About the AlcoCalc project

Why we created AlcoCalc

AlcoCalc aims to explain how much ethanol drinks contain and the assumptions behind a modelled elimination estimate. It shows the inputs, assumptions and timeline behind its calculation. It is not a breathalyzer or a professional assessment.

Inputs used in the calculation

The model uses serving volume (V, in ml), alcohol by volume (ABV, in %), number of servings, body weight (m, in kg), the selected distribution model, and the start and end times. Brand entries have a default ABV; use the product label when it differs from the catalogue.

Amount of pure ethanol

Ethanol (g) = number of servings × V × ABV / 100 × 0.789. The constant 0.789 g/ml is ethanol’s approximate density. For example, 500 ml at 5% ABV contains about 19.7 g of ethanol. This is the amount in the drink, not a blood alcohol estimate.

Distribution ratio r

The modelled dose is calculated as A / (m × r). The tool uses r = 0.68 for the male model and 0.55 for the female model. These values are also used in published eBAC models [1], but an individual’s body composition may differ from the averages. They are not personal measurements.

Elimination: a fixed assumption of 0.15‰ per hour

AlcoCalc subtracts 0.15‰ per hour from a shared modelled level. Jones’s review describes individual blood alcohol elimination rates of about 0.1–0.3 g/L/h [2]. A single rate cannot represent everyone. Grams per litre (g/L) and per mille (‰) are not generally interchangeable; AlcoCalc uses a simplified convention.

Absorption and timing

The model divides each drink into 60 equal portions across the drinking period and models each portion as absorbing over 45 minutes. A zero-length period is treated as one portion. These are implementation assumptions; source [1] uses a different absorption assumption and does not validate AlcoCalc’s 45-minute setting.

Uncertainty and differences from real-world results

The paper “Alcohol calculations and their uncertainty” [3] explains the uncertainty in calculations based on the Widmark approach. Serving size, alcohol strength, timing and the distribution factor can all affect the result. AlcoCalc does not directly account for food, medication, health conditions or individual metabolism, and it does not provide a guaranteed margin of error or confidence interval.

An estimate, not a measurement

All values and times are estimates for information only. They may differ substantially from a person’s actual blood alcohol level or a measured result. The tool provides no medical or legal advice, has no evidentiary value and cannot confirm that someone is fit to drive.

A reading of 0.00‰ does not confirm sobriety

Zero, the green colour and the smiley show the model’s result. The per mille value is rounded to two decimal places. Do not use the calculator, countdown or feeling sober to decide whether you can drive.

Transparency and expert review

The project does not publish fabricated reviews, expert authors or certifications. The cited literature explains the principles behind the calculator; it does not endorse or clinically validate AlcoCalc. Contact information is listed on the privacy page.

Expert resources

• [1] Prevalence of Hangover Resistance According to Two Methods for Calculating Estimated Blood Alcohol Concentration (eBAC), 2020 — Appendix A
• [2] A. W. Jones: Pharmacokinetics of Ethanol — Issues of Forensic Importance
• [3] Alcohol calculations and their uncertainty, 2015

The source publications support the stated principles, not the certification of this calculator.

Related topics

• Calculation methodology and assumptions
• The Widmark formula and estimated blood alcohol level
• Using AlcoCalc safely
• Sources and documentation
• Data and privacy