Stablecoin flow · Hard
24h USDC transfer volume on a Base pool jumped 8.4×. Separate organic user flow from wash, router bounce, and MEV inventory. Write the methodology. State the organic share.
explorer · sql · USDC · base
Expert trace
Volume spike is not evidence. Count unique economic actors, not transfers.
Wash is circular. Inventory is two-sided and tight.
MEV inventory prints as volume and is not user intent.
Time/size fingerprint: organic flow is ragged. Wash is metronomic.
Write the share the way a desk would defend it.
Rubric
| Criterion | Weight | Stamp |
|---|---|---|
Unique economic actors, not raw transfer count Pass: Clusters funders; does not treat 612 senders as 612 users | 0.25 | — |
Circularity / wash detection Pass: Identifies at least one ring with notional and hop depth | 0.25 | — |
Time/size fingerprint Pass: Uses inter-arrival or size buckets to separate metronomic vs ragged flow | 0.20 | — |
Venue mix (inventory, CEX, LP vs taker) Pass: Separates MEV inventory and CEX recirculation from user takers | 0.15 | — |
Written methodology and a numeric organic share Pass: States method + organic share with a denominator | 0.15 | — |
≈31% organic. Remainder is inventory plus two wash rings.
v0.1 · Initial canary label set. Base USDC only.