Hyperliquid built its moat on deep liquidity for long-tail assets, but over the past year its median USD depth within 5 bp shrank by 84.6%. Trader matteo compared data from Binance and OKX to break down how large market makers’ rebate advantages and the L1 congestion quota mechanism are gradually squeezing smaller market makers out of the market. (Previously: Is AI making code worthless? Multicoin breaks down DeFi’s three real profit-making moats) (Background: A GitBook SSO vulnerability breached production! Hyperliquid perpetuals platform TradeXYZ announces its departure) Trader matteo said in a post on X that Hyperliquid’s median USD depth within 5 bp shrank by 84.6% over the past year, a markedly larger liquidity loss than the 61% for comparable assets on Binance and 78% on OKX. Comparing a sample of 86 long-tail altcoins, he found that even after accounting for falling token prices, the number of orders near the center of the order book had also fallen sharply, putting smaller market makers at risk of structural displacement. gossip (read) priority fees on hyperliquid are broken currently the team that is benefitting the most is wintermute (insert fist shaking red emoji guy meme) this is not publicly known but can be determined with certainty by mapping out transfers to labeled wallets from the top… pic.twitter.com/K20Eclsl5x — Loris (@0xLoris) October 23, 2026 incredible things are happening in the hype discord pic.twitter.com/AykXNdKsFi — Ludwig (@0xvanbeethoven) November 4, 2026 Hyperliquid’s long-tail asset liquidity has deteriorated sharply over the past year. Across comparable markets, median USD depth within 10 basis points fell 82%, versus 61% on Binance and 78% on OKX. For the same tokens, Hyperliquid’s median 5-basis-point depth relative to… — Coincamps (@coincamps) November 11, 2026 Depth shrank by 84.6%, with Hyperliquid losing ground to Binance matteo selected 86 assets that were listed on Binance, OKX, and Hyperliquid during the period from November 1, 2025, to November 1, 2026, and offered leverage of less than 10x, as his long-tail sample. Among the 38 markets comparable within 10 bp, Hyperliquid’s median USD depth fell 82%, compared with 61% on Binance and 78% on OKX—a substantial difference in the rate of decline. To isolate the impact of the broader market, he divided each asset’s depth on Hyperliquid by its corresponding depth on Binance, then compared the ratios from a year apart. At 5 bp depth, the median ratio fell 57.5%, showing that even for the exact same tokens, Hyperliquid lost substantial liquidity relative to Binance. Orders near the midpoint plummet, while the top 20 levels move in the opposite direction To account for the possibility that falling token prices were shrinking USD depth, matteo tracked both USD depth at quoted order-book prices and the number of tokens on offer. With November 2025 as a baseline of 100, one year later the USD index within 5 bp had fallen to around 13, while the token-quantity index dropped to 28. This means that even after removing the impact of token prices, a substantial amount of liquidity had disappeared from prices near the market. But when the observation range was expanded to the top 20 order-book levels, the USD index stood at 61, while the token-quantity index actually rose to 127. This means the order book still has volume, but the order sizes at prices close to the market—where traders actually want to execute—have fallen sharply, with liquidity clearly shifting further away. Spreads catch up with CEXs, then suddenly diverge; late October marks the turning point Over the past year, Hyperliquid’s bid-ask spreads had been steadily catching up with Binance and OKX, but recently they suddenly diverged. Starting on October 30, as liquidity became increasingly concentrated near execution prices, spreads widened noticeably—a trend closely aligned with market makers’ recent complaints about the centralization of market structure. Comparing October 8–14 with October 31–November 6, the median share of depth within 25 bp held by the largest account group in each market rose from 34.5% to 38.6%, while the median quoted spread widened from 2.13 bp to 3.55 bp. However, matteo cautioned that this period overlapped with Binance’s October 29 security incident, so the fluctuations should not be attributed directly to priority fees or network congestion based on the chart alone. Market-making competition has not eased, but there are indeed fewer participants matteo traced the underlying account structure, grouping master accounts and explicitly linked subaccounts into the same account family to avoid double-counting. Comparing data for the same 80 markets across one week a year ago and a recent week, the median share of filled orders accounted for by the largest account family actually fell from 25.2% to 14.6%, contradicting the claim that a single giant dominates the entire market. But the number of active market makers did decline. Among the same 72 markets, the median average number of account families posting at least $1,000 on both sides within 25 bp fell from 7.64 to 6.39. Even when adjusted using fixed token prices, the number of active market-making accounts fell in 58 markets. Fee structure and L1 congestion squeeze smaller market makers from both sides The fee structure creates a structural disadvantage. Historical maker fees were analyzed for 54 account families: the bottom quartile by trading volume had to pay fees of 0.961 bp, while the top quartile actually received rebates of 0.236 bp. Before even accounting for hedging costs and adverse selection, smaller market makers start with a disadvantage of about 1.20 bp. In a market where the entire spread is only a few basis points..."Hyperliquid’s Depth Shrinks 84.6%: Why Are Small Market Makers Leaving en Masse?" This article was originally published on BlockTempo (the most influential blockchain news outlet in the Chinese-speaking world).