Method
How accurate are life expectancy calculators?
A calculator can report a population figure exactly and still say little about one person's lifespan. What the number measures, what it leaves out, and what different kinds of calculator can honestly claim.
By The SnapLongevity Team7 min read
Asking whether a life expectancy calculator is accurate mixes two different questions. One is whether the number it reports is a correct summary of a population's death rates. The other is whether that number tells you when you will die. A well-built calculator can score highly on the first and still be uncertain on the second, because the second is not what the statistic was designed to answer.
What a population life table actually measures
A life table takes the death rates observed at each age in one year and asks how long, on average, a group would live if those rates never changed. WHO's 2021 table for Norway gives 82.9 years at birth. That number is calculated from WHO's estimated 2021 mortality rates, not a forecast of anyone's lifespan; the full step-by-step method is in our guide. How a life table is built.
Period life expectancy
Uses the death rates of a single year, applied across a whole hypothetical life. This is what WHO publishes and what SnapLongevity shows.
Cohort life expectancy
Follows people actually born in one year, using the death rates they meet as they age. It can only be completed once a generation has died, or projected with assumptions about the future.
Why the right number changes as you age
Life expectancy at birth includes everyone who dies in childhood or middle age. Someone who has reached 65 has, by definition, survived those years, so the relevant figure is remaining life expectancy at 65: 20.5 years in Norway in 2021, or an expected age of 85.5, not 82.9. At 85 it is 6.0 further years. This is conditional survival, and a calculator that ignores your current age is answering the wrong question. Life expectancy at 65 in detail.
What does a calculator's number describe?
From a population average to one person's uncertain lifespan
Step 1 · Population statistic
82.9 years
Life expectancy at birth in Norway. It averages over everyone born, including those who die young.
Step 2 · Conditional estimate
20.5 more years
Remaining life expectancy for people who have already reached 65. Still an average, now for a narrower group.
Step 3 · Individual outcome
A wide range
One person's actual lifespan. The same life table shows how widely 65-year-olds' outcomes spread.
Of 100 people alive at 65 under Norway's 2021 death rates
- 12 die between 65 and 75
- 29 die between 75 and 85
- 59 alive at 85
Norway 2021: life expectancy at birth 82.9 years; remaining life expectancy at 65 20.5 years. Of people alive at 65, die between 65 and 75 12%, die between 75 and 85 29%, alive at 85 59%.
WHO life tables, 2021, Norway, both sexes. Period figures: they assume 2021 death rates for the rest of life. Shares are from the survivor column: (l65 − l75) ÷ l65, (l75 − l85) ÷ l65 and l85 ÷ l65. The table ends at 85+, so outcomes beyond 85 are not split further.
What uncertainty means here
The same Norwegian life table that gives 65-year-olds an average of 20.5 more years also implies that, of people alive at their 65th birthday, 12% would die before their 75th birthday, 29% between their 75th and 85th birthdays, and 59% would still be alive at 85. The three groups do not overlap and together cover everyone. Being alive at 85 is not the same as dying at 85: that last group includes people who go on to die at 86 and people who reach 100. Likewise, 85.5 is an average age at death for the group, not a predicted age for any one person. All of these people share the same age, country and average. An average with that spread is not inaccurate; it is simply not a date.
There is a second layer of uncertainty in the inputs themselves. WHO's estimates are modelled from national death registration, surveys and statistical methods, and they are more precise where registration is complete than where it is not. WHO notes that its figures may differ from nationally endorsed statistics. Calculators built on any source inherit its uncertainty.
How sex, age and country enter the estimate
Population tools usually condition on exactly three things, because those are what official life tables publish. In Norway in 2021, remaining life expectancy at 65 was 21.3 years for women and 19.6 for men. Country differences at birth can exceed 30 years, but they narrow sharply at older ages because much of the gap comes from deaths earlier in life. Why countries differ.
What a simple population calculator cannot capture
- Your own health: existing conditions, blood pressure, fitness and medication are not in a population life table.
- Behaviour and circumstances: smoking, income, education and occupation shift risk, but the table averages over all of them.
- Future change: medical progress, new epidemics or policy shifts will alter death rates in ways a period table does not anticipate.
- Within-country variation: regions and social groups differ, and a national figure blends them.
What different kinds of calculator can reasonably claim
Swipe the table sideways to see all columns.
| Type | Inputs | What it can reasonably claim |
|---|---|---|
| Population life table | Age, sex, country | The average remaining years and survival chances for that population under one period's death rates. |
| Actuarial or official projection | Age, sex, country, assumed future trends | A cohort estimate that depends openly on assumptions about future mortality improvement. |
| Questionnaire-based risk tool | Population data plus lifestyle or health questions | A rough adjustment, credible only if its risk model and validation are published. |
| Clinical prognostic model | Measured clinical data for a defined patient group | Risk of death over a set period, for the population it was validated in. |
Adding questions does not by itself make a calculator scientifically validated. Validation means the tool's predictions were checked against what actually happened in people it was not built on.
SnapLongevity's calculator is a population tool. It shows WHO's figures for your age band, sex and country, together with survival chances to later ages, so the spread is visible rather than hidden behind one number. Explore the Norwegian example at 65.