← Back to blog
Zi WeiSingapore10 min read

Which Zi Wei Dou Shu star combinations never appear in any chart?

I went through every birth moment in sixty years, and some patterns did not turn up in a single chart.

Under this star-placement method, Huo Xing (火星, the Fire star) and Ling Xing (铃星, the Bell star) never share a palace (meaning the two stars land in the same palace). Under this method, Tian Kui (天魁, the Heavenly Leader) and Tian Yue (天钺, the Heavenly Halberd) never share a palace either. Lu Cun (禄存, the Wealth Keeper) and Tian Ma (天马, the Heavenly Horse) share a palace only for people born in six particular years. The other fifty-four years have zero charts.

By "pattern" here I only mean a few stars landing in the same palace, or flanking the Life Palace (命宫).

Curious about your own chart? Cast it free in 30 seconds →

Most articles about patterns talk about what a pattern does. How many people have it, and who cannot possibly have it, is something hardly anyone mentions.

The charting program is mine. I wrote the star-placement rules into code one line at a time, so this time I had it cast every one of the twelve double-hours of every day across sixty years, then checked the results against the three hundred and ninety-odd real charts in my charting database.

The Complete Book of Zi Wei Dou Shu (紫微斗数全书) has a line: "A gentleman's chart also holds the Ram, the Spinning Top, Fire and Bell; how could a petty man's chart lack the stars of Fame, Wealth and Power?"

The old book admitted long ago that everyone has malefic stars in their chart. What I wanted to count is what "everyone has" actually amounts to as a fraction. And which patterns nobody even gets the chance to have.

First, how I counted.

The exhaustive run goes from 1 January 1965 to 31 December 2024, twelve double-hours a day, 262,980 birth moments in total, one chart per moment. None of the stars counted here depend on sex, so I only cast male charts. Star placement follows the San He (三合) school, and Tian Kui and Tian Yue use the table built into my engine.

The real charts come from my own charting database. I took only the date and hour of birth, sex, calendar type, and longitude and time zone, and no other identifying information. I removed the form's default birth moment and counted each birth time only once. The statistics were run three times, a day apart each time, so the denominator is sometimes 391 and sometimes 392. I say which one I am using each time. Failures: 0 charts.

The script also carries a self-consistency check. For every chart, the earthly branch each star lands on must equal the branch the placement formula gives. Across the exhaustive set and the real charts: 0 violations.

The scripts are under xiaoS/census/, in four files. The results are under xiaoS/data/, with filenames beginning 20261024, 20261025, 20261026 and 20261028.

Let me start with the most ordinary cases.

Lu Cun is set by the year stem. Qing Yang (擎羊, the Ram) always sits one palace forward of Lu Cun, and Tuo Luo (陀罗, the Spinning Top) always sits one palace back.

Year stem Jia (甲) Yi (乙) Bing (丙) Ding (丁) Wu (戊) Ji (己) Geng (庚) Xin (辛) Ren (壬) Gui (癸)
Lu Cun's palace Yin (寅) Mao (卯) Si (巳) Wu (午) Si (巳) Wu (午) Shen (申) You (酉) Hai (亥) Zi (子)

So whichever palace Lu Cun sits in, that palace is flanked by Qing Yang and Tuo Luo. "Flanking the Life Palace" means two stars land in the palaces directly on either side of the Life Palace. Yang-Tuo flanking the Life Palace sounds ominous, and Lu Cun sitting in the Life Palace sounds auspicious, yet on the chart they are the same chart, identical case by case.

In the exhaustive set, Lu Cun, Qing Yang, Tuo Luo and Tian Ma each land in each of the twelve palaces in exactly 21,915 charts, one twelfth, not off by a single chart. Lu Cun in the Life Palace: 8.333%. Of the 391 real charts, 38 have Lu Cun in the Life Palace, against an exhaustive expectation of 32.6. Qing Yang in the Life Palace: 40 charts. Tuo Luo in the Life Palace: 41.

Huo Xing and Ling Xing meeting Qing Yang and Tuo Luo works the same way. Fire with Ram, Fire with Spinning Top, Bell with Ram, Bell with Spinning Top in the same palace: all four are 8.333%, and split by the ten year stems, each stem is still 8.333%. No year gets an edge. Among the real charts, Fire and Ram share a palace in 31 and Bell and Spinning Top in 28, with an expectation of 32.6 for both.

So far, everything is even.

One level up, the stars start picking hours.

