Floorball is a type of floor hockey. During the game, substitutions can be made.
The team is also allowed to change players any time in the game; usually, they change the whole team. Individual substitution happens sometimes, but it usually happens when a player is exhausted or is hurt.
I would like to use an RNN to predict when the next substitution will happen for a team. However, I have no pre-existing dataset to train on. Is there a way that I can start predicting without a dataset and continually improve accuracy as more games are played?