40 to 50% of all fatalities involve the circumstances I've described. The last year I checked was about 36k fatalities, and a little more than 16k of them matched that pattern.
Another confounding, but obvious point, is that alcohol and risky driving is also more of a factor at night than it is during the day. People also might assume that a motorcycle rider is _less_ likely to commit a DUI, when it's actually the opposite.
Once you take risky single vehicle accidents out of the picture, the statistic that really starts to stand out is pedestrians. They go from 16% of all deaths to around 30% of the remaining deaths. These are also most likely to happen at night, and a surprising number happen outside of urban areas; and it wasn't clear from NHTSA data alone whether that was due to speed of the vehicle being higher or the distance to a hospital being greater.
One odd point that I didn't have time to chase down is it seems that elderly drivers and passengers are much safer in the front seat. There was a strong skew for their fatalities to be found in the rear seats of vehicles. The profile was something like 80 year olds in the rear, and 40 year olds in the front. This seemed like a particularly dangerous configuration for elderly passengers.
The other interesting data point I gathered is that Texas had a higher number of fatalities over California, even though it has a smaller population. It suggests that state laws and road design are also a large factor in fatalities.
Anyways.. NHTSA publishes the FARS database. Every fatality is recorded, and if it was investigated all of the data is reported in a standardized format. They even have a "sequence of events leading up to the fatality" coded in there. It's an amazing resource.
Another confounding, but obvious point, is that alcohol and risky driving is also more of a factor at night than it is during the day. People also might assume that a motorcycle rider is _less_ likely to commit a DUI, when it's actually the opposite.
Once you take risky single vehicle accidents out of the picture, the statistic that really starts to stand out is pedestrians. They go from 16% of all deaths to around 30% of the remaining deaths. These are also most likely to happen at night, and a surprising number happen outside of urban areas; and it wasn't clear from NHTSA data alone whether that was due to speed of the vehicle being higher or the distance to a hospital being greater.
One odd point that I didn't have time to chase down is it seems that elderly drivers and passengers are much safer in the front seat. There was a strong skew for their fatalities to be found in the rear seats of vehicles. The profile was something like 80 year olds in the rear, and 40 year olds in the front. This seemed like a particularly dangerous configuration for elderly passengers.
The other interesting data point I gathered is that Texas had a higher number of fatalities over California, even though it has a smaller population. It suggests that state laws and road design are also a large factor in fatalities.
Anyways.. NHTSA publishes the FARS database. Every fatality is recorded, and if it was investigated all of the data is reported in a standardized format. They even have a "sequence of events leading up to the fatality" coded in there. It's an amazing resource.