Differences in Asset Scale and Volume Between CEX and DEX: How Hyperliquid Faces the Challenge?

Centralized exchange ( CEX ) far exceeds decentralized exchanges ( DEX ) in terms of asset accumulation scale and volume. The asset scale of CEX is about a hundred times that of DEX, while the volume is about ten times that of DEX. In the face of such a huge gap, how does Hyperliquid avoid being constrained by certain large trading platforms? This mainly relies on its decentralized structural design and algorithmic mechanism.

When we talk about "decentralized" exchanges, we are actually discussing which aspects need to be decentralized? Is it asset custody, trade matching, price discovery, or settlement mechanisms? These questions are worth deep reflection.

JELLYJELLY Contract Controversy: The Battle Between Centralization and Decentralization

In March 2025, the JELLYJELLY contract triggered a market upheaval on the Hyperliquid platform. The contract price soared by 429% in just a few hours, nearly triggering a massive liquidation. If liquidation occurs, short positions will be pushed into the on-chain liquidity pool HLP, potentially causing tens of millions of dollars in floating losses. As on-chain positions were precarious, certain trading platforms also unusually rushed to launch JELLYJELLY perpetual contract trading.

As the situation was about to spiral out of control, Hyperliquid validators urgently voted to intervene, forcing the delisting of the contract and freezing trades. This action raised questions in the community about "decentralized" exchanges.

This event has not only become the focus of heated discussion in the crypto community, but it also exposes a core issue: what exactly determines the price on decentralized trading platforms? Who bears the risk? Is the algorithm truly neutral?

This article will take the JELLYJELLY event as a starting point to analyze the algorithmic differences in the core mechanisms of perpetual contracts on three major platforms regarding the index price, mark price, and funding rate (, and explore the financial concepts and risk transmission mechanisms behind them. We will see how different algorithms shape trading styles, serve different types of traders, and how they affect traders' survival capability during market storms.

This is not only an analysis of contract technology, but also a philosophical contest about the design of market order.

Three Key Elements of Perpetual Contract Trading

Before diving into the discussion, let's first clarify the three core components of perpetual contract trading:

1. Index Price: Tracks changes in the spot market price and serves as a theoretical benchmark. Hyperliquid refers to it as the Oracle price).

2. Mark Price: The decisive price used to calculate unrealized profits and losses, trigger liquidation, and other key events.

3. Funding Rate: An economic mechanism connecting the spot and futures markets, guiding the contract price to revert to the spot.

Comparison of Core Algorithms of Three Major Platforms

1. Index Price/Oracle Price

The price of Hyperliquid's oracle is completely independent of its own market and is constructed by validator nodes. It uses a weighted median method to resist extreme price fluctuations, demonstrating strong anti-manipulation properties, but the update frequency is relatively slow, updating once every 3 seconds. This design aims to eliminate outliers and fluctuations, making the price smoother.

2. Marking Price Mechanism

Hyperliquid's marked price algorithm integrates multiple price sources:

• The 150-second exponential moving average of the difference between Oracle price and contract mid price
• The platform's own buy and sell prices are the median of the latest transaction price.
• Weighted median of the mid price for multiple CEX perpetual contracts

Validators are responsible for regularly updating the Oracle and Mark Price, and for consistency verification of input sources. This mechanism creates a certain degree of "algorithmic democracy," enhancing resistance to manipulation.

3. Funding Rate Algorithm

Hyperliquid introduces a premium index in the funding rate calculation, sampled every 5 seconds and calculated as an hourly average to prevent short-term volatility. To compensate for the limitations of price reversion, Hyperliquid employs three distinctive settings:

• In extreme cases, the funding rate can reach up to 4% per hour.
• Calculate funding fees based on oracle price instead of mark price
• Charge 1/8 of the 8-hour funding rate every hour

This "small step fast run" approach aims to accelerate price return and compensate for the limitations of the order book depth model.

Trading Strategies and Financial Philosophy Adapted to Different Platforms

1. A large trading platform: The design of rational systemizers

• Core Concept: "Make the market predictable"
• Mechanism Features: Smoothed Mark Price, Fine Modeling of Funding Rate, Risk Buffer Mechanism
• Suitable Audience: Institutional investors and medium to long-term traders seeking stable returns with controllable risks.

2. A trading platform: The design of the trading instinct.

• Core concept: "The market is a reflection of human nature"
• Mechanism Features: Price marking sensitivity, large fluctuations in funding rates, rapid liquidation mechanism
• Suitable for: high-frequency traders, short-term traders, and speculators who are good at capturing instantaneous price deviations.

3. Hyperliquid: The Design of On-Chain Structuralists

• Core Concept: Decentralized Governance + Programmable Pricing Mechanism
• Mechanism features: Validator consensus price, HLP Vault support, high-frequency funding rate, on-chain transparency
• Suitable for: traders pursuing transparency and decentralization, on-chain arbitrageurs, extreme traders

Conclusion

The algorithm designs of different platforms reflect different understandings of the nature of the market. One large platform pursues stability, another emphasizes market behavior, while Hyperliquid attempts to establish a new consensus through on-chain governance. However, the JELLYJELLY incident shows that even decentralized systems may require human intervention in extreme situations.

In the future financial world, algorithms will continue to expand their influence. However, we must recognize that every algorithm carries value judgments behind it. Whether pursuing freedom, fairness, or transparency, traders are ultimately searching for a sense of order. Let us always maintain a sense of awe for the market and take responsibility for our choices.