From the chart, going from 5 or 10 driver traffic deaths to 1 may be a simple statistical anomaly (and perhaps not even, it could be within the distribution). W

From the chart, going from 5 or 10 driver traffic deaths to 1 may be a simple statistical anomaly (and perhaps not even, it could be within the distribution). We would need to know the history of pedestrian and cyclist deaths too. The long term evolution seems meaningful, but that year may not mean anything.