People v. Belkin
Defendant was driving his Maserati at 145mph when he crashed into the back of another vehicle, killing the other motorist. His blood alcohol level measured 0.315.
After pleading guilty, defendant was sentenced to 25 to 40 years' imprisonment for the second-degree murder conviction and 10 to 15 years for the OWI conviction.
Trial court erred by assessing 10 points for offense variable (OV) 1 (aggravated use of a weapon), one point for OV 2 (lethal potential of a weapon used or possessed) when the prosecutor argued that the vehicle was used as a weapon. (Frankly, this was just a dumb argument by the prosecutor.) Nonetheless, correcting these two errors did not change the sentencing guidelines.
An interesting argument was raised on appeal by the defense, claiming that MADD statistics reflect that a drunk driver is not likely to be involved in a deadly motor vehicle accident. The appeals states:
Defendant's reliance on statistics from Mothers Against Drunk Driving (MADD), which he asserts demonstrate that the risk he created while driving drunk did not create a likelihood of causing death or great bodily harm, is misplaced. The data from MADD shows that on a daily basis, there are 300,000 incidents of drunk driving leading to 28 deaths. Defendant argues that this correlates to a 0.009% chance that an incident of drunk driving will lead to death and, therefore, that death or great bodily harm was not the "probable" outcome of his actions.
It is no criticism at all of MADD's laudable mission to say that its citation of the drunk-driving death rate is not designed as a predictive model of the profile of any particular driver, and whether or not that particular driver is likely to cause death by driving drunk. Rather, MADD's use of that particular statistic is part of its mission of public education, to define the scope of the problem, in support of its public policy proposals. A purported predictive model, as defendant seeks to use the statistics, would lie in the realm of the social sciences and the proper application of statistical models which would control for the different variables at play in drunk driving offenses, which include blood-alcohol level; the degree to which a defendant's blood-alcohol level exceeds the legal limit; the differing blood-alcohol levels which states have criminalized; accidents involving alcohol only, drugs only, or a combination of the two; and the manner in which a defendant drives. See generally Bazemore v Friday, 478 US 385, 400; 106 S Ct 3000; 92 L Ed 2d 315 (1986) (discussing the use of regression analysis generally to control for variables regarding a particular party's actions). Defendant's use of statistics gathered by MADD merely lumps all drunk driving offenses together and then purports to draw some meaningful analysis at the granular level as to the likelihood of a particular result in a particular case, a clearly invalid use of MADD's data. Defendant made no attempt to demonstrate the reliability of the data for the purposes for which it was offered.
By choosing data defining the scope of the national problem, defendant also ignores other MADD statistics which might support a different conclusion. For example, MADD notes that "every 51 minutes, someone is killed" as a result of drunk driving, and "drunk driving is still the number 1 cause of death on our roadways." MADD also notes that every two minutes, someone is injured in a drunk driving crash. The MADD data, applied to defendant's conduct, which so egregiously deviates from a typical drunk driving offense, may well support a finding that it was quite likely, even overwhelmingly likely, that death would result. Defendant's blood-alcohol level was 0.315 (or nearly four times the legal limit); he was not merely speeding—he was driving a high-powered sports car in excess of 140 miles per hour; and he was on the phone immediately before the accident bragging about his speed. The victim's car was moving in the same direction as defendant's own at 53 miles per hour; the force of the impact which defendant precipitated was the equivalent of hitting a stationary individual at more than 80 miles per hour. And defendant engaged in such a conduct on a surface street, not a highway, which is designed to handle higher speeds more safely, although certainly not to anything approaching the level engaged in by defendant. In light of the MADD statistic that more than one person an hour is killed by a drunk driver, and the facts of defendant's actions the night of the incident, we think it highly likely that in that one-hour period, no other drunk driver in America demonstrated conduct more likely to cause an accident; and that if a collision were to occur, striking the victim at the equivalent of a car moving at 80 miles per hour striking a person at a standstill, the result would have been the certainty of the victim's death. So while we reject defendant's use of the MADD statistics, and do not rely on any of the MADD data ourselves, we do note that based on defendant's generalized citation to it, and assuming the reliability of it, the data might well support the opposite conclusion from that for which defendant argues.