# Family Tree Statistics Are Most Useful Where They Look Wrong

The Statistics page is live in KleioBase, on every plan. It counts your names, places, lifespans, families and ancestors, and the record holders it picks out are often where the errors in a tree are hiding. Here is what it shows, how it counts, and where the numbers stop.

Claims about KleioBase last verified 2026-10-19.


Most genealogy programs will tell you how many people are in your tree. The more interesting figures sit underneath that one: who in it married youngest, which first name was handed from father to son for five generations, or how many of your sixteen great-great-grandparents you have named.

Those figures are fun to look at, and that is reason enough to want them. They are also a check on your data. A tree of a few hundred people is too big to read one profile at a time, and the figures that look odd are often where the mistakes are.

**The short answer.** KleioBase, an AI research companion that turns scattered historical records into one connected family archive, now has a Statistics page, on every plan including free Explorer. It turns the people and families in your current workspace into six tabs: Overview, Names, Places, Lifetimes, Families and Ancestry. They cover common names, maps of where people were born and died, record holders such as the youngest parent and the longest life, the average lifespan by century, ancestors found per generation, and couples in your tree who were related by blood. The page calculates every figure from what you have entered, at the moment you open it, so read the extremes as a list of things to check as well as a list of facts.

## A record holder is either a remarkable life or a typo

The Families and Lifetimes tabs pick out the extremes: most children, most marriages, youngest and oldest at marriage, youngest and oldest parent, earliest and latest born, longest life and shortest adult life.

**Some of them are real.** Some families were very large, and some people married three times. Seeing them named in one place is part of the pleasure of the page.

**Some of them are mistakes with a story attached.** A woman with four marriages may be two women of the same name who were merged into one profile. A father of 74 may be a son and father who shared a name. A mother of 14 may be a birth year copied from the wrong line, or a date read in the wrong calendar, which our guide to [reading dates in old records](/blog/reading-dates-old-records-julian-gregorian-feast-days) covers. Each record holder names the person, so one click takes you to the profile to find out which it is.

The page does not show values that cannot be right. A lifespan over 120 years, a parent younger than 12 or older than 80, and a marriage before age 12 count as data errors, and the page leaves them out of the figures so one mistyped year does not take over a card. The flip side is that the most obvious errors never reach the page, and you catch those on the profile itself.

## Your ancestors double every generation, and your tree does not

You have two parents, four grandparents, eight great-grandparents and sixteen great-great-grandparents. Each generation back doubles the number of possible ancestors.

The Ancestry tab shows, for each generation, how many of those possible ancestors your tree names. A fan chart lets you scroll back through the older generations. The figure tells you where your research thins out. A tree that names every great-grandparent but only five of sixteen in the next generation has a clear list of eleven people to look for. Our earlier post on [a family tree that shows what is missing](/blog/family-tree-that-shows-what-is-missing) made the same argument at the level of a single profile.

**Some ancestors appear more than once.** In small or rural communities, cousins sometimes married, and the same couple can turn up on two branches of one pedigree. Genealogists call this pedigree collapse. The Repeated ancestors figure counts how often it happens in your home person's pedigree.

## Shared ancestry puts a number on cousin marriages

When both members of a couple in your tree descend from the same ancestor, the Shared ancestry section lists them with their relationship, such as "second cousins", and a percentage.

The percentage is the share of ancestry a child of that couple would inherit from the same source twice, the measure the geneticist Sewall Wright called the coefficient of inbreeding in a paper in *The American Naturalist* in 1922. For first cousins it is 6.25 per cent. For second cousins it is 1.56 per cent. The section also lists people in your tree whose parents were related.

Beside each related couple sits an estimated DNA range, taken from the Shared cM Project, the crowd-sourced survey of known relationships compiled by Blaine Bettinger, in its fourth version from March 2020. For first cousins that range is 396 to 1,397 centimorgans, and for second cousins 41 to 592. DNA comparisons themselves happen at the testing companies, and our guide to [where to upload your DNA](/blog/gedmatch-alternatives-where-to-upload-dna) covers where you can do it.

## What we built

The Statistics page sits in the sidebar, after Insights. It reads only the workspace you are in, and it uses no AI and no credits.

**Six tabs.** Overview gives the size of the tree, gender and living or deceased split, occupations and religions, your Tree Health Score, and how many people have a birth date, a birth place, both parents, a photo or a linked record. Names ranks surnames and first names, shows names passed down from parent to child, and shows the earliest-born person with each surname. Places has a births map and a deaths map by country, the furthest moves, and how many people died in the country they were born in. Lifetimes adds lifespans by century, births by decade and month, and the most shared birthday.

**Lived through.** The Lifetimes tab also lists up to eight historical events that the most people in your tree were alive for, chosen by the same rules as the timelines in the Knowledge Base. An event counts for a person only if it applies to the places in their life and to who they were. For a person with no recorded death, the page assumes they lived to 100 at most.

**Everything clicks through.** Every person named links to their profile. Click a surname, first name, occupation, religion or place and the Knowledge Base People page opens filtered to those people, and that page now has multi-select filters for surname, first name, gender, occupation, religion and birth and death place.

**Everyone, or your connected tree.** If you have set a home person, a switch lets you leave out people who are not linked to them, such as a separate branch from an imported file. The address keeps the tab and the switch, so you can bookmark a view. The [Tree Statistics guide](/docs/tree-statistics) covers every section.

There are three honest limits.

**It counts what you have entered.** A merged profile, a wrong date or a missing family link changes the figures, and the page cannot tell you which ones are wrong. The page counts only dates with a readable year and leaves living people out of lifespans, and when several people tie for a record, the card shows one of them and how many others share it.

**The DNA range assumes one relationship.** DNA Painter, which publishes the Shared cM Project as an interactive tool, says that its statistics "do not cater for pedigree collapse or endogamy". A couple related in more than one way, the case the Shared ancestry section exists to find, is a case the range was not built for. Treat the figure as an estimate from published averages.

**Very large or heavily intermarried trees can be too much for some sections.** The page does its sums when you open it, and where a section would take too long, as Shared ancestry can on a tree with a great many cousin marriages, it says "Too large to compute here" instead.

## The practice, whatever you use

The page is ours, but the habit works with any program that can count.

**Start with the extremes.** Open the record holders before the averages, and check each one against its source. The oldest father and the most-married woman are worth five minutes each.

**Count each generation.** Write down how many of your sixteen great-great-grandparents you can name, then the thirty-two after them. The gap between the two numbers is your next research list.

**Treat a repeated ancestor as a finding.** If the same couple appears twice in your pedigree, that is something about the community your family lived in, and it changes how you read your DNA matches.

**Look at the surname that appears first.** The earliest person with each surname is usually where that line stops in your tree, and the document that names them is where the next search starts. Our guide to [FAN club research](/blog/breaking-genealogy-brick-walls-fan-club-research) covers what to do when that line will not go further back.

The people who first recorded these lives were parish clerks, registrars and census takers, each writing one entry about one household. None of them could see the patterns on this page, because those appear only once the entries for one family are gathered in one place, and gathering them is the work you have been doing in your tree.

Canonical: https://kleiobase.com/blog/family-tree-statistics-find-errors