Does comparing raw number of deaths here make sense without taking into account the total distance travelled / time spent per person on these different kinds of roads? Not to mention other factors like e.g. how tired/inebriated/distracted the average person using these roads is, which is unlikely to be the same for short distance and long distance travel.
Also 744 < 606 + 318 so it doesn't even seem to be "more than half" even with some road types missing.
Also 744 < 606 + 318 so it doesn't even seem to be "more than half" even with some road types missing.