In The News

In The News

Wednesday, November 8, 2017

The Fallacy of the Coin Toss: .08 is “On the Money”

By Michael Nichols
Categories: Michael J. Nichols


Many lawyers, even judges have said to me: “what do you do with that if the guy is right on the line?” The line means a chemical test result that is right at the threshold of the so-called “per se” legal limit. There are some things that need to be discussed because the justice system has allowed statistics and measurement to be misused for too long.

Lawyers who practice heavily with whom I commiserate frequently on national listservs, in emails and elsewhere tell me what I already know: “a .08 is a lot harder to get a not guilty verdict than people think” (.08 means .08 grams of alcohol per 210 Liters of breath). That is my experience a well. The reason in my experience is that when the prosecutor is “stuck” with a .08 they drill down harder on certain areas during testimony. One of those areas is a complete fallacy in many respects and I was fortunate enough to have a judge who “got it” in a recent trial.

The Michigan Breath Alcohol Testing Program features the Datamaster DMT. This instrument was designed and manufactured by National Patent Analytical Systems (NPAS). John Fusco and Dave Radomski (mostly Mr. Radomski) designed the instrument so that it reports the measurement result to the thousandth decimal place i.e. .08 is actually .080 or .084 for example. The Michigan State Police (MSP) opted to “hide” that 3rd digit.

What I find with a .08 case is that the prosecutor will often call in someone with slightly more knowledge than the arresting officer/breath test operator, who then testifies about the datamaster “truncating” the 3rd digit. The prosecutor then says “does the datamaster always round down?” 2 things are offensive and misleading that every lawyer and every member of the public, who is a potential future juror, must learn.

To say that it “always rounds down” simply is not true either scientifically or statistically. Ask yourself: “what is the probability that the number is greater than .08? Well, it must be more likely than not that it is .081 or higher, right?” Wrong! Wrong. Wrong. Wrong. Wrong.

The probability that the number is anything other than “0” is absolutely random. We simply do not know what it is. Recently in a trial in mid-Michigan, the prosecutor asked Class IV operator Marvin Gier, who performs the 120 day checks and services the instruments on the east side of the state, “does the datamaster always round down?” Marv: “yes.”

I objected and moved to strike. The judge called us to the bench. I explained the above. Before we went to the bench I asked to voir dire (ask foundational questions) of Mr. Gier. “Mr. Gier: is it a completely random probability that the 3rd digit is 0, 9 or some digit in between?” Mr. Gier: “YES.” “To testify that it is any of those numbers would be speculating by you, right?” Mr. Gier: “Yes.” When we approached, the judge just wanted to know what my concern was and what I wanted struck from the record. The prosecutor begged. The judge said “no.”

A powerful moment followed. The judge said “the jury will disregard the witness testimony that the datamaster rounds down.” Mr. Gier was honest and refused to speculate.

Later in the trial, an expert in the science of measurement testified that the chance that the true value of the test subject’s breath alcohol content being EXACTLY .08 is zero. Put differently, there is a non-zero chance that the measurement result, such as .08, is exactly .08. The reason is a very simple concept that so many people misconstrue. When we measure something, we never really know exactly what the true value of the thing that we are measuring “the measurand” really is.

The reported value, such as the bathroom scale, really only represents a range of values, within which the true value falls with a high probability. Your weight for example. You stand on the scale and it reveals 150 lbs. You do not like that report. You step off and step back on and you get 145. What is the true value? We know what the mean value is – which is 147.5. The true value is more likely to be closer to 147.5 but we still do not know whether we are 150 or whether we are 145.

If you were to weigh yourself once and you are right at 150 lbs, then the chance you are above 150 or below 150 is exactly “50/50.” Here is why: the measuring instrument is never perfect even if it is working as well as it is supposed to work. If you want perfection, you will never get it. If you want to be able to make a decision about whether, for example, you need to lose 5 lbs in order to make a certain weight class for a competition then you need to repeat your measurements and understand the measurement uncertainty of your bathroom scale. Is it 10%? Is it 7%? And how confident that I can be that if I weigh myself 10 times in a row, the weight reflected will be an accurate depiction of what my weight really is within that percentage?

Here is where this discussion could delve deeply into statistical analysis, student’s t-test and standard deviations. However, the point to leave you with is the basic concept that a chemical test result in any toxicological discipline that is used forensically carries with it a coin-toss probability of reflecting a true value that is above or below a legal threshold. Put differently, at .08, can we “infer” that the person is guilty of operating in violation of the “legal limit?” Flip a coin.

As a postscript: the prosecutor offered in the trial story here to dismiss the drunk driving allegation and allow a young woman, who recently graduated from college, to benefit from treatment under the Holmes Youthful Trainee Act (HYTA) with a guilty plea to “disturbing the peace.” This means that if she stays out of trouble for 30 days, the case will be dismissed and her arrest record will be suppressed and available only to courts or law enforcement in the future … that is absent a bureaucratic snafu in the Michigan State Police records division. Nothing is perfect.

Need a Lawyer?

Get an online consultation or call 517-432-9000

Online Consultation »

Do the Medicines You Take Criminalize Your Driving?

Specially-trained police officers called "drug recognition experts" or "DREs" are spreading across Michigan...

More »

Personal Injury / Traumatic Brain Injury Experts

Traumatic Brain Injury (TBI) is one of the most frightening and serious forms of injury...

More »

Criminal Defense

We are skilled, experienced and committed to resuts in both the serious and misdemenor criminal case

Personal Injury

We have successfully represented clients with serious and traumatic injuries

What our clients are saying

more testimonials »

Peer Recognition

Mike Nichols is a national leader in drunk driving defense. He is a member of the Forensic Committee and Michigan delegate to the National College for DUI Defense. He is also a Sustaining Member of the College. Nichols is also a founding member of the Michigan Association of OWI Attorneys; a member of the American Chemical Society; an associate member of he American Academy of Forensic Science, Adjunct Professor of Forensic Evidence in Criminal Law and OWI Law and Practice at Cooley Law School. He is also author of the West OWI Practice book and several chapters in other books on science and the law.

Mike Nichols is recognized by his peers in Michigan as a “SuperLawyer” in DUI/Criminal Defense. Nichols has also been asked to speak at conferences by groups such as the NCDD; Various Bar Associations in other states.