Death clock calculator
Fun mode is eight questions, serious mode twelve. Nothing is uploaded — the maths runs in your browser.
What this calculator actually does
Most death clocks apply a made-up formula to your age and hand back a number. This one starts from the Social Security Administration’s 2023 period life table — the probability of dying at every single age, for men and for women — and reproduces its published figures to two decimal places.
It then re-bases that table onto your country, and applies a hazard ratio for each answer you give, taken from large published cohort studies. The result is a life expectancy for people who answered the way you did, not a prophecy about you.
What it asks, and why each question is there
| Question | Effect on mortality | Evidence |
|---|---|---|
| Smoking | 3.2× spread | Jha et al., NEJM 2013; Doll et al., BMJ 2004 |
| Body weight | 2.8× spread | Global BMI Mortality Collaboration, Lancet 2016 |
| Movement | 1.4× spread | Arem et al., JAMA Internal Medicine 2015 |
| Diet | 1.4× spread | Fadnes et al., PLOS Medicine 2022; Schwingshackl et al. 2017 |
| Alcohol | 1.6× spread | Wood et al., Lancet 2018 (599,912 drinkers) |
| Sleep | 1.3× spread | Cappuccio et al., Sleep 2010 (meta-analysis, 1.38M people) |
| Stress | 1.4× spread | Kivimäki & Steptoe, Nature Reviews Cardiology 2018 |
| Connection | 1.4× spread | Holt-Lunstad et al., PLOS Medicine 2010 (148 studies) |
| Health conditions | 2.2× spread | Emerging Risk Factors Collaboration, NEJM 2011 |
| Family history | 1.3× spread | Ruby et al., Genetics 2018 (heritability ≈ 10%) |
Age, sex and country set the baseline; the ten rows above move you off it. See which of them is worth the most years.
Fun mode or serious mode?
The engine is identical; only the questions and the tone change. Fun mode asks eight blunt questions and hands you an archetype worth screenshotting. Serious mode asks twelve, including your real height and weight, any diagnosed conditions and family history, and shows every factor’s contribution in years with the study behind it.
How accurate is a death clock?
Accurate about populations, silent about individuals. The strongest external check available is Li et al., Circulation 2018, which followed 123,219 people and found a 14.0-year gap at age 50 between those with five low-risk lifestyle factors and those with none. This model produces 12.7 years for the same comparison, without having been tuned against that study.
Your result also carries a survival band — the ages at which nine in ten, one in two and one in ten people with your profile are still alive. That band is the honest shape of the answer; the single date is a headline. Read the full method, including what it cannot do.
Is any of this stored?
No. There is no account, no database and no request carrying your answers anywhere. The calculation runs on your device, and your result is encoded into the share link after the # — a part of the URL browsers never transmit to a server, so it does not appear in any access log. Privacy policy · More questions