Data Research: How big is the liquidity gap between Hyperliquid and CME crude oil?

Original author / Castle Labs

Compiled by / Odaily Planet Daily Golem

Editor's note: This article systematically studies the differences in crude oil contract trading data between Hyperliquid and CME during weekdays and weekends, drawing some important conclusions. Currently, Hyperliquid indeed cannot compete with CME in absolute metrics such as liquidity depth or slippage, with overall liquidity being less than 1%, which is related to the fact that the main users of the RWA trading platform are still crypto-native retail investors.

What sets Hyperliquid apart is that the trading volume of crude oil contracts on Hyperliquid significantly increases during the weekend. This indicates that besides retail investors with speculative demand, traders who want to gain exposure to crude oil trading and hedge their positions before Monday are also trading on Hyperliquid. This trend is becoming increasingly evident, giving Hyperliquid the capability for price discovery in commodities.

However, for institutional investors, the high trading costs on the Hyperliquid platform compared to CME remain a major barrier to its expansion in the commodity trading field. If Hyperliquid does not improve its ability to handle institutional-level orders soon, it can only serve as a temporary trading venue for traditional traders on weekends, ultimately becoming a minor supplement in the traditional financial landscape.

Research Methodology and Data Sources

This analysis evaluates the microstructure of the crude oil market through two studies, covering both weekdays and weekends, utilizing tick data from two trading venues: Hyperliquid's xyz:CL perpetual contract and CME's CLJ6 (April 2026 NYMEX WTI crude oil futures) contract.

CME data is sourced from the Databento trading data source, which captures tick data rather than order book snapshots. Therefore, all depth and slippage estimates for CME are based on actual transaction volumes rather than quoted depth. Hyperliquid data is sourced from Hyperliquid's publicly available S3 database, which contains complete on-chain transaction records.

Thus, the analysis of both trading venues is based on actual transaction volumes. All depth data represents explicit liquidity, i.e., the volume within a specific basis point range around the VWAP midpoint over a 5-minute window, rather than the complete resting depth on the order book.

Research Period and Market Background

The research period is from February 27, 2026, to March 16, 2026, coinciding with the geopolitical turmoil following Iran's attack on February 28, 2026.

• Market close before the attack: The last CME trading day before the attack event.
• Monday opening: The market reopened under significant pressure, with CME prices gapping up sharply, while the Hyperliquid xyz:CL market was constrained by discovery boundaries.
• Subsequent weekends: Due to persistently high oil prices, market volatility led to sustained crude oil trading volumes on the Hyperliquid platform.

xyz:CL was launched in early 2026, meaning the observation period for these three weekends covers the early maturation phase of the Hyperliquid market. Observed trends, including increased liquidity depth, rising trading volumes, and growing user numbers, partially reflect market maturation. However, we believe that on-chain exchanges currently cannot compete with traditional exchanges in absolute metrics such as liquidity depth or slippage.

The goal of our research is to track directional trends: whether the price spread between the two is narrowing, how quickly it is narrowing, and under what conditions it narrows.

Data Analysis

Data analysis is divided into two parts by time period:

• Weekday period: Covering a complete three-week period, comparing the depth, slippage, and premiums/discounts traded by Hyperliquid and CME during weekdays. For Hyperliquid, we also analyzed its financing rates throughout the period.
• Weekend period: Within the given time frame, covering three weekends, we analyzed price discovery and the price gap deviation of Hyperliquid relative to CME's opening price.

Weekday Period Data Analysis

This analysis covers a complete three-week period, focusing on the times when both exchanges are active.

Liquidity depth is measured by the dollar trading volume within ±2, ±3, and ±5 basis point ranges around the VWAP midpoint in each 5-minute interval, summarized as the median for all weekday intervals. As mentioned, this reflects the volume within the interval rather than the resting quote depth. This method may underestimate the liquidity depth of both CME and Hyperliquid.

Execution slippage is estimated using a synthetic order book sorted by execution price. In each 5-minute period, observed market orders are arranged in ascending order of price (simulating sell orders), and sell orders are executed sequentially until the target order size is reached. The target price is set as the lowest transaction price within that period (representing the best selling price at the time the order arrives). Slippage is calculated as the difference between the volume-weighted average price (VWAP) and the target price, expressed in basis points. This method is applied to incremental order sizes ranging from $10,000 to $1,000,000.

