The Hockey Brain
AI & Claude5 min read

What is MCP, and why it matters for hockey analytics

The open protocol that lets an AI safely query your data — explained in one diagram.

Client

AI (e.g. Claude)

  • Plans the task
  • Requests a tool
Protocol

MCP

  • Standard interface
  • Tools + resources
Server

Your MCP server

  • Read-only SQL
  • Your gold tables
Result

Grounded answer

  • Real numbers
  • Traceable query
MCP is the standard socket between an AI client and your data tools.

The 30-second version

  • MCP is a standard way to expose tools and data to an AI model.
  • Instead of pasting numbers into chat, the model calls your MCP server for real data.
  • It turns 'chatting about hockey' into 'querying your hockey database'.

The problem MCP solves

Large language models are great at reasoning but have no built-in access to your data. The naive fix — pasting rows of stats into the chat — is error-prone and does not scale. MCP (the Model Context Protocol) is an open standard that lets a model discover and call tools, so it can fetch exactly the data it needs instead of guessing.

How it works

Think of MCP as a universal socket. Your side runs an MCP server that exposes tools (for example, a read-only SQL function over your gold tables) and resources. The AI client speaks the same protocol, so it can list the available tools, call one, and read the structured result — all without custom glue for every model.

Why it matters for analytics

With MCP, "ask a question about our team" becomes a real database query the model runs against your governed data, returning grounded, traceable answers. You keep control: expose only read access, scope the tables, and log every call. It is the safe, standard way to let AI sit on top of the stack you already built.

Build it yourself

  1. 1Expose your gold tables through a small read-only SQL tool.
  2. 2Wrap it as an MCP server so any MCP-capable client can call it.
  3. 3Connect your AI client and grant only the read tools it needs.
  4. 4Log every query the model runs for auditability.

Key terms

Expected goalsxG

Newsletter

Weekly hockey analytics

Data-driven takes on performance, scouting, and team strategy. No fluff — just the numbers that matter.

No spam. Unsubscribe anytime.

Related explainers