← Back to list
kelly-criterion Checked
Kelly criterion optimal sizing with fractional variants, edge estimation, and practical applicati…
8.3Overall rating
Installation
🤖 Install via AI
Copy the prompt below and send it to your AI assistant (e.g. Claude Code) — it will follow the instructions and install automatically.
Install the "kelly-criterion" skill by following the instructions at https://skill123.me/install/kelly-criterion.
⌨️ Command line install
Run in your terminal — downloads and installs to ~/.claude/skills/.
curl -fsSL https://skill123.me/install/kelly-criterion.sh | bash
📦 Download ZIP
Download the zip and extract it into your skills directory (e.g. ~/.claude/skills/), then restart your session.
⬇ Download v1.0.0 · 20 KBAbout this skill
Overview
Kelly criterion optimal sizing with fractional variants, edge estimation, and practical application for crypto trading
Source
- Repo: https://github.com/agiprolabs/claude-trading-skills
- Path: skills/kelly-criterion
Score breakdown
Trigger
6.0
Description states What (optimal sizing, fractional variants, edge estimation) and domain (crypto) with good keywords; no Use-when, no boundary, no trigger variants.
Structure
8.7
265-line SKILL.md with two well-scoped references (derivation, practical estimation) and two scripts; clean separation and annotation.
Workflow
8.4
Excellent discipline for a sizing skill: explicit 'never full Kelly' doctrine, sample-size gates before trusting estimates, confidence-interval conservatism, correlation adjustment, cost subtraction from edge, and policy-cap precedence over Kelly output; kelly_from_trades provides rolling stability analysis as a verification loop; scripts run offline with demo data.
Content
9.0
Dense quantitative exposition with tables everywhere; minor typos ('overbets' twice) but high actionability.
Engineering
10.0
Frontmatter valid; shebangs, dependencies, env vars documented; all referenced paths exist.
Security
9.0
6.0 manifest: fully offline — no network, no file writes, no credentials, no persistence. Capabilities match declared scope. No injection or keys. Minor: no not-financial-advice disclaimer for bet-sizing recommendations.
Rigorous Kelly-criterion treatment: formula variants, quick-reference fraction table, why-fractional-Kelly analysis (estimation error, drawdown scaling, over/under-betting asymmetry), minimum sample-size gates (50/100/200 trades), correlation adjustment, assumption-vs-reality mitigation table, and continuous Kelly. Two offline calculators including bootstrap confidence intervals and rolling stability analysis. Deductions: description lacks When/When-Not; no financial disclaimer despite bet-sizing output; typographical errors ('overbets').