Analysis of Perpetual Futures Trading Mechanism: The Algorithm Battle of Three Major Platforms

In the cryptocurrency derivatives market, Perptual Futures trading has become an important component. This article will delve into the algorithmic differences in the core mechanisms of the three major platforms in Perptual Futures, exploring the financial philosophies and risk management approaches behind them.

Three Core Elements of Perpetual Futures Trading

Perptual Futures trading mainly consists of three key components:

1. Index Price: Reflects the changes in the spot market price and is the theoretical benchmark 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 that connects the spot and Futures Trading markets, guiding the Futures Trading price back to the spot.

Comparison of Algorithms of Three Major Platforms

Index Price Mechanism

A certain trading platform uses validator nodes independent of its own market to construct index prices, employing a weighted median method to combat extreme volatility, with an update frequency of every 3 seconds. This design enhances anti-manipulation, but the update speed is relatively slow.

Mark Price Algorithm

The marking price Algorithm of a certain trading platform is based on the median of three types of prices, including the mid-price of the best bid/ask in the Futures Trading market, the transaction price, and the impact price. The median constructed after EMA processing makes the marking price changes smooth and resistant to spikes, suitable for large capital stable allocation.

Another trading platform only uses the mid-price of the best bid/ask as the source of the marking price. The price is extremely sensitive to small trades, making it more suitable for high-frequency and short-term operations.

A certain decentralized platform combines the two aforementioned methods, integrating multiple price sources to calculate the marked price, including the EMA of the difference between oracle prices and the average price in the contract, the median price of the platform itself, as well as the weighted median from several centralized exchanges.

Funding Rate Algorithm

A certain decentralized platform has introduced a premium index based on traditional models, sampling every 5 seconds and calculating the hourly average to prevent short-term fluctuations. Its distinctive settings include:

1. Funding rate can reach up to 4% per hour in extreme cases.
2. The funding fee is based on the oracle price rather than the mark price.
3. The funding fee is charged step by step on an hourly basis.

The funding rate of a certain centralized platform relies on a longer settlement period, combined with order book depth and lending rates, to provide institutional investors with smooth funding cost expectations.

The algorithm of another centralized platform is relatively simple, based on the calculation of price deviation from the order book, with large fluctuations, making it suitable for high-frequency short-term strategies.

Trading Philosophy of Different Platforms

Design of System Rationalizers

A large centralized platform aims to "make the market predictable" by providing a controllable environment for institutional investors and medium to long-term traders through smoothing mechanisms and fine modeling.

Design of the Trading Instincts

A certain centralized platform adopts a "fast, fierce, and precise" strategy, accepting market irrationality, attracting high-frequency traders and short-term speculators to seek opportunities amidst intense volatility.

On-Chain Structuralist Design

A certain decentralized platform attempts to create a new financial paradigm by attracting traders seeking verifiable code and distributed governance through validator consensus pricing, on-chain transparency, and extreme funding rate design.

Conclusion

The algorithm designs of different platforms reflect varying understandings of the essence of the market. Whether it is institutional buffering, market behavior supremacy, or on-chain consensus, all attempt to build a trustworthy trading order amidst imperfection. In the future, algorithms will continue to expand their territory in the financial world, but value judgments will always exist behind every line of code.