information at once in our brains when making decisions. This is where computer models come in handy. They can comfortably handle a large number of variables and data points at the same time, helping to synthesize a multitude of weights for different factors before collating them all into a numerical value or ranking order.

That’s why most MLB teams now use algorithms and computer modeling to help align their draft boards. The difference between current project boards and AI-driven boards of future potential is that current models are programmed by humans who determine how different information is weighted.

In the future, it’s possible that artboards will be built entirely by computers.