miles driven, not market share, is the correct comparison here (and even then you'd want to include many other covariates, such as where the cars are operated,

miles driven, not market share, is the correct comparison here (and even then you'd want to include many other covariates, such as where the cars are operated, socioeconomic status of drivers, etc, etc) if you really want to compare different vehicles.