SAPIR 861
Calculate
Open Scriptures Hebrew Bible · 54 portions · exhaustive, not sampled

Does a name meet its word?

A Hebrew name has a numeric value. So does every word in the Torah. In the week a person is born, one portion is read. How often does that portion actually contain a word of the same value? Not an opinion — a count.

EN · RU · עברית · ES · PT

Engine self-test

Running two probes against the live /api/birth…

Type a Hebrew name, or a number. Each square is one week of the reading year.

The number

51.8%

Averaged over every one of the 749 distinct proper nouns in the Pentateuch, this is the share of birth weeks whose portion contains a Torah word of the same value as the name. The computation covers all 54 weekly portions of the Torah, so there is no sampling error: the figure is exact under the stated rule.

It is not a high number, and it is not meant to be. A good part of the time the name and the week do not meet, and saying so is the point. And one thing matters more than the figure itself: it is a frequency, not evidence of design. We have not compared it against a random model, did not state a hypothesis in advance, and it has not been independently replicated. Until that comparison exists, the number says «this is how the text is built», not «this was intended».

By size of the name

GematriaNamesMeets its week
1-192267.2%
20-20028255.9%
201-50033552.2%
501-100010738.0%
over 100034.9%

The decline is arithmetic, not mystical: large values are simply rarer among the words of any single portion.

The counting rule

Every published figure depends on what is allowed to count as a word. Ours is the strict reading, and it lowers the result:

Consonants only
Vowel points and cantillation marks are stripped before counting.
Prefixes removed
Using the OSHB morphological segmentation, the form וְאוֹר counts as אור (207), never as 213. A conjunction glued to a word does not make a new word.
Content words only
Nouns, verbs and adjectives. Particles, prepositions and pronouns are excluded.
Standard values
Mispar hechrachi. Final letters keep their base value: ך=20, ם=40, ן=50, ף=80, ץ=90.

Relax the first two rules — count every word as written, prefixes included, any part of speech — and the same corpus yields 62.7%. Restrict that further to names between 20 and 200 and it yields 72.8%. Both are true statements about a looser question. We publish the strict one because a reader who checks should find what we said.

Two things worth knowing

Every name meets its word somewhere. Of the 749 biblical names tested, none has a value absent from the Torah entirely. A name is never left with nothing; only the week may not be the meeting place.

A meeting is rarely trivial. Only 10 of those 749 values occur in all 54 portions. When a name and its week do meet, the coincidence is usually specific to that week.

The dataset

For every value from 1 to 1500: how many of the 54 portions contain a word of that value under the rule above. Free to use with attribution.

Download the coverage table (CSV)

Cite as:

SAPIR 861 (2026). Verse-of-meeting coverage in the Pentateuch: an exhaustive measurement. Dataset and method. https://sapir861.com/method

Sources

Text and morphology: the Open Scriptures Hebrew Bible, Westminster Leningrad Codex, CC BY 4.0. Portion boundaries: the hebcal leyning data. Both are public; anyone can repeat this computation and get the same number. Note on a different system: the Latin table printed by Agrippa in 1533 is a Renaissance invention, not the Jewish tradition — see Agrippa 1533 vs Jewish gematria.

Your own name

The same calculation, run on your name and your birth date, with every step shown.

Find your birth code

SAPIR 861 · sapir861.com · we count, we do not guess
Dataset CC BY 4.0. Text and morphology © Open Scriptures, CC BY 4.0.

Share

WhatsAppTelegramFacebookX