# Recipe Paper Strategy Backtest

> Backtest a trading strategy using paper trading against live prices.

- Skill: `yejiming/recipe-paper-strategy-backtest` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add yejiming/recipe-paper-strategy-backtest`
- Raw SKILL.md: https://api.skillmd.com/api/skills/yejiming/recipe-paper-strategy-backtest/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: yejiming (https://skillmd.com/u/yejiming)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/yejiming/recipe-paper-strategy-backtest

---


# Paper Strategy Backtest

> **PREREQUISITE:** Load the following skill to execute this recipe: `kraken-paper-strategy`

Run a strategy through multiple paper sessions to validate consistency before live deployment.

## Important: Paper Results Overstate Live Performance

Paper trading fills at the exact quoted price with no fees, no slippage, and no partial fills. Live trading on Kraken incurs maker/taker fees (0.16%/0.26% at base tier), slippage on market orders, and possible partial fills on limit orders. When comparing session results in step 9 below, subtract at least 0.26% per fill from paper returns to get a more realistic estimate. A strategy that is only marginally profitable on paper is likely unprofitable live.

## Steps

1. Define strategy parameters (pair, entry/exit rules, position size)
2. Initialize paper account: `kraken paper init --balance 10000 -o json 2>/dev/null`
3. Run session 1: execute strategy logic using paper buy/sell commands
4. Record session 1 results: `kraken paper status -o json 2>/dev/null` + `kraken paper history -o json 2>/dev/null`
5. Reset: `kraken paper reset -o json 2>/dev/null`
6. Run session 2 with same parameters at a different time (different market conditions)
7. Record session 2 results
8. Repeat for 3-5 sessions minimum
9. Compare results: win rate, average P&L per trade, max drawdown, Sharpe-like ratio
10. If results are consistent and positive, the strategy is a candidate for live promotion
11. If results vary wildly, adjust parameters and re-test