subset-performance

Compute hit rate and ROI across betting strategy subsets over defined periods.

1|Updated May 24, 2025
One-click install
npx skills add https://github.com/najicham/nba-stats-scraper --skill subset-performance
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subset-performance
Source: https://github.com/najicham/nba-stats-scraper/tree/main/.claude/skills/subset-performance
Command: npx skills add https://github.com/najicham/nba-stats-scraper --skill subset-performance

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Compare performance across all defined dynamic subsets to identify which betting strategies work best.

Core Features & Use Cases

  • Compute ROI and hit-rate across subsets to rank strategy performance over a specified period.
  • Filter and compare subsets by pattern, system_id, edge, confidence, and ranking to guide decision-making.
  • Use Case: Analysts can evaluate which dynamic subset yields the highest ROI in the last 7-30 days and adjust betting strategies accordingly.

Quick Start

Invoke the subset-performance skill to generate a last 7 days performance snapshot across all subsets. You can also run:

  • /subset-performance --period 14
  • /subset-performance --period 30
  • /subset-performance --subset v9_high*

Frequently Asked Questions about subset-performance

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I calculate ROI and hit rate for NBA betting predictions over a specific period?

You can compare betting strategy performance by computing cross-subset metrics over defined periods like the last 7 to 30 days. It subsetting by pattern, system_id, edge, confidence, and ranking to identify which dynamic subset yields the highest ROI and hit rate.

How do I compare betting strategy performance across different subsets?

You can compare betting strategy performance by computing cross-subset metrics over defined periods like the last 7 to 30 days. It subsetting by pattern, system_id, edge, confidence, and ranking to identify which dynamic subset yields the highest ROI and hit rate.

Can I filter NBA analytics data by pattern, system_id, and confidence to evaluate picks?

Yes, you can generate a performance snapshot for the last 14 or 30 days by applying a period filter. This excludes pushes and today's incomplete data to ensure your hit rate and ROI calculations accurately reflect settled predictions.

How do I exclude pushes and today's incomplete data when analyzing prediction performance?

Performance analysis is limited to historical periods and filters by pattern, system_id, edge, confidence, and ranking. It focuses strictly on computing hit rate and ROI for settled data, excluding pushes and incomplete daily entries.

What is the best way to benchmark dynamic subsets to optimize NBA picks?

You need access to nba_predictions and nba_analytics datasets to compute cross-subset metrics. These datasets provide the historical foundation required to apply period filters and calculate accurate hit rate and ROI.