Skip to main content

Livepeer Foundation — Review summary (IDOL / Arrakis DEX liquidity proposal)

As-of snapshot: 2025-12-20 (re-run scripts to refresh numbers; results depend on onchain state and prices).

This document summarizes (1) what the proposal asks for, (2) the research we ran to evaluate it, and (3) the key results + recommended guardrails.

1) Proposal overview (what’s being asked)​

From the forum pre-proposal (“IDOL – Improving Dex / Onchain Liquidity”, topic 3151):

  • Allocate 250,000 LPT (~$1M) from the DAO treasury.
  • Move funds to a Livepeer Foundation-controlled multisig (proposal describes a 2/3 multisig).
  • Deposit into an Arrakis Pro-managed concentrated-liquidity vault for LPT/ETH.
  • The proposal references “Uniswap v4”, while also referencing an “existing 0.3% LPT/ETH pool” (which, on Arbitrum today, is a Uniswap v3 pool).
  • Fees described:
    • AUM fee: 1% / year (waived for first 6 months)
    • Performance fee: 50% of trading fees generated by the vault
  • Strategy intent: begin with one-sided LPT and bootstrap toward ~50/50 LPT/ETH inventory over time.

2) Research performed (reproducible artifacts)​

All results below are derived from scripts + cached artifacts in this working folder:

  • Forum snapshot: cached under data/forum-posts-3151/
  • Onchain slippage quoting (live state via Uniswap v3 Quoter) for the Arbitrum LPT/WETH 0.30% pool:
    • Script: scripts/onchain_slippage_univ3.py
    • Output: outputs/onchain-slippage.csv
  • Onchain swap analytics (volume/fees + directional flow) via Swap logs:
    • Script: scripts/univ3_swap_analytics.py
    • Outputs: outputs/lpt-univ3-swap-analytics-24h.json, outputs/lpt-univ3-swap-analytics-30d.json
  • Address concentration analytics (“wash-trade surface”) via Swap logs:
    • Script: scripts/univ3_swap_address_analytics.py
    • Outputs: outputs/lpt-univ3-swap-address-analytics-24h.json, outputs/lpt-univ3-swap-address-analytics-30d.json
  • Empirical reversion + chunking analysis (how quickly the pool reverts after small swaps; best-case chunked execution):
    • Scripts: scripts/univ3_reversion_analysis.py, scripts/chunked_trade_analysis.py
    • Outputs: outputs/univ3-reversion-lpt-1d.json, outputs/chunked-trade-lpt-25k-by-1k.json
  • Fork-based “does ~$1M liquidity move the needle?” sanity check (Hardhat Arbitrum fork; mint additional 50/50 liquidity into the existing pool; re-quote via Quoter):
    • Output examples: outputs/fork-liquidity-impact-782k-range60.json, outputs/fork-liquidity-impact-782k-range200.json, outputs/fork-liquidity-impact-1m-range200.json
  • IL + breakeven economics model (historical LPT/ETH ratio volatility; passive 50/50 IL to highlight required volume):
    • Script: scripts/il_models.py
    • Outputs: outputs/il-model-summary.json, outputs/breakeven-volume.csv
  • Audit PDF extraction + summaries (Arrakis modular core + Uniswap v4 module):
    • Script: scripts/audit_summaries.py
    • Output: outputs/arrakis-audit-summary.md

3) Key results (as-of 2025-12-20)​

A) Onchain slippage is severe at $25k–$50k sizes (today)​

Arbitrum LPT/WETH Uniswap v3 0.30% pool: 0x4fD47e5102DFBF95541F64ED6FE13d4eD26D2546

Quoted against live onchain state via Uniswap v3 Quoter (note: these “impact” numbers include the 0.30% LP fee):

Notional (USD)Buy impactSell impact
$1,000~0.60%~0.59%
$5,000~1.78%~1.73%
$10,000~3.25%~3.12%
$25,000~7.67%~7.14%
$50,000~15.04%~45.26%

Source: outputs/onchain-slippage.csv

B) Current DEX flow is modest (fee income is small at today’s volume)​

Computed from onchain Swap logs (same pool, 0.30% tier):

  • 24h: ~$74k volume, ~$223 total fees
  • 30d: ~$1.35M volume, ~$4,053 total fees

Illustration (if a vault captured ~100% of fees, and the DAO receives 50% of trading fees):

  • 24h: DAO fees ≈ $112/day
  • 30d: DAO fees ≈ $2,027/30d

Sources: outputs/lpt-univ3-swap-analytics-24h.json, outputs/lpt-univ3-swap-analytics-30d.json

C) “Chunking” can reduce effective slippage for sophisticated users, but it’s not one-click UX​

Empirical reversion (24h window; swaps $500–$2,000 notional; “revert” defined as returning within ±2 ticks within next 10 swaps):

  • candidates: 32
  • revert rate: 46.9%
  • median revert latency: ~22s (p90 ~61s)

Best-case chunked execution (assume full revert between chunks):

  • $25k single swap impact: ~7.67% buy / ~7.14% sell
  • $25k split into 25× $1k chunks: ~0.60% buy / ~0.59% sell

Sources: outputs/univ3-reversion-lpt-1d.json, outputs/chunked-trade-lpt-25k-by-1k.json

D) Adding ~$1M liquidity can plausibly “fix depth”, but placement matters (fork sanity check)​

Hardhat fork experiment: mint additional 50/50 liquidity into the existing pool, then re-quote.

