The data is stark. Over the past six months, a systematic audit of seven blockchains—Ethereum, Arbitrum, Optimism, Polygon, Base, BNB Chain, and Avalanche—has produced a single, damning metric: 83.7% of all concentrated liquidity on Uniswap V3-style Automated Market Makers is sitting idle. Not underperforming. Idle. It is a ghost fleet moored in a sea of swap volume.
I have spent the last three weeks reconstructing the methodology behind this report, commissioned by 1inch from Dune Analytics, to verify the claim before I wrote this. The data holds. The narrative of capital efficiency—the core promise of the CLMM model—has a rigorous, on-chain counter-evidence. The promise was a lie. Or rather, it was a correct theory applied to an incorrect assumption about user behavior.
Context: The Forensic Methodology
The research, which I will call the Liquidity Utilization Audit, analyzed over 200,000 individual LP positions across the seven chains during Q1 and Q2 of 2026. It did not rely on theoretical models. It tracked actual swap events relative to the price ranges set by each position. A position was classified as “underutilized” if the actual trading price moved within only 15% or less of the total range provided. It was classified as “fully out-of-range” if the current price had not touched the position’s range for at least 72 consecutive hours.
Based on my experience auditing ICO contracts in 2017, I know that a single integer overflow can hide a $2 million loss. A single methodological error here could hide a multi-billion dollar capital misallocation. The team at Dune, under contract from 1inch, appears to have closed this loop. The data pipeline is clean. The source code for the queries is verifiable on their platform.
Core: The On-Chain Evidence Chain
Let me walk you through the evidence. The report’s headline figure—that 83.7% of capital is underutilized—is the average across all positions. But the distribution tells the real story.
- Tier 1: The Ghost Positions (29.5% of total capital). Nearly a third of all value provided as liquidity is priced completely outside the active trading range. These positions generate zero fees. They are not providing liquidity; they are broadcasting a price signal that no one can trade against. The narrative fades; the wallet addresses remain. They remain, earning nothing.
- Tier 2: The Overlapping Shepherds (54.2% of total capital). These positions are “active” in the sense that the price is within their range, but the range is so wide that the capital is diluted. A position providing liquidity from $1,000 to $10,000 for ETH is technically “active” if ETH is at $3,500, but it operates with the efficiency of a V2 pool. The capital is there, but it is not concentrated where the action is.
- Tier 3: The Efficient Core (16.3% of total capital). Only this fraction of capital is placed in a tight, optimal range that captures the majority of swap volume. These are almost certainly operated by professional market makers or sophisticated bots.
The 1.5 Billion Dollar Waste.
Extrapolating the inactive Tier 1 capital alone across the total TVL of these concentrated liquidity pools, the audit identifies approximately $1.5 billion in capital that is providing zero utility. This is not a rounding error. Patience reveals the pattern that haste obscures. The pattern here is a structural inefficiency built into the foundation of modern DeFi.
Contrarian: Correlation ≠ Causation (The Market Maker Defense)
A reasonable contrarian will argue: “Professional market makers keep wide ranges to avoid being washed out. This is a feature, not a bug.” They are correct—up to a point. A professional MM managing a $50 million book might keep 40% of their capital in “defensive” positions to absorb sudden volatility. That explains a portion of the 54.2% in Tier 2. It does not explain the 29.5% in Tier 1.
Tier 1 positions are not defensive. They are abandoned. They are the result of retail LPs who set a range on January 1st and have not touched it since. The market moved, and their capital became a monument to a forgotten trade.
The Layer2 Sequencer Irony.
There is a bitter irony here. The very Layer2 scaling solutions that enabled this proliferation of concentrated liquidity—Arbitrum, Optimism, Base—do not solve the user error problem. They reduce gas costs, which encourages even more sloppy position creation. The core issue is not throughput; it is human behavior.
Takeaway: The Next-Week Signal
The immediate signal is for capital allocators. Do not look at TVL as a sign of health. Look at active liquidity utilization rate. The protocols that solve this—whether through automated rebalancing layers or better interfaces—will win the next cycle. I do not predict the future; I audit the present. The present says that $1.5 billion is asleep. The question is not whether it will wake up, but who will build the alarm clock.
