How we compute timing

Mingpath's timing windows come from a published, deterministic scoring rule: each year is compared with your natal chart and luck pillar, matching stars add points and each adverse sign subtracts one, and consecutive years scoring 3 or more form a window. Same birth data and question, same result; the AI only writes it up.

The timing windows in a question book are not guessed by an AI. They come from a published scoring rule: the same birth data and the same question give the same result, whoever runs it and whenever. The AI only puts the computed result into words.

1. Evidence, year by year

For each year asked about (usually ten), the year's stem and branch are compared with your natal chart and the luck pillar of that time. What is compared depends on the question: partner star and spouse palace for relationships; authority, resource, output stars and the month pillar for career; wealth stars and the wealth store for money; favourable elements and the day branch for wellbeing. Each finding is one evidence line, labelled BaZi or Zi Wei.

2. What adds

3. What subtracts

Each adverse finding counts as one mark, −1, however dramatic it sounds. On purpose: adverse signs are what tradition most often overstates.

4. Net score and windows

Net = sum of adds − number of marks. Consecutive years with net ≥ 3 form a window; the one with the highest peak is "strongest", others "second". Years with adds ≥ 1 but two or more marks are mixed. If this year is in neither, it is marked watch. Scattered years with net 1–2 are listed as residual, not windows.

5. Worked example

Male, born 25 Jan 1999 23:26 in Shanghai (see conventions), asking when he will meet the right person, 2026–2035:

YearPillarAddsMinusNetReads as
2026丙午+0−1-1mixed
2027丁未+3−21mixed
2028戊申+3−12residual
2029己酉+202residual
2030庚戌+404strongest
2031辛亥+404strongest
2032壬子+101residual
2033癸丑+303second
2034甲寅+000
2035乙卯+1−10

Every evidence line for every year is printed in the book; every statement must point back to them, and a checker compares them line by line before the book is delivered.

6. Monthly flow

The window year (or the year asked about) is cut into twelve solar months and each month is scored the same way (BaZi only): good (net ≥ 2), watch (peer stars), moving (the month's branch clashes the chart), even. Companion members get the table refreshed every month.

7. Why transparent

The commonest problem with fortune-telling is that the story bends to the result. With fixed rules and the evidence laid out, you can judge for yourself which line you don't accept. The preview wall runs the same rules in your browser, so the preview and the book always agree.

8. Methods diverge

Timing is where schools disagree most: luck-pillar-first, clash-and-combine, symbolic stars, the "blind school"; Zi Wei has its own schools. We chose a rule that can be published and recomputed; that does not make it the only right one. If a teacher gave you different years, that is normal, and both can inform you. When BaZi and Zi Wei disagree about a year, the book says so instead of forcing one answer.

9. Date selection

The Xieji Bianfang Shu system: twelve officers, Yellow/Black Path spirits, 28 mansions, daily suitable/unsuitable lists, solar terms. A Yellow-Path day whose suitable list includes agreements, trade, contracts, travel or opening is a good day to begin; breaking and closing days and solar-term days are left open.

What we don't do

No prediction of specific events, no claims of accuracy, no promises about health, wealth or relationships; no fear or urgency; no per-minute chat, no virtual coins. Birth data is used only for what you ask for, never profiled, and you can delete it yourself in My account.

Cite this page: Mingpath, “How we compute timing”, https://mingpath.app/en/method/, updated 2026-11-08.