Also: "1% of people used that thing that we broke" is often a willful misinterpretation of metrics where the actual situation is more like "100% of people used that thing that we broke, 1% of the time."
The real interpretation is going to be somewhere in between, but the people choosing the direction the want the project to go are going to choose the interpretation that supports the decision that has already been made.
edit: The best way to kill a train network is to tear up all of the least utilized track. The vast majority of that track consists of the last few miles of travel, and although only 1% of people use a particular piece, 80% of trips use one of the pieces. If people are going to have to buy a car to get to the heavily utilized backbone, they're going to ask themselves why they're driving to a train rather than just driving to the destination. This leads to a decay of the backbone.
The real interpretation is going to be somewhere in between, but the people choosing the direction the want the project to go are going to choose the interpretation that supports the decision that has already been made.
edit: The best way to kill a train network is to tear up all of the least utilized track. The vast majority of that track consists of the last few miles of travel, and although only 1% of people use a particular piece, 80% of trips use one of the pieces. If people are going to have to buy a car to get to the heavily utilized backbone, they're going to ask themselves why they're driving to a train rather than just driving to the destination. This leads to a decay of the backbone.