If automobile accidents were even remotely related to how much drivers drink, then it’s precisely when drivers drink that they would be expected to have an accident.  If in the past 30 days an average of 12% of all drivers drove within 2 hours of drinking, and 21% within the last year did, then we would expect at least 12-21% of all accidents to involve drinking drivers.  The simple fact that “police reported alcohol involvement” proves that 94% of fatal accidents involve ONLY sober drivers, and that only 6% involve drinking drivers, indicates that the DRINKING driver is the SAFER driver.