How to design and read backtests honestly — lookahead/leakage traps, point-in-time discipline, multiple comparisons, CLV vs ROI, sample size.
EnglishVerified bookmaker entry points (competition/event ids) for cross-book price lookups, starting with AFL.
EnglishWorked example of modeldevelopment for match-winner (h2h/moneyline) markets — ratings + logistic regression, home advantage, draw handling.
EnglishWorked example of modeldevelopment for totals (over/under) markets — scoring-process baseline, pace adjustment, holdout evaluation.
EnglishHow to evaluate and fix probability calibration — holdout discipline, Brier/log-loss reading, Platt-style rescaling.
EnglishCross-bookmaker comparison workflow — find the same selection at every book, surface the best price, and flag value.
EnglishDFS lineup workflow — site rules first, projections with stated sources, deterministic optimisation via optimizelineup, stacking/ownership as explicit judgment.
EnglishThe general model-building method for ANY market — framing, sample-size discipline, feature selection (user priors + data-driven), baselines, leakage-safe validation, calibration, persistence.
EnglishWorking definitions of the core quant concepts — Brier, log-loss, calibration, logistic regression, gradient boosting/XGBoost, regularization, walk-forward CV, CLV — and when each tool fits.
EnglishHow to read a racecard honestly — barrier/weight/form/scratchings, converting prices to fair probability (overround/vig), cross-book best price, and exotics.
EnglishInterpreting captured line movement — steam vs drift, why the close is the benchmark, and how movement context changes a value report.
EnglishRemove the bookmaker margin (vig/overround) to estimate fair probabilities and spot value.
English
AGENT SKILLS ATLASIncludes Backtest Design, Book Navigation, Build a H2h Model, and other workflows.
Claude Code
Codex
Cursor
GitHub Copilot
Gemini CLISupported by the Atlas install prompt. Skill-specific requirements vary.
Pack stats
What’s included
Backtest Design
How to design and read backtests honestly — lookahead/leakage traps, point-in-time discipline, multiple comparisons, CLV vs ROI, sample size.
Book Navigation
Verified bookmaker entry points (competition/event ids) for cross-book price lookups, starting with AFL.
Build a H2h Model
Worked example of modeldevelopment for match-winner (h2h/moneyline) markets — ratings + logistic regression, home advantage, draw handling.
Build a Totals Model
Worked example of modeldevelopment for totals (over/under) markets — scoring-process baseline, pace adjustment, holdout evaluation.
Calibrate Probabilities
How to evaluate and fix probability calibration — holdout discipline, Brier/log-loss reading, Platt-style rescaling.
Compare Odds
Cross-bookmaker comparison workflow — find the same selection at every book, surface the best price, and flag value.
Dfs Lineup Building
DFS lineup workflow — site rules first, projections with stated sources, deterministic optimisation via optimizelineup, stacking/ownership as explicit judgment.
Model Development
The general model-building method for ANY market — framing, sample-size discipline, feature selection (user priors + data-driven), baselines, leakage-safe validation, calibration, persistence.
Quant Concepts
Working definitions of the core quant concepts — Brier, log-loss, calibration, logistic regression, gradient boosting/XGBoost, regularization, walk-forward CV, CLV — and when each tool fits.
Reading a Racecard
How to read a racecard honestly — barrier/weight/form/scratchings, converting prices to fair probability (overround/vig), cross-book best price, and exotics.
Reading Line Movement
Interpreting captured line movement — steam vs drift, why the close is the benchmark, and how movement context changes a value report.
Vig Removal
Remove the bookmaker margin (vig/overround) to estimate fair probabilities and spot value.
Quick start
- Choose your skills. Select individual skills above, or choose the entire pack.
- Copy the install prompt. Use Copy install prompt above the list.
- Send it to your agent. Paste into your coding agent’s chat. Your agent reviews the source and guides the setup.
Try it with your agent
After installation, start with a request like:
Use the Backtest Design skill to help with my task.
Ask me for the context you need before getting started.Requirements & license
- A coding agent that can access your project files and install skills.
- Your agent checks each skill’s instructions, dependencies, and permissions during setup.
- Reported license for the skills shown: MIT. The author’s terms apply.