CLMM Vault策略回测 (专业版) v5

Explanation of the CLMM (Concentrated Liquidity - Market Maker) Strategy Backtesting Model Developed for the Sui Chain Vaults Protocol

Why Are We Doing This?
Conducting strategy backtesting is a crucial step for us to make data-driven decisions, validate the feasibility of strategies, and manage potential risks before committing real funds and significant development resources. A strategy that appears to have a high APY may perform entirely differently once real-world frictional costs (such as rebalancing fees and slippage) are deducted. The goal of this backtesting model is to quickly and cost-effectively identify which strategy parameter combinations have the potential to be profitable and which ones pose risks before formal development, thereby avoiding significant losses and providing data support for the project's direction.

Core Features of the Backtesting Model
We have built a "pro version" (v5) strategy simulator using TradingView's Pine Script. It can quickly simulate the core performance of our auto-compounding and rebalancing Vaults on historical price data, with the following main features:
Auto-Compounding: Continuously adds the generated fee income to the principal based on the set profit range (e.g., 0.01%).
Auto-Rebalancing: Simulates automatic rebalancing actions when the price exceeds the preset profit range and deducts the corresponding costs.
Smart Filtering Mechanism: To make the simulation closer to our ideal "smart" decision-making, it integrates three freely combinable filtering mechanisms:
Buffer Zone: Tolerates minor and temporary breaches of the profit range to avoid unnecessary rebalancing.
Breakout Confirmation: Requires the price to be in the trigger zone for N consecutive candles to confirm a breakout, filtering out market noise from "false breakouts."
Time Cooldown: Enforces a minimum time interval between two rebalances to prevent value-destroying high-frequency trading in extreme market conditions.
Important: Simplifications and Assumptions of the Model

To quickly prototype and iterate on the TradingView platform, we have made some key simplifications to the model.
A fully accurate backtest would require a deep simulation of on-chain liquidity pools (Pool Pair), calculating the price impact (Slippage) and impermanent loss (IL) caused by each rebalance on the pool. Since TradingView cannot access real-time on-chain liquidity data, we have made the following simplifications:
Simplified Rebalancing Costs: Instead of simulating real transaction slippage, we use a unified input parameter of single rebalance cost (%) to "bundle" and approximate the total of Gas fees, slippage, and realized impermanent loss.
Simplified Fee Income: Instead of calculating fees based on real-time trading volume, we directly input an average fee annualized return (%) as the core income assumption for our strategy.

How to Use and Test
Team members can load this script and test different strategies by adjusting the input parameters on the panel. The most critical parameters include: position profit range, average fee annualized return, single rebalance cost, and the switches and corresponding values of the above three smart filters.