Wen Chang (文昌, the Literary Prosperity star) counts backward from the Xu palace to the birth hour, and Wen Qu (文曲, the Literary Arts star) counts forward from the Chen palace to the birth hour. The two stars are symmetric about the Chou-Wei axis. They only coincide when the count lands on that axis. So only two kinds of people have Wen Chang and Wen Qu in the same palace: those born in the Mao hour (5 to 7 am), with both in the Wei palace, and those born in the You hour (5 to 7 pm), with both in the Chou palace. Exhaustive count: 43,830 charts, 16.667%, one sixth.

Di Kong (地空, the Earthly Void) and Di Jie (地劫, the Earthly Robbery) work the same way. Di Kong counts backward from Hai, Di Jie counts forward from Hai, and the axis is Si-Hai. People born in the Zi hour (11 pm to 1 am) have both in Hai; people born in the Wu hour (11 am to 1 pm) have both in Si. Also 43,830 charts, 16.667%.

Zuo Fu (左辅, the Left Assistant) and You Bi (右弼, the Right Assistant) share a palace for only two kinds of people: those born in the fourth lunar month, in the Wei palace, and those born in the tenth lunar month, in the Chou palace. Exhaustive: 16.815%. Real charts: 61, 15.56%.

Flanking the Life Palace is pinned down by the axis too. With two symmetric stars, the palace they flank can only lie on the axis of symmetry. For Chang-Qu flanking and Fu-Bi flanking, the Life Palace can only be in Chou or Wei, about 2.778% each in the exhaustive set, one at 2.768 and the other at 2.787. For Kong-Jie flanking, the Life Palace can only be in Si or Hai: 2.739% exhaustive, 10 real charts, expectation 10.7.

This is where the real charts drift. Chang-Qu in the same palace shows up in 34 real charts, 8.67%, against an expectation of 65.3, nearly half missing.

The missing half is entirely in the hour table. Only 14 people entered the Mao hour and only 20 entered the You hour, which adds up to exactly 34. Kong-Jie in the same palace drifts the other way: 70 charts, 17.86%, expectation 65.3. Zi hour 33 plus Wu hour 37 is exactly 70.

The hours real people entered pile up at the Chen hour with 52, the Shen hour with 60 and the Xu hour with 35, and run thin at the Chou hour with 17, Yin 21, Mao 14 and You 20. That reflects the people filling in the form, not the charts. Birth months, on the other hand, are flat: against a uniform distribution over twelve cells, chi-square is 5.71.

After the hours, the years take their turn.

Lu Cun and Tian Ma share a palace in only six years of the whole sixty-year cycle: Jia-Zi, Jia-Chen, Jia-Shen, Geng-Yin, Geng-Wu and Geng-Xu. Everyone born in those six years has it.

Tian Ma is set by the year branch: Shen, Zi and Chen years put it in Yin; Yin, Wu and Xu years in Shen; Si, You and Chou years in Hai; Hai, Mao and Wei years in Si. Tian Ma only ever lands in Yin, Shen, Si and Hai. The stems that put Lu Cun in those four palaces are only Jia, Geng, Bing and Wu, and Ren. Bing, Wu and Ren are yang stems, and cannot pair with the two yin branch groups, Si-You-Chou and Hai-Mao-Wei. That leaves Jia with Shen-Zi-Chen and Geng with Yin-Wu-Xu, exactly six years.

Exhaustive: 10.125%. Real charts: 46 of 391, 11.77%, expectation 39.6. Geng-Wu 26, Jia-Shen 11, Jia-Zi and Jia-Chen 4 each, Geng-Yin 1. The 345 charts outside these six years: zero.

Those 46 took a roundabout route. At first I had not deduplicated, and 175 of the 614 charts in the whole database had Lu Cun and Tian Ma together, 28.5%, well above the exhaustive figure. My first thought was that I had made a mistake in the lookup table.

Only when I dug through the database did I see that the form's default birth moment, in 1990, a Geng-Wu year, had been submitted unchanged 65 times. Geng-Wu happens to be one of the six Lu-Ma years. Sixty-five copies had magnified the shape of one chart. With the default and the duplicates removed, it came back to 46 of 391, right next to the exhaustive figure.

For Lu Ma Jiao Chi (禄马交驰, Wealth and Horse Galloping Together) I took the widest reading: Lu Cun and Tian Ma both within the Life Palace's San Fang Si Zheng (三方四正, the three directions and four squares), that is, the Life Palace, the opposite palace, and the palaces four steps away on either side, four palaces in all.

