LooksRare / X2Y2 wash-trading reward-farming cohort — Ethereum — 2022

Summary

LooksRare launched on January 10, 2022 with a "vampire attack" on OpenSea: users who had traded on OpenSea received LOOKS token airdrops, and ongoing trading activity on LooksRare earned LOOKS rewards. The reward pool was substantial — the LOOKS token emission schedule allocated a significant share to trading rewards.

The incentive design created a structural wash-trading surface:

1. Wash-trade profitability = reward value minus marketplace fee cost. If the LOOKS rewards earned per ETH of trading volume exceeded the 2% marketplace fee (plus gas), a trader could profit by washing NFTs between controlled addresses. The marketplace collected fees; the trader collected rewards; the NFT transfer was economically meaningless but on-chain indistinguishable from a legitimate trade.

2. The wash-trading volume was vast. By February 2022, CryptoSlam and Dune Analytics dashboards estimated that 80-95% of LooksRare volume was wash-traded. The absolute volume was so high that LooksRare briefly surpassed OpenSea in nominal trading volume, even though the genuine economic activity was an order of magnitude smaller.

3. X2Y2 launched a structurally identical incentive model. X2Y2 (February 2022) replicated the trading-reward design, producing a second wash-trading wave. The wash-traders migrated between platforms based on which reward rate exceeded the fee+gas cost.

4. The extracted rewards entered circulation as clean tokens. LOOKS and X2Y2 tokens earned through wash trading had no on-chain traceability to the synthetic nature of the volume that generated them. The extraction path was: purchase floor-price NFT → wash-trade between controlled addresses → claim LOOKS/X2Y2 rewards → sell LOOKS/X2Y2 for ETH/USDC. The NFT leg is the laundering mechanism; the reward token is the clean output.

The wash-trading activity was not invisible — Dune dashboards by hildobby, sealaunch, and others tracked the wash-trade percentage in near-real-time, and CryptoSlam adjusted its volume rankings to flag wash-traded marketplaces. The failure was not in detection but in incentive design: the marketplace contracts rewarded volume without any volume-authenticity check, and the wash-traders were acting rationally within the incentive structure as designed.

The wash-trading volume declined as LOOKS and X2Y2 token prices declined (the reward-per-volume equation turned negative), and both marketplaces eventually adjusted their reward mechanisms. The structural lesson — that volume-based token rewards without wash-trade detection are a laundering surface — recurred at Blur (October 2022 onwards) with a different reward mechanism that was marginally more resistant to simple self-trade cycles but still wash-trade-able at scale.

Timeline

When	Event	OAK ref
2022-01-10	LooksRare launches with LOOKS trading rewards	T7.004 surface deployment
2022-01 to 2022-03	Wash-trading volume peaks; 80-95% of LooksRare volume estimated as synthetic	T7.004 execution
2022-02	X2Y2 launches with structurally identical reward model; second wash-trading wave	T7.004 extension
2022-Q2	LOOKS and X2Y2 token prices decline; wash-trade profitability drops; volume normalises	(reward-economics closure)
2022-10	Blur launches with Points-based reward model; wash-trading adapts to the new incentive structure	(next-cycle surface)

What defenders observed

• Volume-based token rewards without wash-trade detection are a T7.004 surface by design, not by accident. The marketplace contracts correctly executed every transfer and correctly distributed rewards. The failure was at the incentive specification layer — rewarding "volume" rather than "legitimate price discovery" — and the wash-traders were rational economic actors within that specification.
• Community dashboards (Dune, CryptoSlam) were the detection layer. The wash-trade percentage was tracked in near-real-time by independent analysts using self-trade and cycle-detection heuristics on the NFT transfer graph. The community analytics layer, not the marketplace contract layer, was where T7.004 detection lived at v0.1.
• The NFT itself is the laundering instrument. Unlike T7.001 (where coins pass through a mixer) or T7.003 (where coins hop across chains), T7.004 uses the NFT transfer as the obfuscation primitive. The NFT changes hands between controlled addresses; the marketplace records a trade; the reward contract distributes tokens. The reward token's on-chain provenance terminates at the reward contract — there is no direct link from the reward token back to the synthetic nature of the volume that earned it.

What this example tells contributors writing future Technique pages

• LooksRare/X2Y2 wash-trading is the canonical T7.004 cohort at the highest nominal volume scale. $18B+ in synthetic wash-trade volume across two marketplaces in ~6 months makes this the largest-by-volume T7.004 instance documented at v0.1.
• T7.004 is structurally distinct from T5.002 even though both involve reward-farm extraction. T5.002 (Slow LP Trickle / Value Extraction) extracts from LP-provider rewards; T7.004 extracts from marketplace trading rewards. The extraction mechanism is structurally parallel but the on-chain primitive (LP provision vs NFT transfer) is different, and the detection methodology (LP-position-duration analysis vs NFT-transfer-cycle detection) is correspondingly different.
• The Blur airdrop wash cohort (examples/2023-02-blur-airdrop-wash-cohort.md) is the next-cycle surface. Contributors tracing T7.004 evolution should note that Blur's Points-based reward model was a response to the LooksRare wash-trade critique — but the Blur model was itself wash-traded at scale, demonstrating that T7.004 is resilient to reward-mechanism changes as long as the marketplace rewards a metric that can be synthetically inflated.

Public references

• Dune Analytics. hildobby dashboard: "LooksRare & X2Y2 Wash Trade Analysis." 2022.
• CryptoSlam. NFT volume rankings with wash-trade flags, 2022.
• Dune Analytics. sealaunch dashboard: "NFT Wash Trading on LooksRare." 2022.
• Cross-reference: T7.004 (NFT Wash Laundering) at techniques/T7.004-nft-wash-laundering.md.
• Cross-reference: T5.002 (Slow LP Trickle / Value Extraction) at techniques/T5.002-slow-lp-trickle-value-extraction.md.
• Cross-reference: examples/2023-02-blur-airdrop-wash-cohort.md — Blur airdrop wash cohort (next-cycle T7.004 surface, 2023).

Proposed new BibTeX entries

@misc{hildobby2022looksrare,
  author = {{hildobby}},
  title = {LooksRare \& X2Y2 Wash Trade Analysis — Dune Analytics Dashboard},
  year = {2022},
  note = {80–95\% of LooksRare volume estimated as wash-traded; the canonical community-side T7.004 detection layer}
}

@misc{cryptoslam2022wash,
  author = {{CryptoSlam}},
  title = {NFT Volume Rankings with Wash-Trade Flagging},
  year = {2022},
  note = {Adjusted NFT marketplace volume rankings flagging wash-traded venues; the venue-side T7.004 response}
}