Enter your weight and physiology setting to enable the educational estimate.

Not a sobriety test. This is a rough calculation, not measured BAC. It can be too low, especially soon after a drink. Never use this number or a consumer breathalyzer to decide whether to drive.

After finishing your last sip of beer, wine, or spirits, tap below. Do not start the timer when you begin drinking.

Start the timer after your last sip.

Not a sobriety countdown. This only reduces false-high readings from mouth alcohol. Your BAC can still rise; neither result tells you whether it is safe to drive.

Times between recorded drinks will appear here.

Accuracy review · selected day

Separate test conditions, early estimates and late clearance. Timing groups describe the records; they do not prove absorption is complete or that a device reading is accurate. No automatic calibration.

BAC comparison · 24 hours

━ Active estimate   ━ Old v2.3 model   ● Device reading   ▾ Meal
Hollow dots: less than 20 minutes after a completed drink.

The optional band varies recorded amounts only (±20% with pour details, ±30% without). Physiology and absorption assumptions stay fixed. It is not an overall BAC accuracy range. Device values can lie outside it.

Colored dates contain a drink record or symptom entry. A blank date does not mean zero drinks.

BAC estimate settings

Used only for a simplified calculation. Drink amounts matter: a half shot of 80-proof whiskey is 0.5 U.S. standard drinks, not 1.

Watson adult body-water equations add height and age (the female equation has no age term). Still a population estimate, not a measurement. Supported weight: about 66–441 lb; implausibly low body-water estimates are rejected. Current profile recalculates historical estimates; saved reading snapshots stay unchanged. Method and current conversion guidance

Account

Signed in

Firebase provides login and any recovery backup linked to your account. That backup is restored once per device, keeping newer local entries. New entries are saved locally; continuous cloud sync is not enabled. Local records are not encrypted.

Breathalyzer details

Defaults for new readings only. Existing readings keep their recorded device details.

App updates

Drink Log v2.4 · Simple login & logging · build 7 · estimator unchanged from build 5. Check for the latest deployed version without restarting your phone. Saved records and the breathalyzer timer stay on this device. Finish and save any open form first.

Export & restore

Drink, symptom and device-reading records stay in this phone or browser, not in Cloudflare. Back up before changing phones or clearing app storage. Updates on the same Cloudflare URL preserve local records.

About the estimate

One U.S. standard drink ≈ 14 g ethanol. Pours with volume/ABV use oz × 29.5735 × ABV/100 × 0.789 g ethanol. Unknown pours retain their recorded standard drinks × 14 g. Spirits are modeled as a shot consumed over 30 seconds before the completion timestamp. Other drink types retain a 15-minute consumption assumption. First-order absorption has a 15-minute time constant after ingestion; finishing a shot does not mean it is absorbed. These durations are assumptions, not measured start/absorption times. The 15-minute absorption constant was selected after retrospective device comparisons; this is not blind or clinical validation. Average distribution factors remain 0.68 male / 0.55 female. Optional Watson body-water estimation uses height, weight and age (male equation), with blood-water fractions 0.825 male / 0.838 female (UKIAFT v4.4). It does not personalize metabolism. Explicit per-drink duration replaces default consumption assumptions when recorded. Elimination now slows at low concentrations using Michaelis–Menten kinetics: rate = 0.015 × BAC / (0.008 + BAC) %BAC/hour, with one-minute integration. This avoids removing alcohol at a full constant rate while very little has been absorbed. Parameters are population assumptions, not fitted to your device. Kinetic model reference. The main number is the central estimate. The optional band varies dose ±20% for pour metadata or ±30% without, holding physiology fixed. These percentages are sensitivity assumptions, not measurement precision. It excludes uncertainty in distribution, metabolism, absorption, missed drinks and device error; it is not a confidence interval or an overall accuracy bound. The previous combined-extremes envelope has been removed rather than presented as a likely personal BAC range. Untimed legacy records are preserved and excluded. Current profile applies to all historical estimates. No automatic calibration.

A device reading is stored as a separate observation. It does not automatically calibrate the estimate or verify a device's accuracy.

Alcohol-related impairment can occur at any displayed BAC. Do not use this app to assess driving or machinery safety.