It's the opposite. In a FSM you have clearly defined states and all possible transitions between them. If you write the same code as normal imperative code tha

It's the opposite. In a FSM you have clearly defined states and all possible transitions between them. If you write the same code as normal imperative code that modifies some variables - to understand what it does you have to "run it" in your brain and reverse engineer the states and transitions yourself.