Breaking

University of Tokyo Baseball Team Develops Its Own AI to Predict Batted Balls

November 7, 2026 at 01:36 EST

Research & Papers

The University of Tokyo baseball team has developed its own artificial intelligence system to predict the trajectory of batted balls, a tool aimed at helping the squad analyze and anticipate play on the field.

University of Tokyo · Collegiate baseball

Tokyo's University Baseball Club Builds Its Own AI to Predict Batted Balls

The club developed the system in-house rather than buying a commercial product. The goal is to help the squad anticipate where a ball will land, so fielders can position themselves before the ball is hit.

Why landing-point prediction matters

Fielders have fractions of a second after contact to decide where to move.

STEP 1

Bat meets ball

Contact starts the trajectory.

→

STEP 2

AI estimates landing point

The in-house model predicts where the ball will come down.

→

STEP 3

Defenders position

Better estimates allow more effective fielder placement.

→

STEP 4

Fielder reacts

Movement happens within fractions of a second.

What remains unconfirmed

Key details about the system have not been independently verified.

Training data

What data the model learned from is not confirmed.

Prediction accuracy

How closely predicted landing points match actual results is unknown.

In-game use

How players and coaches apply the tool in practice and games is unclear.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.