Joseph White’s latest Eyes on the Road column looks at the VTTI‘s new crash research, where they “rigged up 100 cars with video cameras and other data recording devices to chronicle 42,300 hours of data and images of people driving in the general vicinity of Washington, D.C., and the Virginia suburbs”:
The study captured video of a total of 241 drivers ranging in age from 18 to over 55. Over the course of a year, the test cars got into a total of 82 accidents (69 of which were fully recorded), 761 near crashes and 8,295 “incidents,” defined as events requiring an evasive maneuver.
[...]
Based on data gathered from police reports, researchers had thought that 25% to 30% of accidents involved some kind of driver distraction. But the Virginia Tech study suggests that 80% of crashes involve some kind of driver distraction or drowsiness within the three seconds prior to the crash, the NHTSA said in a statement.
Further, many, many accidents are caused by a very, very small proportion of drivers:
One 18-year-old woman was involved in three crashes, 53 near-crashes and 401 “incidents.” A 41-year-old woman was involved in four crashes, 56 near-crashes and 449 incidents. Pause and take that in. In the space of a year or so, two people were involved a total of 116 crashes or near-crashes, and a combined 850 incidents that involved some sort of swerving or emergency avoidance maneuver. Consider that this track record of bad driving was compiled even though the motorists knew that they were being watched by a camera. This makes you wonder what the world would be like if really bad drivers could somehow be taken off the road.