Jamie Allen's Fabulous Pedigree -- Statistics

I've not updated most of these statistics since the database was much smaller. Version 83 has 262,590 pedigree pages, but this page was prepared when the database was a fraction this size.

Statistics of Database Version 3

In addition to several explanation/discussion pages, the site contains 95 index pages, and ...

The following table was prepared while Version 4 was in development, when there were 38,908 pedigree pages.

The total number of surnames I send to the Gendex index is 15582, but about one-third of these (forms with prefixes like von, de, of the , etc., and forms with hyphens like Waldeck-Eisenberg) do not have separate index headers in the website index.

The 10201 surnames recognized for the website index are broken down as follows:

The various other statistics which follow reflect Version 1. I've been too lazy to update them.


Statistics of Database Version 1

Although the Version 2 database has 33111 pedigree pages (2165 more than the Version 1 database), I've left the old statistics for now. In many cases, the Version 2 statistic can be estimated by just increasing the old statistic by 7%. (In some cases the Version 3 statistic can be estimated by increasing the old statistic by 22%, but many ancestries have been demoted from ``probable'' to ``possible.'')

An exception is the lists of most common names which I've revised since they seem to have been miscalculated previously.

31,031 Nodes for 29,920 Individuals.

Counting some private and placeholding nodes that aren't uploaded to the website, my database has 31,031 nodes, reflecting 29,920 individuals.

Not only is 031,031 a doubled-string number, and 029,920 a palindrome, but there is a special reason to consider these numbers unusual.

First, note that ... , 5 , 7 , 11 , 13 , 17 , ... are consecutive primes.

Now get this:
        7*11*13 * 31 = 31031
        7+11+13 = 31
        5*11*17 * 32 = 29920
        5+11+17 = 33

I decided this was ominous, and made an easy decision for Final Freeze. (Well ... I changed my mind after a few weeks, and Version 2 contains over 2000 important additions!)

Size of the Database.

There are other ways to count the size of my database. There are 30,946 pedigree pages of which 29,808 are indexed at Gendex.com. (This last number excludes the additional people (or abstractions) include ``(Miss; Coheiress) de VAUX de Diriton'', ``son of Gerard of WETTERAU'', and ``First Speaker of Indo-Uralic'' none of whom has a first name that can be provided to Gendex Index users.)

The Gendex index also omits people like or: Martha Allen SHELBY who appear only without their own pedigree page as an ``or: '' alternate. (There is a surname entry for the alternate in my website's own Index (unfortunately the primary person's forename is used for the sub-sort).

Two factors that might make the ``number of individuals'' misleading are

The number of over-counted and uncounted people are roughly equal; thus 29,920 is a good estimate for the number of named people.

Some people have two pages; some zero.

On the pedigree page of Teutberge de TROYES, her children are shown to include possibly Constance of Arles, but there is a notation that Constance's mother may instead have been ``Teutberg, daughter of Sanchez I Mittara.'' This Teutberge does not have her own pedigree page; in fact if you click on her possible daughter, Constance of Arles, the alternate Teutberge's name is not even written in full -- only the Teutberge de TROYES page contains the full mention of ``Teutberg, daughter of Sanchez I Mittara.''

There are about a thousand instances in my database, where a possible parent does not have his/her own pedigree page.

On the page of Donald, 8th Lord of the Isles, you will see that two distinct fathers are shown for Donald: Eoin MacAngus or John MacDonald, but these are just two names for the same person, the same 7th Lord of the Isles, whose pedigree seems to be uncertain. (You don't even need an ``aka'' for the police blotter: John is simply the Anglicization of Eoin and most nobles had two or three surnames to choose from.) If you click on the two ``different'' fathers of Donald, the 8th Lord, you may conclude, like me, that ``(NN) MacDONNEL'' is likely to be Agnus MOR himself! If I were more sure, I'd shrink my database by five pages -- the alternate John MacDonald would disappear, along with ``his'' five ancestors, and, so that the single pedigree then reflects both sources, a generation would be inserted between the 6th and 7th Lords reading
    ``(NN) ... poss. one or two missing generations.''

Similarly it seems disputed whether Andrew or Archibald Livingstone was the father of Sir William Gorgyn, but when the respective pedigrees are compared, one guesses that the ancestors Thurston and Livingus are the same in the two trees, so each is duplicated and has two pedigree pages at the web site (none of these pages with much info.)

Because of such cases there are a thousand or more individuals in the database who are represented with two pedigree pages.

The thousand pages with two individuals, and the thousand individuals with two pages, roughly cancel each other out.


Fictitious and unnamed people.

One objection to stating that my data base has 29,920 individuals is that that includes fictitious people. There are almost one hundred immortals from Greek myth in the database, and hundreds of fictitious people from other myths (Irish, Brythonic, Germanic, Persian, etc.). (All these names are sent to Gendex.) I'm afraid there may be hundreds of other fictitious individuals in the database where the fiction is less obvious ... and the joke's on me.

The number of named fictitious persons is in the same ballpark as the number of unnamed real persons in the database, so 29,920 is a good estimate for either ``number of real people'' in the database, or ``number of named people.''


Children.

There are 6054 pedigrees showing 2 or more children (2.6 on average); those with most children are Henry I Beauclerc, and Ralph Nevill d. 1425, 1st Earl Westmoreland.

In addition there are 297 pages showing children who don't have pedigree pages; an average 2.8 such children are shown when any are; of these, the two with most such children are King Edward I of England and Nyx (Goddess of Night in Greek myth).


Database Size.