The Hypeliquid-CME basis during weekdays: Tracks the signed price difference between the Hyperliquid midpoint and the CME latest price within 5-minute windows across all weekdays. This reflects any structural premium or discount of Hyperliquid relative to CME's reference price during active periods. The Hyperliquid midpoint is derived from the volume-weighted average price (VWAP) of transactions within each 5-minute trading period, rather than real-time order book quotes.

Hyperliquid financing rates are priced hourly, expressed in basis points per hour.

Weekend Period Data Analysis

This analysis focuses on three different weekend market closures for CME:

• W1: February 28 to March 1, 2026
• W2: March 7 to March 8, 2026
• W3: March 14 to March 15, 2026

During W1 and W2, the Hyperliquid perpetual contract was constrained, so the marked price could not exceed the "range limit boundary" (DB). When the oracle price freezes (for example, when the main reference market (CME) is closed and external price data sources stop updating), the protocol effectively limits the price to a narrow range.

For each weekend window, we report key data metrics for Hyperliquid xyz:CL, including price, volume, and number of trades. To measure the price gap deviation at Monday's opening, for each weekend, we measure the price difference between Hyperliquid and CME at three reference points:

• 3 hours before CME reopens
• 1 hour before CME reopens
• At CME opening (T=0)

All price gaps are expressed in basis points, with positive values indicating Hyperliquid is above CME's opening price and negative values indicating a discount.

Quantitative Analysis

This section first conducts an analysis comparing the liquidity conditions of the Hyperliquid xyz:CL HIP-3 crude oil market with NYMEX CLJ6 during overlapping weekday periods.

Liquidity Depth: Hyperliquid Less Than 1% of CME

There is no doubt that the liquidity situation of on-chain exchanges is starkly different from CME. The average liquidity depth of CL on Hyperliquid is less than 1% of CLJ6, with consistent liquidity depth across price ranges (109 times at ±5 bps). In the ±2 bps range around the midpoint price, CME's executable depth is $19 million, while Hyperliquid's is only $152,000, a difference of 125 times.

Given the novelty of the CL market on Hyperliquid and the different target user groups, this result is not surprising. The primary value of on-chain exchanges lies in providing permissionless trading channels for users who are traditionally excluded from institutions like CME.

However, as weekend trading volumes on DEXs like Hyperliquid grow, perceptions of these platforms are beginning to shift, and institutional investors are increasingly interested in hedging positions during non-trading hours. Therefore, creating a market environment suitable for traditional investors and retail traders is becoming increasingly important for Hyperliquid.

For retail traders with a trading volume of $10,000, this cost difference is negligible. But for institutional investors with trading volumes exceeding $1 million, the on-chain trading costs of CL (and most other markets) remain prohibitive.

In fact, the inherent differences in user groups are reflected in the median trade sizes during these overlapping market periods.

The 166-fold difference in median trade size (90,450 USD vs 543 USD) clearly demonstrates the fundamental differences in the user groups served by these trading venues. The median trade size of CLJ6 is comparable to a standard crude oil futures contract (approximately $94,000 nominal value at current prices), while Hyperliquid's median trade size is $543, reflecting the leveraged directional bets of crypto-native retail traders.

We expect that as these markets become increasingly legitimized in the eyes of more traditional investors and funds shift to on-chain, the median trade size in the Hyperliquid commodity market will reach a turning point.

To further distinguish different trade sizes, we conducted order simulations with order size limits ranging from $10,000 to $1,000,000.

For a $10,000 order, CLJ6 traders experience no slippage, which is consistent with expectations, while Hyperliquid users have a median execution slippage of less than 1 basis point, at 0.77 basis points. The gap appears at the $100,000 order size, where Hyperliquid users' slippage rises to 4.33 basis points, approaching the 5 basis point threshold, while CME CLJ6 shows no slippage.

Notably, this is above the median trade size of the CLJ6 market (90,450 USD).

At a trading size of $1 million, Hyperliquid's 15.4 basis points is about 20 times CME's 0.79 basis points, confirming that this trading venue currently lacks the capacity to handle institutional-level orders. Considering Hyperliquid's average trade size, the platform could easily provide the same quality of service to users without incurring slippage.

CLJ6 orders only begin to show significant slippage around a trading size of $500,000, affecting their execution.

