Scientific & Legal Methodology

Transparency, accuracy, and privacy

CalcGrain is built as a teaching layer on top of published toxicology and licensing rules — not as a black-box score. Every core calculator runs in the browser from documented constants in our first-party script files. We commit to three things:

• Mathematical transparency — the same Widmark-style BAC path, biphasic EtG decay, and regional point tables you see here are the ones the tools actually call.
• Educational accuracy — inputs follow NIAAA-style standard drinks (14 g ethanol), SAMHSA-context EtG cutoffs that programs commonly print (100 / 300 / 500 ng/mL), and simplified DMV/DVLA/provincial schedules for the regions we list.
• Privacy by architecture — drink counts, body weight, and violation checkboxes are not posted to CalcGrain servers as part of a calculation. See Data privacy promise.

Editorial copy is reviewed against public NIAAA, SAMHSA, CDC, and licensing-authority material. Corrections with official source links are welcome via contact.

The math: BAC (Widmark-style)

Blood alcohol concentration on CalcGrain starts from US standard drinks, then a Widmark distribution volume, then a linear elimination term. A compact form of the teaching model is:

BAC ≈ (A × 0.806) / (BW × WR) − (β × T)

Where A is alcohol mass in grams, BW is body weight in kilograms, WR is the sex-based Widmark r factor, β is the elimination rate in % BAC per hour, and T is elapsed drinking or post-peak time in hours. The 0.806 factor appears in some published Widmark restatements that convert alcohol mass into a blood-concentration unit system; CalcGrain’s implementation is algebraically equivalent for teaching purposes:

peak BAC (%) = (grams / (kg × 1000 × r)) × 100 − (β × hoursDrinking / 2)

• Standard drink: 14 g ethanol (NIAAA definition: 12 oz 5% beer, 5 oz 12% wine, or 1.5 oz 40% spirits).
• Widmark r: 0.68 (male) and 0.55 (female) in js/etg-data.js.
• Elimination β: 0.015% BAC/hour default (Time to Sober also offers 0.012 / 0.015 / 0.018 for infrequent / moderate / heavier-habit teaching bands).
• During drinking: we subtract roughly half of β × duration so absorption and elimination overlap instead of pretending the entire session is an instantaneous peak.

After peak, remaining BAC is max(0, peak − β × hours). Legal-limit chips compare that estimate to common per se lines (0.08% many US drivers 21+, 0.04% many CDL contexts, 0.02% many under-21 rules, 0.05% caution). Impairment can begin below any of those numbers. Never drive after drinking because a calculator is green.

The math: EtG clearance

Ethyl glucuronide is a urine metabolite used in monitoring programs. CalcGrain does not treat EtG as a 31–40 hour single half-life. Urinary EtG literature typically discusses half-lives on the order of a few hours; the detection window after a session is much longer because the tail sits near sensitive cutoffs.

What the live EtG tools actually do:

1. Estimate peak BAC as above, then scale to a teaching peak EtG: peakETG ≈ peakBAC × 1000 × 62.5.
2. Decay with a biphasic curve: fast half-life ≈ 3.5 hours for the first 8 hours, then slow half-life ≈ 9 hours (see js/etg-common.js).
3. Compare ng/mL to 100 (sensitive), 300 (common clinical), and 500 (less-sensitive) cutoffs that appear on many US panels. SAMHSA has warned that low-level EtG cannot reliably separate beverage use from some incidental ethanol exposures — we surface that as education, not a loophole.
4. Optional hydration multipliers change a modeled concentration (dilute vs concentrated urine). Extra water does not add a new enzyme. Labs may flag low creatinine.

The dedicated Half-Life calculator is a different job: you already have a lab ng/mL and slide a 2.5–5 hour teaching half-life. That is not the same as inventing a peak from four beers. Detection-time tools solve for hours until each cutoff on the biphasic path. Population sketches for light-to-moderate intake often land around 24–45 hours under 300 ng/mL and 45–72 hours under 100 ng/mL — those are model bands, not a promise.

NIAAA’s ~0.015% BAC/hour figure is the ethanol elimination teaching rate used on BAC pages. It is not an EtG half-life. Mixing those two clocks is the most common error in “how long until I pass” searches.

Traffic point systems (DMV / DVLA / provinces)

Driving tools use simplified schedules aligned with published licensing guidance for the regions in js/data.js. They are not a live abstract.

• Totals: checked violations add published point values. Optional offence dates drop events that fall outside the region’s rolling period.
• License class: provisional, learner, and commercial modifiers can lower the ceiling (for example NSW P-plate limits).
• Auto-suspend flags: some offences (DUI-class rows) force a critical gauge even when the raw total is below the numeric cap.
• Expiry: endorsement visibility (UK 4 vs 11 years) is not always the same clock as “counts toward totting-up.” Insurers may look longer than the DMV suspension window.
• Insurance estimator: directional surcharge coefficients — not a binder quote.

Always verify with the statute, the citation, and an official driving record. We expand regions as schedules are checked; request a missing jurisdiction via contact.

Data privacy promise

Calculator architecture is client-side:

• No CalcGrain account is required to run a tool.
• Drink counts, weights, and violation selections are not written to our database as calculation logs. We do not keep a server-side history of your scenarios.
• Optional “save my profile” on some driving tools uses localStorage in your browser only. You can clear it with the forget control or by clearing site data.
• Contact-form messages you choose to send are the exception: that is voluntary correspondence, described in the Privacy Policy.

Analytics and ads, when you consent, may still see page URLs and coarse device data — not the numbers you typed into a calculator field. Full detail: Privacy Policy.

What we do not claim

• We do not certify sobriety, abstinence, or fitness to drive.
• We do not replace LC-MS/MS, evidential breath, or blood draws.
• We do not bind insurance premiums or guarantee a court outcome.
• What-If chips (wait 4 hours, food, water) are educational overlays. Food does not delete absorbed ethanol; water does not flush EtG from metabolism.

Related reading: About CalcGrain, EtG calculator, BAC legal limit calculator, points calculator, and the guides library.