What a night of drinking might do to blood alcohol over time — and how little anyone can say about it.
estimate only no accuracy claimed
RegionGlobal. Drink-drive limits are listed for a range of jurisdictions as information about the law only. They change, they are lower or zero for new and professional drivers in most places, and being below one is not the same as being fit to drive.
Read this before you touch anything below
Do not drive after drinking. Not tonight, not tomorrow morning, not at any number this
page shows you. This tool cannot tell you whether you are fit to drive, and neither can
any calculator, app, watch or breathalyser you own. The only reliable way to be certain is
to not drink and drive at all.
This page makes no accuracy claim whatsoever. It is a teaching illustration built on
a population-average formula. Real people differ from it by margins large enough to make any
individual number meaningless — food, medication, illness, liver function, genetics, sleep,
hydration, how accurately you remember what you drank, and what the drink actually contained
all move the answer, and none of them are in here.
No level shown here is described as safe, and none is. Impairment of judgement,
reaction time and vision begins below every legal limit in the world, and legal limits are
not safety thresholds — they are the point at which a particular jurisdiction decided to
prosecute. The limits listed further down are shown as information about the law, not
as targets, and never as permission.
If you have been drinking: get a taxi, take public transport, walk, or stay where you are.
If you are worried about your drinking, speak to a doctor.
The inputs
metric
Alcohol distributes into body water, and the two published formulas differ because average body
composition does. Neither describes any individual, and a person whose body composition sits away
from the average their formula assumes will be badly served by it. Pick whichever is closer, and
treat the difference between them as part of the uncertainty rather than a fact about you.
This slider is arbitrary and you cannot know your own value. It spans the published
population range, from a light or infrequent drinker at the slow end to someone with a long
heavy-drinking history — whose liver clears alcohol faster, which is a symptom of harm and not an
advantage. Moving it changes the answer by more than doubling it. That is the point of it being
here: if one unknowable slider can do that, no number on this page is a fact about you.
Food slows how fast alcohol reaches the blood, which lowers and delays the peak. It does not reduce the total.
The drinks
3 drinks
Drag the blocks on the timeline to move a drink in time. Times are 24-hour; a drink that starts
before midnight and finishes after it is fine.
What the model shows
—
—
This is not a measurement of you. It is what one population-average
equation produces from the numbers typed on the left. Do not use it to decide anything, and in
particular do not use it to decide to drive.
Detail
Drink-drive limits by place, for information only
These are legal thresholds, not safety thresholds.
They are prosecution limits chosen by legislatures, they change, they differ for new and
professional drivers, and several places listed are effectively zero. Nothing here is permission
to drive at any number.
Where
mg/100 mL
Also written
Note
Where these numbers come fromThe distribution factor is the Seidl (2000) height-and-mass regression; elimination is zero-order across the published population range of 0.010–0.025 g/100 mL per hour; absorption is first order. Drink-drive limits are the statutory prosecution limits published by each jurisdiction and are listed as information about the law only — they change, and they are lower or zero for new and professional drivers in most of them. No level shown is safe, and this page cannot assess anyone’s fitness to drive.
Field notes
Why no calculator can answer the question you want answered
How it works
One equation from 1932, and everything it leaves out
The curve is a Widmark model. Alcohol is treated as spreading through the water in the body, so the
concentration is the dose divided by that volume; the liver then removes it at a roughly constant
rate per hour regardless of how much is there. Drinks arrive through the stomach rather than
instantly, so this page passes each one through a first-order absorption step before it reaches
the blood. That is the entire physiology in it.
What the band is
The shaded region is not decoration and it is not a confidence interval in any statistical
sense. It is what the same drinks produce when the two most uncertain inputs are moved across
their published population ranges — the distribution factor by ±15% and
the elimination rate across 0.010 to 0.025 per hour. Individuals fall
outside it. Read the band, not the line.
If the band is wide enough to make a decision ambiguous, that is the honest output, and it is
the usual case rather than the exception.
Can this tell me if I am safe to drive?
No. Nothing on this page describes any level as safe, because none is. Impairment
begins below every legal limit that exists. This is a population-average illustration with
uncertainty far larger than the differences people try to use it to decide between, and it
knows nothing about you beyond the numbers you typed. Do not drive after drinking.
Can it tell me when I will be under the limit?
No, and it deliberately does not try. It will not print a time, because that number
would be wrong for a large share of people who read it, and because being under a legal
threshold is not the same as being fit to drive. The morning after a heavy night is the case
people misjudge most often — alcohol is still being cleared long after the drinking stopped,
and sleep does not speed that up.
Why does the elimination slider change the answer so much?
Because the published range across healthy adults really is that wide: roughly 0.010 to 0.025
grams per 100 mL per hour, a factor of two and a half. You cannot know your own figure
without a blood test series, so the honest reading of the slider is that it shows how little
the model can pin down, not that you should pick your favourite value.
Does coffee, a cold shower, food or exercise sober you up?
No. None of them changes how fast the liver clears alcohol. They can make somebody feel more
alert while remaining just as impaired, which is worse than feeling drunk. Food eaten
before drinking slows absorption and lowers the peak; it does not reduce the total.
Why does it ask for a male or female formula?
Because the published equations for how alcohol distributes were derived separately, from
groups with different average body composition — the relevant variable is body water, not
identity. Neither formula describes any particular person. If you sit away from the average
your formula assumes, it will be wrong about you, which is one more reason to read the band
rather than the line.
No accuracy is claimed and no responsibility is accepted. This is an educational
illustration of a 1932 population model, not a measurement, not medical advice, and not a
fitness-to-drive assessment. It cannot tell you whether you are impaired, whether you are over any
limit, or when you will be under one. Do not drink and drive.
Version history · unchanged
No change to this page at all since the earliest archived release (v4.64). The full history for the site is in the changelog.