BTC Coin-M DCA Ladder Backtest

Overview

A deterministic backtester for coin-margined BTC perpetuals implementing a DCA ladder with explicit perp mechanics (inverse PnL in BTC), funding, fees, and liquidation.

Equity Curve and Drawdown

(Above: Strategy equity vs. BTC price, with liquidation price, average entry, and cycle markers. Log scale for better visualization of growth.)

Data

• BTC-USD daily OHLC (source: , timezone: )
• Funding: constant 10% p.a. unless series provided.

Methodology

• Entry/scale-in via ladder rules (deterministic fills with slippage/commission).
• Leverage cap, maintenance margin, liquidation model.
• IS window: 2020-05-11 → 2021-10-24 (objective end date).
• OOS window: 2024-04-20 → 2024-10-20 (halving + 6m).

How to Run

• Python 3.10+
• pip install -r requirements.txt
• Open MS_backtesting.ipynb and run all cells.

Key Results (Summary)

Window	CAGR	Sharpe	Sortino	MDD	VaR(95%)	ES(95%)
IS	936	3	4.4	-49	-6.7	-10.4
OOS	179	1.5	2.6	-38	-7.1	-9.4

(Insert equity curve and drawdown plot here)

Limitations

• Isolated regime
• Constant funding unless series supplied.
• Daily bars; no intraday path dependency.

Next Steps

• Optional T+1 execution.
• Historical funding feed.