This is not a forecast of Arrakis performance (no active management), but it is useful for magnitude:

  • Add ~$782k 50/50 liquidity:
    • “Moderate” range (±60×tickSpacing): $25k impact ≈ ~1.24%, $50k sell ≈ ~2.12%
    • “Wide” range (±200×tickSpacing): $25k impact ≈ ~2.40%, $50k sell ≈ ~4.26%
  • Add ~$1M 50/50 liquidity (±200×tickSpacing): $25k impact ≈ ~2.05%, $50k sell ≈ ~3.62%

Sources: outputs/fork-liquidity-impact-782k-range60.json, outputs/fork-liquidity-impact-782k-range200.json, outputs/fork-liquidity-impact-1m-range200.json

E) Fee economics: breakeven volume is far above today’s volume under a 50% fee split​

Using a simplified passive 50/50 IL model to illustrate scale (for $1,000,000 capital, 0.30% fee tier, DAO retains 50% of trading fees → effective capture ~0.15% of volume):

  • 180d “median” IL (4.56%): requires **$169k/day** sustained volume to breakeven on fees alone
  • 180d “worst” IL (16.78%): requires **$622k/day**

Source: outputs/breakeven-volume.csv (see also outputs/il-model-summary.json)

F) Asymmetric / hidden-risk highlights (even with honest actors)​

  • Incentive mismatch: vendor fee is on gross trading fees; treasury bears IL/LVR + tail risk (fees ≠ net performance).
  • “Exit liquidity” dynamic: deeper DEX liquidity disproportionately benefits large sellers; inventory drift can leave the vault LPT-heavy.
  • Wash trading is not free by default, but becomes profitable quickly if any external per-volume reward exists.
    • Rule-of-thumb break-even external reward rate ≈ (DAO fee share) × (pool fee tier) (e.g., 50% × 0.30% ≈ 0.15% of volume), plus gas/slippage.
  • If Uniswap v4/hooks are involved, audit summaries highlight approvals + hook surfaces that require strict role/approval hygiene.

Supporting docs: reports/risk-assessment.md, reports/asymmetric-opportunities.md, outputs/arrakis-audit-summary.md

4) Recommendation: treat as a pilot with explicit guardrails​

If improving onchain liquidity is a priority, a staged pilot is the highest-signal / lowest-regret structure:

  • Clarify chain + exact deployment: v3 vs v4, pool/vault/module/hook addresses, and whether anything is upgradeable.
  • Tranche funding: start smaller (e.g., $100k–$250k equivalent), time-boxed (e.g., 6–8 weeks), with explicit scale/stop checkpoints.
  • Pre-commit KPIs:
    • repeated quoter-based slippage at $5k/$10k/$25k both directions
    • 7d/30d volume + fee generation
    • inventory drift bounds + time-to-reach target inventory ratio (if bootstrapping)
    • net performance vs a benchmark (e.g., HODL and/or passive LP proxy)
  • Stop conditions: mandate breach, drift outside bounds, material underperformance, or any security incident.
  • Fee-structure improvements (if possible): lower/cap the performance fee and/or tie performance fee to net performance vs benchmark.
  • Operational safety: Foundation multisig should retain practical “kill switch” withdrawal ability; enforce least-privilege approvals/roles.

5) Supporting materials (full writeups)​

  • Executive summary: reports/executive-summary.md
  • Risk assessment: reports/risk-assessment.md
  • Adversarial deep dive: reports/asymmetric-opportunities.md
  • Cross-pool case studies: reports/arrakis-case-studies.md

6) Re-run / reproduce (from blockchain/proposal-review/livepeer-idol-dex-liquidity/)​

  • Forum snapshot: python3 scripts/fetch_forum_topic.py
  • Onchain slippage (includes tick-depth estimate): python3 scripts/onchain_slippage_univ3.py --pool 0x4fD47e5102DFBF95541F64ED6FE13d4eD26D2546 --include-tick-depth
  • Pool volume/fees:
    • 24h: python3 scripts/univ3_swap_analytics.py --pool 0x4fD47e5102DFBF95541F64ED6FE13d4eD26D2546 --token0-coingecko-id livepeer --token1-coingecko-id ethereum --days 1
    • 30d: python3 scripts/univ3_swap_analytics.py --pool 0x4fD47e5102DFBF95541F64ED6FE13d4eD26D2546 --token0-coingecko-id livepeer --token1-coingecko-id ethereum --days 30
  • Address concentration: python3 scripts/univ3_swap_address_analytics.py --pool 0x4fD47e5102DFBF95541F64ED6FE13d4eD26D2546 --token0-coingecko-id livepeer --token1-coingecko-id ethereum --days 30
  • Reversion + chunking: python3 scripts/univ3_reversion_analysis.py --pool 0x4fD47e5102DFBF95541F64ED6FE13d4eD26D2546 --token0-coingecko-id livepeer --token1-coingecko-id ethereum --days 1 --include-time and python3 scripts/chunked_trade_analysis.py --pool 0x4fD47e5102DFBF95541F64ED6FE13d4eD26D2546 --token0-coingecko-id livepeer --token1-coingecko-id ethereum --total-usd 25000 --chunk-usd 1000
  • IL + breakeven model: python3 scripts/il_models.py
  • Audit summaries: python3 scripts/audit_summaries.py