The author of the original article misquotes the 30% statistic. According to the PDF he linked to [1], 30% of American motorist fatalities are "speeding related

The author of the original article misquotes the 30% statistic. According to the PDF he linked to [1], 30% of American motorist fatalities are "speeding related." This is not the same thing as "caused by speeding." According to the NHTSA, speeding in this context is defined as "Speeding is reported in the Fatality Analysis Reporting System (FARS) as a driver-level attribute that combines 'driving too fast for conditions' or 'in excess of posted speed limit.'" [2] Note that, in multi-vechicle accidents, only one party needs to have been speeding for the accident to be considered "speeding related". The 30% figure quoted also seems to vary quite a bit depending on the data you look at: "The percentage of all crashes that were speeding-related (DTFFC or EPSL) varied considerably among the States, from about 6 percent to about 20 percent of all crashes" (ibid) The NHTSA distinguishes between two types of speeding: 1) Exceeding Posted Speed Limit (EPSL) 2) Driving Too Fast For Conditions (DTFFC) The former simply occurs whenever one's speed is greater than the posted maximum. The latter is a subjective call by the police officer, etc investigating the case and factors in road, traffic, and weather conditions. The NHTSA has found (ibid) that about half of speeding-related fatal crashes occur for EPSL (55%) and the other half for DTFFC (45%). For speeding-related crashes that result in injuries, the respective ratios are 26% and 74% (ibid). The same source mentions that "The relative proportion of crashes that occurred on the curved sections of the road was much higher in speeding-related (DTFFC or EPSL) crashes" which suggests that many of these accidents were actually caused by either some form of driver inattention or adverse weather affecting road conditions, and that speeding was a contributing factor in the accident (i.e. less ability to correct for an unexpected event) but not the sole cause. Since driver inattention is almost impossible to prove absent records of an untimely text message or tweet, it's likely that the true causes of most of these accidents (most likely driver inattention or overly agressive driving) go unreported and are instead attributed to the only measurable data point that can be discerned: speeding. [1] http://www-nrd.nhtsa.dot.gov/Pubs/812006.pdf http://www-nrd.nhtsa.dot.gov/Pubs/812006.pdf [2] http://www-nrd.nhtsa.dot.gov/Pubs/811090.PDF http://www-nrd.nhtsa.dot.gov/Pubs/811090.PDF