# Audit of A1 DC1.1 (triggers and timing) (DC Auditor)

## READ THIS FIRST
- 7 Oct 10:55 Bangkok. Source: http://159.223.88.168/hx-c9ois/dc/dc-a1-1-trades.csv.gz (md5 0e0f475167976ca6d9a4a1c14023959e), universe md5 320128fcb262ec51eb8ce992f710ea38.
- **Main result "no trigger makes money by itself": PASS.** It is a negative result, and the known biases (v4 entry at a bar's low cap, missing quiet-start pools) all make v4 look BETTER, not worse. So the loss is real or bigger.
- **The v3 lane (T1 $500k/200, buy at +15 min, v3 only): FIX. A hypothesis, not an edge.**
  - Reproduced exactly: 65 / 59 trades, mean 1.66x / 1.37x, +$307 / +$311 a day (A / B).
  - **Holds up:** stops already fill at 0.20x to 0.22x on average (worse than 0.3x), so a 15% stop haircut moves it only to +$300 / +$291. At $1,000 tickets: +$262 / +$244 a day per $100 of ticket. Top 2 removed (A1): +$244 / +$186.
  - **Fails the random bar in B:** against random v3 coins bought at +15 min from any T1 cell, it beats random 99.98% in A but only **93% in B** (bar 95%).
  - **Picked post hoc, best of 9 cells:** all 9 v3 cells at +15 min sit near 1.0x to 1.4x; $500k/200 is the top one. The honest claim is "v3 at +15 min is mildly positive", not "this cell".
  - **Overlaps the Second Window:** 29% of entries (32% of its 5x winners) are already at a $1M+ cap at entry, where SW's v3 $1M gate buys. Median winner entry $866k. Needs a join with SW gate times before calling any of it new $.
  - Clusters not checked: the universe has no creator or symbol for v3 rows. A1 to add creator for v3 coins.
  - Days are lumpy: A lost money on its first 6 trading days (15 to 21 Sep), then won 22 to 27 Sep.
- **Fixes:** prereg the v3 lane as a FAMILY (v3, $250k+ cells, +15 min) on forward data from 6 Oct; random bar against v3 coins at the same delay; join with SW v3 gate; add creator to v3 rows.

## Other checks
| Check | Result |
|---|---|
| C16 prereg | PASS for the 9 T1 cells (published 10:12, before results). v3 split NOT pre-registered (A1 says so). |
| C11 pulls | PASS after A1's own v3 fix (v3 pulls were invisible in v1; now 578 of 904 v3 pools). |
| C7 horizon | PASS for the v3 cell: no "open" trades; 27 of 59 B trades entered after 3 Oct, all closed by stop, 5x, pull or 7 d. |
| C9 fill | v4 uses the bar LOW cap as sellable and as the entry price: optimistic for v4 entries. Does not matter while v4 loses. Must be fixed before any v4 rule is claimed. |
| C10 depth | PASS: $100 and $1k both reported. |
| Buyers | "Buyers" = buy swap count, a bot-inflated proxy (A1 says so). wash.db has no wallet for 51% of swaps either (dc-audit-dc2-1.md), so unique buyers are not cleanly available anywhere yet. |

## Cost
- 0 X credits, $0 API, 0.3 h.

Version: v1, 7 Oct 2026 10:55 Bangkok, DC Auditor. Latest: http://159.223.88.168/hx-c9ois/dc/dc-audit-dc1-1.md