diff --git a/docs/HANDOFF.md b/docs/HANDOFF.md
index a5fe256..7c77f74 100644
--- a/docs/HANDOFF.md
+++ b/docs/HANDOFF.md
@@ -8,7 +8,7 @@ log.
 
 Canonical Iris `main` is **`ca2b4b2`** (#17, the headless rig). **#18
 `split/18-position-chain`** is open in `/home/bob/repos/iris-pr18`; its local
-head is **`45a7176`**, eighty-one commits, pushed. Built-in alignment is
+head is **`5f16617`**, eighty-two commits, pushed. Built-in alignment is
 complete there; see "Built-in alignment" below for the retained-layout
 details. No PR reviews were present when checked on 2026-09-15.
 
@@ -167,7 +167,7 @@ in, and it is compiled out of the release runs above.
 
 ## Verification
 
-**At `45a7176`**, the current head:
+**At `5f16617`**, the current head:
 
 - `cargo fmt --all --check`, `cargo clippy --workspace --all-targets --
   -D warnings`, `cargo test --workspace`: sixteen targets green, 83 suite
@@ -768,20 +768,8 @@ pixels and `Holds`. Decided with Bryan on 2026-09-15.
     **One half step further is arithmetically available and deliberately not
     taken.** The `Holds` assertion is quiet at one, and the whole-of-a-box
     case becomes an exact identity. What stops it is shrinker seed 220 on
-    `reorder`, which then lays out differently warm than cold: too narrow a
-    range is supposed to cost a redraw and nothing else, and there it
-    re-breaks a wrapping text, whose reported width moves a `Branch` onto
-    its other subtree. **That is the next thing to chase** -- it is the
-    unsettled-text family rather than a rounding question, it reduces to ten
-    widgets, and closing it is what lets this go lower:
-
-    ```sh
-    SHRINK_SEED=220 SHRINK_DEPTH=5 SHRINK_CASE=reorder \
-      cargo test --release --test shrink -- --ignored --nocapture
-    ```
-
-    The tree is `Scroll(X) { Pad as a region node { Span(X) { Branch,
-    Span(X){ wrapped text, one-line text } } } }`.
+    `reorder` -- diagnosed on 2026-09-16 and written up under "The offer
+    chain breaks at a region node" below. It is not a rounding question.
   - **The multiply on the way in** gets one more step at the top of the range
     and nothing at the bottom, since truncation only ever drops. The whole of
     a box has no multiply in it, however many pixels are added to it, and
@@ -800,6 +788,65 @@ pixels and `Holds`. Decided with Bryan on 2026-09-15.
   as floats and are put on the grid where they arrive. `Vec2` stays what the
   GPU and the platform speak; `PxVec2` is what layout decides in.
 
+## The offer chain breaks at a region node
+
+**The open defect at `5f16617`**, and the reason `Holds::through` keeps a
+half step it does not need. It is the fourth thing this session to come down
+to the offer against the placed box, after `aea878d`, `d8ae9c3` and the
+`Stack`/`Pad` overrides in `2bc6bdf`.
+
+`offered_region` composes a widget's offer through its ancestors' offers,
+which is the invariant. Where the parent is a **region node** it does not:
+it falls back to `UiRegion::FULL`, and `redraw` then resolves that against
+the node's slot entry -- which holds the node's **placed** box. So a widget
+under a region node is re-asked in the box its parent *placed* that node in,
+not the box its parent *offered* it.
+
+That is only wrong where the two differ, and `Scroll` is where they do: it
+offers its content the viewport and places it in a box as long as the
+content. So the offer becomes a function of the answer being re-derived, and
+the old answer confirms itself.
+
+Reproduced from the shrinker's ten-widget reduction:
+
+    Scroll(X) { Pad as a region node { Span(X) {
+        Branch(threshold 213), Span(X){ wrapped text, one-line text } } } }
+
+```sh
+SHRINK_SEED=220 SHRINK_DEPTH=5 SHRINK_CASE=reorder \
+  cargo test --release --test shrink -- --ignored --nocapture
+```
+
+Traced, at a 900x1200 window, with `reorder` rotating both spans:
+
+- Each order **does** settle, and to a different answer: `[0,1]` gives the
+  pad a 2080.18px box and `[1,0]` gives it 1789.90. So this is not an
+  unsettled tree; both are fixed points.
+- Cold, rotated: the texts are asked at 900 and 594.06, shape, and report
+  305.94 and 583.97, so their span reports 889.90.
+- Warm, rotated: the outer span is re-asked in **2080.18**, the pre-rotation
+  content box. Its texts are asked at 305.94 and 977.42, both reuse retained
+  answers without reshaping, and the span reports 1283.36 -- the same as
+  before the rotation, so no answer changed and nothing propagated up to the
+  `Scroll`. Widget 0 lands 290.27px out and the `Branch` picks its other
+  subtree.
+- Marking every widget dirty converges warm onto cold, which is what says
+  the retained path and not the arithmetic is at fault.
+
+**What makes the fix awkward** is coordinate frames, and it is worth knowing
+before starting. `ActiveData::offer` is in the parent *widget's* coordinates
+and `ActiveData::region` is in the parent *node's*, so "was this node placed
+where it was offered" is not a comparison either field can answer alone. Two
+shapes look plausible:
+
+- Compose offers through a region node by expressing the node's offer as a
+  part of its placed box -- an inverse composition, with a rounding of its
+  own.
+- Let `offered_region` say it cannot answer, and have `redraw` escalate to
+  the parent the way `parent_must_place` does. Correct and small, but it
+  costs region nodes their point for partial repaint unless the "placed
+  where offered" case is detected, which is the frame problem again.
+
 ## The leftover boundary
 
 Fixed in `5ed9e87`, which closed the shrinker's `reorder` case. Neither of its
@@ -1011,8 +1058,14 @@ Queued from this work, in order:
   allowance went from three half steps to two, and the whole of a box now
   maps back to one step rather than one step per level of nesting.
 
-  **And it was free.** 1,880M instructions and 755M cycles against 1,908M
-  and 760M, medians of 25 with all twenty-five work counters identical.
+  **And it made layout faster.** 1,840M instructions and 735M cycles against
+  1,908M and 760M -- 3.6% and 3.3% -- medians of 25 with all twenty-five
+  work counters identical. `5f16617` is where the win is: a draw already
+  descends past every move entry on its way in, so `DrawInfo` carries what
+  the slot composes to and a widget's own box is a select rather than a walk
+  back up a mean of 2.8 levels, eight hundred times a frame. Composing on
+  the fine grid alone cost 1,880M and 755M, so the widening was paid for
+  twice over by not doing the walk.
   Two things paid for the widening and are the reason to keep them: a
   *length* composes on its own in two multiplies a level rather than the
   four both ends cost, since where the parent sits falls out of the