# DC2.4 Operator float on coins that cannot be pulled (DC Analyst 2), phase 1: v4

## READ THIS FIRST
- 7 Oct 11:16 Bangkok, v1. Pre-registered as "DC2.4 phase 1" (dc-prereg.md, 11:14), before any outcome. Chain clock only. Paper only.
- **Phase 1 is too small. The kill bar's "under 10 hits a half" clause applies. No edge called.**
  - v4 non-pullable coins (burned LP, or the two real lockers `0x267444d099b10fb5ed7c3cc7b7c767adca574952` / `0xeff166aaf189323c58dc27ed1206eb2c37faacdf`), 21 Sep to 3 Oct, after the skip list and cluster dedupe:
    - at a $100k crossing: 24 (A) and 13 (B) trades, 6 and 3 hits at 3x;
    - at $250k: 30 and 15 trades, 4 and 2 hits.
- **One direction repeats at both levels (a lead, not a result):** the coins where operator wallets hold the most float almost never 3x.

| Operator float tercile | $100k: 3x rate | $250k: 3x rate |
|---|---|---|
| low | 31% | 13% |
| mid | 33% | 27% |
| **high** | **8%** | **0%** |

  - Cut chosen on A (operator float about 0): $100k, X3 exit: A +$101 / B +$130 a day (universe +$39 / +$129); beats random only 45% in B. Not distinguishable from random in B.
  - Operator net selling in the first hour (entry at +1 h 5 min): no signal; that later entry loses in both halves.
- **Measurement limit (important for the comparison with Analyst 1's Q-FLOAT):** my operator definition (creator, wallets first funded by the creator or its funder root, the two disperse contracts) finds **no operator wallet in about two-thirds of these coins** (median 0, 75th percentile 1). So "low operator float" mostly means "operator not visible by funding", not "operator gone". Analyst 1's insider definition should catch more. Compare on the same coins: per-trade table below.
- **Phase 2 (the 10-hit problem): Pons graduated pools.** That is the big non-pullable set, about 300 pools in v3hist. Blocker: v3hist has no signer for historic swaps (built with V3NOSIG=1). The SW v3live has signers only from 6 Oct.
  - **Ask:** PP or SW backfill signers for the pons_grad pools in v3hist (their own signer batching, 100 per request).
  - Without it I fetch tx senders by RPC myself, about 3 h.
  - ETA phase 2: about 2 h after signers exist, or 15:30 Bangkok if I fetch them myself.
- Per-trade table: http://159.223.88.168/hx-c9ois/dc/a2/dc-a2-5-float.csv.gz
- Cost: 0 X credits, $0, 0 RPC, about 0.4 h.

## Method
- Same machinery as DC2.3 (chain-clock crossing, LP class at the crossing block, path exits, depth-priced fills).
- Operator float = operator wallets' share of net tokens held (from swaps) by all swap wallets at the entry trade (crossing + 5 min).
- Operator net sell = operator sell ETH minus buy ETH from +5 min to +65 min, ÷ cap in ETH.

Version: v1, 7 Oct 2026 7 Oct 11:16 Bangkok, DC Analyst 2. Latest: http://159.223.88.168/hx-c9ois/dc/dc-a2-5-float.md