The birth years for 30,900 individuals can be broken down very roughly as follows:

It may be useful to break down the 2600 people in the database born between 4000 BC and 200 AD. They constitute four broad groups of roughly equal sizes:


Names.

The surnames most frequently encountered in this pedigree are national designations like ``of FRANCE; of the FRANKS; of BRITAIN'' -- some of the other biggies of this type are EGYPT/PHARAOH, BAVARIA, DENMARK, SAXONY, BURGUNDY, ENGLAND.

Among more normal surnames, STEWART/STUART, NASSAU, BRUNSWICK/WELF, GREY/GRAY, DOUGLAS, BEAUCHAMP, MONTFORT, NEVILLE, BRANDENBURG, MONTGOMERY, BASSET, BEAUMONT, COURTENAY, FERRERS, MECKLENBURG, are the top fifteen in order. (I'm not descended from German nobility, so the large number of Nassau's, Brunswick's, Brandenburg's, and Mecklenburg's is a result of pursuing Queen Elizabeth's pedigree.)

The top fifteen forenames seem to be, in order, John/Jean/Johann, William/Guillaume/Wilhelm, Elizabeth/Isabella, Margaret/Margery/Marguerite, Jane/Joan/Joanna/Jeanne, Robert/Rupert, Henry/Henri/Heinrich, Thomas, Mary/Maria/Marie, Matilda/Maud/Mahaut/Mathilde, Agnes, Richard, Alice/Alicia, Hugh, Roger.


Relationships.

Another way to summarize the pedigree pages is to say that in addition to 199 abstract pedigrees whose subject is not an individual, and the pedigrees of myself and HRH William Windsor (future King William V of England) who are used as the roots of these pedigrees, there are 30,745 other individual pedigrees, broken down as follows:

  Relation to HRH William 
Relation
to Jamie
 HRH William's
certain ancestor
HRH William's great-
uncle or -aunt
HRH William's uncertain
or distant relation
 Totals
My certain ancestor  8054 20 724  8798
My great-aunt or -uncle  779 970 13  1762
My poss. ancestor  3526 228 6658  10412
My poss. great-aunt or -uncle  283 108 127  518
My distant cousin  6067 101 3087  9255
 
Totals  18709 1427 10609  30745

Notes: (1) Half-aunts are distinguished from Full aunts in the pages, but the designation is not always reliable and would have complicated the above table. (2) Many of those shown as My great-aunts or -uncles are also my possible ancestors, but the certain aunt/uncle relationship takes priority.

As seen, HRH William and I share many distant ancestors, but we are not closely related. According to this database King James I (1566-1625) of the United Kingdom is 3rd cousin of Hew Kennedy (b. ca 1567) -- both are descended from Earls of Angus -- and this provides my closest relationship to the recent Kings. Even that assumes that the matriarch of Allen Dale Farm was indeed the great-niece of the Second Earl of Dundonald, which as far as I know is unproven. (In addition to being 3rd cousins, King James I and Hew Kennedy were 4th cousins using another line. Because of the ubiquity of such multiple relationships I stop at great-aunt and -uncle and do not depict cousin relationships in these pedigrees.)

Surnames for Gendex.

The Gendex.com website reports my database size as 54,170, but actually there are only 30,946 pedigree pages and many of these are placeholder or abstract nodes not sent to Gendex. Only 29,807 individuals are sent as `jamesdowallen' index entries to Gendex, but many are sent with two or more indexing surnames; these break down as follows:

(For my own purposes I count named individual in a slightly different way and get 29,920 individuals.)

Here are the two individuals that happened to get seven different surnames at Gendex. Maria de Loon-Heinsberg (1424 - 1502) is shown with

Sophia Charlotte von Brunswick-Luneburg (1668 - 1705) is shown with

It's somewhat arbitrary which spelling variations are used for which person -- many Von Brunswick's also show up under BRAUNSCHWEIG. (For my own index here, most spelling variations are unified in the Index, e.g. with ``see'' tags.)

Duplicated individuals.

Because ancient names and surnames were not standardized, it is common for the same person to appear with very different names. Two ancient pedigrees, one French and one German, may lead back to the same person but because my database is large, and because the names may be very different, I often don't notice and the same person ends up in the database twice.

For example in December 2002 I noticed that Eudes of LOGENAHE, shown as agnatic ancestor of Emperor Conrad II in some genealogies, and Udo II of LAHNGAU, ancestor of the Wetterau Dukes of Swabia, are actually the same person. Now he has a single pedigree page, but in the earlier version he had two. `Eudes' and `Udo' are forms of the same name and, I guess, the names `Logenahe' and `Lahngau' are also cognate.

Perhaps you, the Pedigree Surfer, can find other individuals with two separate pages here. I'd offer a cash reward, but I'm afraid there might be too many.

Other ways to break down Database Population.

The 29,920 individuals can be separated as follows:

Some of the pedigree pages mention alternative parents for some children; these include

  • 300 links to some of the duplicate individuals mentioned above.
  • 400 links to a child, sibling, etc.
  • 600 links to other pedigreed individuals.
  • 120 links to database individuals, but without an HTML link, e.g.
    ``or: poss. Maud de BETHUNE. q.v.''
  • 600 mentions of named non-database individuals.

    Recent Database Growth.

    I started this project several years ago with just a few dozen names, and it has grown steadily. I've been installing new versions every few months; for details check at archive.org. Here's a summary of recent growth:


              Start.           Pedigree.           Index.           Sources.           Explanation.           Resume.
    Please take the time to E-mail your comments and corrections to me.