Blood alcohol calculator
An estimate of blood alcohol concentration from Widmark's equation — given as the range the published constants actually produce, rather than as a number with three decimal places that looks like a measurement.
By Alex Seote, Built and maintains Rule Calculator
Constants checked against NIAAA and the forensic literature · How we check
This is an estimate and nothing more. It is not a breathalyser, not evidence, and not a basis for deciding to drive. Impairment starts below every legal limit, individual readings deviate from any formula in both directions, and only time lowers blood alcohol. If you have been drinking, do not drive.
Estimated blood alcohol, somewhere in
0.035–0.056%
A range rather than a figure, because Widmark's equation carries a personal constant whose published values disagree, and because people clear alcohol at 0.012 to 0.02 percentage points an hour. Your real reading can fall outside this band.
Time, which is the only thing that lowers it
| Peak before elimination | 0.077% |
| Under 0.08% no sooner than | already below it |
| Back to zero no sooner than | 4.7 more hours |
Both waits are computed from the top of the band at the slowest elimination rate people show, because that is the only direction it is safe to be wrong in. Coffee, food, a cold shower and sleep change none of it — the liver works at its own pace.
Why the single number is the wrong answer
Widmark's equation divides the alcohol consumed by the mass of body water available to dilute it. That water fraction — the constant r — is not measurable at a bar, and the two most-cited published values for it differ by 8% for men and 20% for women. The elimination rate varies by nearly a factor of two between individuals, and rises with habitual drinking. Multiply those uncertainties and the honest output is an interval.
Two further things the arithmetic cannot see: absorption, which food and carbonation shift by tens of minutes, and the fact that the peak arrives after the last drink rather than with it. Someone who stops drinking at closing time is often still rising an hour later.
Questions people actually ask
- Can I use this to decide whether to drive?
- No. Not because the arithmetic is poor, but because the arithmetic cannot be that good: the personal constant in Widmark’s equation differs between published sources by up to a fifth, absorption depends on food, and measured readings deviate from estimates in both directions. Impairment also begins well below any legal limit — reaction time and judgement degrade from the first drink. If the question has occurred to you, the answer is not to drive.
- What counts as one standard drink?
- In the US, 14 grams of pure alcohol — 0.6 fluid ounces. That is a 12 oz beer at 5%, a 5 oz glass of wine at 12%, or a 1.5 oz shot of 40% spirits. A pint of 7% craft beer is nearly two standard drinks and a generous pour of wine is often one and a half, which is the single most common reason people underestimate what they have had.
- Why does the calculator give a range?
- Because the inputs to the formula are ranges. Widmark's own distribution factor for men is 0.68 and the Forrest value is 0.73; for women the two are 0.55 and 0.66. Elimination runs between 0.012 and 0.02 percentage points an hour depending on the person, their liver and how regularly they drink. A calculator that reports one number has chosen one end of each range and hidden the choice.
- Does coffee, food or a shower sober me up?
- No. Food eaten before or during drinking slows absorption, so the peak is lower and later — that happens before the alcohol is in the blood. Once it is there, only the liver removes it, at a rate that nothing available to you changes. Coffee makes an impaired person a more awake impaired person, which is worse rather than better.
- Why is the wait time longer than other calculators say?
- Because it is computed from the top of the band at the slowest elimination rate people show. Every other assumption has a failure mode where someone drives too early. The published forensic work — including Bielefeld and colleagues on measured against estimated concentrations — is clear that individual results scatter around the estimate, so the estimate has to be built to be wrong in the safe direction.
- Is a legal limit a safety threshold?
- No. It is a line the law draws for enforcement, and impairment does not begin at it. Crash risk rises measurably below 0.08%, several US states enforce lower limits, and commercial drivers and drivers under 21 face far lower ones. Being under a limit is a statement about prosecution, not about being safe to drive.
Sources
- NIAAA — What is a standard drink?
- Bielefeld L, Auwärter V, Pollak S, Thierauf-Emberger A. Differences between the measured blood ethanol concentration and the estimated concentration by Widmark’s equation in elderly persons. Forensic Sci Int. 2015;247:23–7.
- Yang CT, Fung WK, Tam TW. Alcohol study on blood concentration estimation: reliability and applicability of Widmark formula on Chinese male population. Leg Med (Tokyo). 2009;11:163–7.