Even that wide, only the twenty-four years with the four stems Jia, Yi, Geng and Xin can have it. The thirty-six years with the six stems Bing, Ding, Wu, Ji, Ren and Gui: zero charts.

The reason is parity. In yang years Tian Ma is in Yin or Shen, in yin years in Si or Hai, so the parity of Tian Ma's branch follows the year branch. On the Lu Cun side, Jia-Yin, Yi-Mao, Geng-Shen and Xin-You share parity with the year branch, while Bing and Wu in Si, Ding and Ji in Wu, Ren in Hai and Gui in Zi have the opposite parity. Any two of the four palaces in San Fang Si Zheng are an even number of palaces apart. Two stars an odd number of palaces apart cannot both be in the same set. For people born in Bing, Ding, Wu, Ji, Ren and Gui years, Lu Cun and Tian Ma are always an odd number of palaces apart, and no matter how wide the definition, they never meet.

Table-based expectation 9.167%, exhaustive 9.178%. Real charts: 32 out of 401, expectation 36.8. This count was run a few days later, so the denominator has a few more charts. Lu Cun and Tian Ma exactly opposite each other also happens in only six years: 10.112% exhaustive, 40 real charts, expectation 40.5.

There is a whole string of other patterns that pick years.

Huo-Ling flanking the Life Palace only comes to Yin, Wu and Xu years. Of the four starting-palace groups, only in Yin-Wu-Xu years does Fire start in Chou and Bell in Mao, two palaces apart. That group is 8.335%, the other three groups zero, 2.101% overall. Real charts: 4 out of 392, all within the 110 charts from Yin-Wu-Xu years, expectation 9.2.

Ling Xing and Di Kong in the same palace happens only in Yin-Wu-Xu years, at the Chen hour (in the Wei palace) and the Xu hour (in the Chou palace). That group is 16.667%, the other three zero, 4.202% overall. Real charts: 25, also all within those 110, expectation 16.5.

Huo Xing and Di Kong in the same palace is the reverse: zero in Shen-Zi-Chen years, 16.667% in each of the other three groups, 12.486% overall. Real charts: 55, expectation 48.9, and not one among the 104 charts from Shen-Zi-Chen years.

Kui-Yue flanking the Life Palace only occurs with the stems Bing, Ding, Ren and Gui, 8.333% for each, zero for the other six stems, 3.336% overall.

Even the decade Four Transformations (四化) have blind spots. I checked ten year stems against twelve Life Palace branches, 120 cells. Under these rules, a decade Hua Ji (化忌, transformation into obstruction) landing on the birth-year Hua Lu (化禄, transformation into wealth) never happens for the three stems Bing, Ren and Gui; a decade Hua Lu landing on the birth-year Hua Ji never happens for the two stems Ji and Xin.

The real charts across the four year-branch groups: Yin-Wu-Xu 110, Shen-Zi-Chen 104, Si-You-Chou 98, Hai-Mao-Wei 80, chi-square 4.81, flat.

The last level.

Not a single chart.

Huo Xing and Ling Xing in the same palace: 262,980 exhaustive charts, zero. Huo Xing and Di Jie in the same palace: zero. Ling Xing and Di Jie in the same palace: zero. Real charts, 392: zero on all three. Tian Kui and Tian Yue in the same palace: zero exhaustive, zero real.

When I saw zero for Fire and Bell, my first suspect was the script. In an earlier count of the Body Master stars, I had left out the key for Ling Xing, and the twelve cells only summed to 91.6%, which is how that bug gave itself away. Had I dropped another one this time?

I added a check to the script: for every chart, the branches of Huo Xing and Ling Xing must equal the branch computed from the starting palace plus the birth hour. After the 262,980 exhaustive charts and the real charts ran through: 0 violations.

The zero is real.

Huo Xing and Ling Xing each get a starting palace from the year branch's San He group, then count forward by the hour. In Yin-Wu-Xu years Fire starts in Chou and Bell in Mao; in Shen-Zi-Chen years Fire starts in Yin and Bell in Xu; in Si-You-Chou years Fire starts in Mao and Bell in Xu; in Hai-Mao-Wei years Fire starts in You and Bell in Xu. In none of the four groups do they start in the same palace. The two stars move the same number of steps for the same hour, so however far apart they start, that is how far apart they land.