When we extend the order size analysis to weekends, slippage for all order sizes decreases, especially for $100,000 and $1 million order sizes, indicating that the market has matured. Over the three weeks analyzed, the slippage reduction for simulated orders is as follows:

• $10,000: -16%
• $100,000: -75%
• $1,000,000: -86.9%

Financing Rates

The financing rates for CL fluctuate significantly during CME's closing periods but fluctuate less during delivery periods. This helps us reveal the internal pricing dynamics of the market during non-trading hours. Weekend openings mean that the CL market can utilize internal price discovery mechanisms (supported by DB and other risk-reduction mechanisms). Therefore, financing rates are expected to be more volatile, as highlighted below.

During active trading periods, Hyperliquid's xyz:CL closely follows the trends of CME's CLJ6, but as oil prices rise, a structural discount appears and expands, likely due to financing rate pressure from accumulated long positions. During weekends, with CME closed, Hyperliquid's price discovery is further constrained by the price range mechanism (DB), which limits the volatility of the marked price in the absence of a real-time reference market.

Weekend Period Analysis: Hyperliquid Has Price Discovery Capability

These three weekends showcase the rapid maturation process of the Hyperliquid market:

W1: February 28 to March 1, 2026 (Iran Attack)

Prices on Hyperliquid rose from around $67.29 near CME to about $70.80, accounting for approximately 45% of the final gap up to $75 on Monday (+1146 basis points).

It is important to note that due to the ±5% price range limit mechanism (DB) mentioned above, price discovery during this weekend was constrained. This explains why the curve in the chart is relatively flat and why there was a gap up on Monday. Nevertheless, in the first second after the paired data was released, the gap between Hyperliquid xyz:CL ($73.89) and CME CLJ6 ($75) was within 1.5%.

This is not a "mistake" or "failure," but rather a risk protection achieved through market design. Therefore, from a data perspective, the correlation during the first weekend is the lowest, but it highlights that xyz:CL reacted to the initial shock of the Iranian attack while also recognizing the importance of DB as a weekend price discovery mechanism, especially for emerging markets.

W2: March 7 to March 8, 2026

The second weekend was the real test, as xyz:CL reached the boundary price of the range at the market close. The opening price for CLJ6 was $98 (up 737 basis points from the $91.27 closing price), while xyz:CL peaked at around $95.83, capturing only 68% of the increase.

During the second weekend, xyz:CL better captured market trends and was closer to CME's opening price than the previous weekend.

W3: March 14 to March 15, 2026

Data from the third weekend indicates that in a calmer market environment, Hyperliquid can more reliably predict CME's final opening direction**.**

This weekend, the convergence of xyz:CL and CLJ6 reached its best: up 226 basis points from CME's closing price, slightly above the opening price on Monday by 62 basis points. CLJ6's closing price on Friday was $99.31, with an opening price of $100.93 (up 163 basis points), while xyz:CL opened at $101.56.

Overall, these three snapshots show the structural changes in the xyz:CL market on the Hyperliquid platform, transitioning from an emerging market constrained by DB price discovery (weekends 1 and 2) to increasingly free price discovery, with overshooting and pullbacks (weekend 3).

Analyzing the price deviation errors at different times before CME's opening (3 hours, 1 hour, 0 hours) across different weekends reveals that W3's data is the most reliable, as the xyz:CL market was influenced by DB during the first two weekends. In W3, the errors for xyz:CL were approximately +70 and -139 basis points, respectively, 3 hours and 1 hour before CME's opening, indicating its price discovery capability was superior to the previously analyzed weekends.

Other Indicators

We also provide additional indicators for the weekend summary analysis, including trading volume, total number of trades, and average trade size. These indicators varied between weekends and showed continuous growth over several weekends.

The total trading volume of the xyz:CL market grew from $31 million to over $1 billion over the three weeks, reflecting an increase in user numbers and the market's eventual maturation.

Additionally, the total number of trades increased from 26,000 in the first weekend to over 700,000 in the third weekend.

Notably, the average trade size over the weekend actually grew from the median we previously mentioned to $534. All three weekends observed the same growth trend, which may indicate more institutional funds flowing into the market.

The average trade size for the first weekend was $1,199, growing to over $1,500 by the third weekend.

This may indicate that the user group using the platform on weekends has changed, with fewer retail users and more traders needing to gain exposure to crude oil trading before Monday, thus weekend trading is more aligned with hedging needs rather than speculation.