The gallery
Two kinds of entry. The classics are sequences you can check: each one is measured on our text, with the exact start, skip and count. The documents are published experiments, reported as records of what their authors did.
The classics
These are the observations that started the subject, from Rabbeinu Bachya's sequence at 42 to the five-book pattern told in the name of Rabbi Michael Dov Weissmandl. Our engine checks each one on the Leningrad text. Each card gives what was found, what chance alone would produce, and how shuffled texts compare. A retelling that does not reproduce is flagged. It is not quietly fixed.
Please read them for what they are. They are old and well loved, and they were noticed before anyone counted the alternatives. A count beside a chance figure tells you how ordinary each one is. Several are very ordinary. The count in Exodus is less so, and the pattern across the five book openings is less so again. Their entries give those ranks and explain how far they can be trusted.
בהרד at skip 42 in Genesis
Found 1. Chance alone 0.51.
Reproduces on this textתורה at skip 50 in Genesis
Found 3. Chance alone 2.2.
Reproduces on this textתורה at skip 50 in Exodus
Found 8. Chance alone 2.3.
Reproduces on this textThe Name at skip 8 in Leviticus
Found 7. Chance alone 4.4.
Reproduces on this textתורה backward at skip 50 in Numbers
Found 2. Chance alone 1.6.
Reproduces on this textתורה backward at skip 49 in Deuteronomy
Found 2. Chance alone 2.0.
Reproduces on this textמשנה and תורה in Exodus 11–12
Found 3. Chance alone 0.15. First letters 613 apart.
Reproduces on this textThe documents
These are the published experiments that matter most: the 1994 rabbis experiment, the War and Peace list, the communities test and its committee, and one cluster study, chosen because its authors set out their choices in the open. We report what was done and what came of it. We do not reproduce these claims as our own, and each entry says which text was used and whether we have re-run it.
Our gallery has no matrices about living people, elections, wars or current events, and nothing from the popular Bible code books. You get the record, the measurement and the means to check it for yourself.
The rabbis experiment, 1994
Rank 4 in 1,000,000 on Koren. Replications near chance.
Published experimentWar and Peace, 1999
Same test, changed names. Rank 57 in 100,000,000 in Tolstoy.
Published experimentThe communities test, 2004
Ranks 463,406 and 578,075 in 1,000,000.
Published experimentA cluster study, 2006
Choice factors listed by the authors: 3 × 6 × 64 × 2 × 8.
Published experimentHow to use this page
Each classic has an Open in the laboratory button that sets the term, scope and skip for you. From there you can tap any letter to open its verse, run shuffles, or change the skip and see what happens. Each entry also says what to look at next. If you are new, start with תורה in Genesis, then read how the chance figure is worked out.
If something here looks wrong to you, send us the edition and the exact letters. Nothing in this gallery is faked, and a figure that fails a check comes down or gets a flag.
What gets in
A classic enters this gallery if it has a traditional or published source, and if it can be stated as an exact term, start and skip. We then run it on our text. If it reproduces, it gets a page with its count and baseline. If a telling of it does not reproduce, the page says so and shows what does. We test claims as they are told, and we do not adjust them until they work.
A document enters if it was published, if it changed the argument, and if the reader needs it to judge the subject. We do not re-create a matrix from these papers on our own text, because that would imply a claim the paper did not make about this edition. We add entries slowly. A popular matrix does not qualify on popularity. Every entry gets the same treatment, whichever side it seems to help.
The classics at a glance
| Observation | Book | Skip | Found | Chance alone | Shuffles at or above |
|---|---|---|---|---|---|
| בהרד | Genesis | 42 | 1 | 0.51 | 393 of 1,000 |
| תורה | Genesis | 50 | 3 | 2.2 | 404 of 1,000 |
| תורה | Exodus | 50 | 8 | 2.3 | 6 of 1,000 |
| The Name | Leviticus | 8 | 7 | 4.4 | 138 of 1,000 |
| תורה | Numbers | −50 | 2 | 1.6 | 480 of 1,000 |
| תורה | Deuteronomy | −49 | 2 | 2.0 | 608 of 1,000 |
Read down the Found and Chance alone columns. In most rows the two sit close together, and that is the main finding of this page. The sequences are there. They are also about as common as their letters would make them. People remember these because of where they start, and a starting place is a thing the eye picks out afterward.
Each entry goes further than this table. It gives the verses, tests the claims that retellings make, and says what we can and cannot measure. The shuffles use seeds 1 to 1,000, so you can repeat every figure on this page yourself in the laboratory, with the same results.
Sources
- Rabbeinu Bachya ben Asher, commentary on the Torah, on Genesis 1:2.
- Rabbi Michael Dov Weissmandl, Torath Hemed, Yeshivath Mt. Kisco, 1958.
- Doron Witztum, Eliyahu Rips and Yoav Rosenberg, "Equidistant Letter Sequences in the Book of Genesis", Statistical Science 9(3), 1994, 429-438.
- Brendan McKay, Dror Bar-Natan, Maya Bar-Hillel and Gil Kalai, "Solving the Bible Code Puzzle", Statistical Science 14(2), 1999, 150-173.
- The text: the Westminster Leningrad Codex, as published by the Open Scriptures Hebrew Bible project.
Checked against its sources on 2026-10-31.