Rooting: Whether a Stem Is Standing on Anything at All
A stem on the surface of a chart is a claim. Whether it is backed by anything depends on what sits below it in the branches — and the table is small.
In a BaZi chart the four stems sit in a row along the top and the four branches sit below them. It is tempting to read the stems as the substance of the chart, since they are the visible characters and the day master is one of them.
The tradition reads it the other way round. The branches are the ground, and a stem only has real presence to the extent that something in the branches supports it. A stem with no root is described as floating — visible on the surface, backed by nothing underneath.
A stem with no root is a promise
Rooting, 通根, is the test of whether a stem's element appears among the hidden stems of any branch in the chart. If it does, the stem is rooted; if it does not, the stem is unsupported.
This is why two charts that look similar at the stem level can read very differently. A helpful stem sitting above a branch of its own element is a supply. The same stem above an unrelated branch is a statement the chart cannot cash — the classical image is a tree with nothing under it, which looks fine until anything pushes.
The concept earns its place because it feeds directly into strength assessment. Counting supportive stems without checking whether they are rooted is how a chart gets read as strong when the support is only on the surface.
Branches hold more than one thing
Rooting would be simple if each branch carried a single element. It does not. Every branch contains one, two or three hidden stems — the main qi, sometimes a middle qi, sometimes a leftover qi — and those hidden stems are what a stem can root in.
A concrete example: the branch 丑 holds 己 as its main stem, with 癸 and 辛 in the middle and leftover positions. So a Water day master can root in 丑 through the 癸 hidden there, and a Metal day master can root in the same branch through the 辛. Same branch, different answers depending on what you are testing for.
This is also why the branch pages on this site show hidden stems rather than a single element label. A branch described as 'Earth' is still carrying Wood and Water inside it, and a reading that ignores that will miss roots that are plainly there.
The three depths
| Element | Strong root (本气) | Medium (中气) | Weak (余气) |
|---|---|---|---|
| 木 甲 乙 | 寅 卯 | 辰 亥 | 未 |
| 火 丙 丁 | 巳 午 | 寅 未 | 戌 |
| 土 戊 己 | 丑 辰 未 戌 | 巳 午 | 寅 申 |
| 金 庚 辛 | 申 酉 | 戌 | 丑 巳 |
| 水 壬 癸 | 子 亥 | 丑 申 | 辰 |
Not all roots are equal, and the depth is decided by the hidden stem's position. A branch whose main stem is your element gives a strong root. The middle position gives a medium one. The leftover position gives a weak one — the element is present but in the weakest slot the branch has.
The table makes the pattern visible. Wood roots deeply in 寅 and 卯, the two branches whose main stem is Wood, and only shallowly in 未. Earth is the outlier, holding four deep roots because it is the element of the transitional branches. Water roots deeply in 子 and 亥 and shallowly in 辰.
Depth matters when a chart is close. A day master with one deep root is in a different position from one with two shallow roots, even though a count of supporting branches would score them the same.
Rooting ignores polarity
Here is the part that surprises people. The table has five rows, not ten. 甲 and 乙 root in exactly the same branches; so do 丙 and 丁, 戊 and 己, 庚 and 辛, 壬 and 癸.
The reason is that rooting is tested by element, not by polarity. A branch either contains Wood or it does not — it does not contain Yang Wood specifically. So the yang and yin stems of the same element share every root they have.
This is worth holding on to because it separates two things that are easy to blur. Polarity absolutely matters in BaZi — it is what produces the ten gods, and it changes what a relationship means. But it does not change where a stem can stand. That is a property of the element, which is why the reference table is half the length you might expect.
How to check a chart
Take the stem you want to test, note its element, then walk the four branches and ask of each one whether that element appears in its hidden stems. Record how deep the position is rather than just whether it is present, because the depth is the part that carries information.
Do it for the day master first, and then for any stem whose role in the chart you are trying to assess. In practice the interesting cases are usually stems that look important on the surface and turn out to have nothing underneath — a wealth stem with no root, or a supportive stem that evaporates on inspection.
Where the tables disagree
Hidden stem tables are not perfectly uniform across lineages. The disagreements are small and specific: whether a particular branch carries two hidden stems or three, and which stem occupies the leftover position when a branch is contested. Most branches are agreed on by everyone.
The practical effect is limited, because the disputed slots are almost always middle or leftover positions — the shallow end, where a root is weak anyway. A disagreement there can shift a close chart without changing an obvious one.
The version used here is the common one, and it is the same table the branch pages on this site display, so a chart you cast here will match what you read here. If you compare against another source and find a difference, check that slot before assuming something has been calculated wrongly.