That gap can never shrink to zero.

Di Jie starts from Hai and counts forward, and none of the four Fire and Bell starting palaces is Hai, so Fire-Robbery and Bell-Robbery never meet either.

Tian Kui and Tian Yue come from a year-stem table: Jia, Wu and Geng in Chou and Wei; Yi and Ji in Zi and Shen; Bing and Ding in Hai and You; Xin in Wu and Yin; Ren and Gui in Mao and Si. None of the five pairs share a branch. Flanking the Life Palace requires exactly two palaces apart, and among the five pairs only Hai-You flanking Xu and Mao-Si flanking Chen qualify, corresponding to Bing-Ding and Ren-Gui. That is why Kui-Yue flanking appeared for only four year stems earlier.

The Complete Book of Zi Wei Dou Shu has another line: "The ominous is not wholly ominous; the auspicious is never purely auspicious."

The book says good and bad fortune are never pure. What I counted is another kind of impurity: the same chart is called Yang-Tuo flanking the Life Palace and also Lu Cun sitting in the Life Palace, and the auspicious and the ominous differ only by a name.

So when your own chart lacks some pattern, first look at which category it is in. If it is one that everyone has a one-in-twelve share of, not having it simply means not having it; only one person in twelve was ever going to get it. If it only comes with certain years or hours and yours do not match, you cannot have it, and that has nothing to do with whether your chart is good or bad. If no chart has it, nobody has it: the pattern in the book cannot be produced by this placement method.

That is about all these numbers can say, and there is more they cannot.

This time I only covered the sixty years from 1965 to 2024, exactly one full sexagenary cycle. Earlier and later years use the same rules and should give the same numbers, but I did not run them here, so I will not extrapolate.

Star placement used only the San He school and my engine's Tian Kui and Tian Yue table. Schools disagree on how to place Tian Kui, Tian Yue, Huo Xing and Ling Xing, and with a different table the list of "never" would change. I have not run the other schools' tables.

The real charts were entered by the people who came to have their charts cast, so they are not population statistics. The hours entered pile up at Chen, Shen and Xu. All I could do was remove defaults and duplicates; whether the entries are accurate, I cannot verify.

"Never" means never within this set of placement rules. Where a pattern named in the books cannot be produced under these rules, this article does not judge whether the book or the rules are right. It only reports the numbers.

Schools define Lu Ma Jiao Chi differently. I only counted three readings: same palace, opposite palaces, and both within the Life Palace's San Fang Si Zheng. I only used the birth-year Tian Ma, not the annual Tian Ma, and did not count any other reading.

All the shares are just shares. Whether you have a given pattern, and whether that is good or bad, this article does not say a word about.

Are Yang-Tuo flanking the Life Palace and Lu Cun sitting in the Life Palace the same thing? They are the same chart. Qing Yang is one palace forward of Lu Cun and Tuo Luo one palace back, so when Lu Cun sits in the Life Palace, Qing Yang and Tuo Luo are on either side of it, identical case by case.

Which years do I need to be born in to have Lu Cun and Tian Ma in the same palace? The six years Jia-Zi, Jia-Chen, Jia-Shen, Geng-Yin, Geng-Wu and Geng-Xu. Everyone born in those six years has it; the other fifty-four years have zero charts under this placement method.

Why don't I have Huo-Ling flanking the Life Palace? Start with your birth year's branch. Only in Yin-Wu-Xu years does Fire start in Chou and Bell in Mao, exactly two palaces apart; the other three groups have zero charts. Even among people born in Yin-Wu-Xu years, only 8.335% have it.

Can Tian Kui and Tian Yue share a palace? Not under this placement method. None of the five branch pairs coincide; the exhaustive set of 262,980 charts has zero, and so do the real charts. With another school's table, the conclusion might differ.

I went through every birth moment in sixty years, and some patterns did not turn up in a single chart.

Next time you notice a pattern missing from your chart, first work out whether it is one everyone has a one-in-twelve share of, one that only comes with certain years or hours, or one nobody has at all. Then think about whether it is good or bad.

⚠️ FateStar generates and interprets your chart based on the traditional Chinese discipline of Zi Wei Dou Shu (紫微斗数). All content is for informational and reflective purposes only.

About the Author

Louis
Louis

Founder of FateStar. He began structured study of San He Zi Wei Dou Shu in 2020 and is responsible for product method, source organization, and content review. Personal experience is not presented as scientific proof.

More about Louis →