NegativeEV

NegativeEV

Grades sports bets against thousands of play-by-play simulations, reporting simulated probability, implied probability, and edge.

Category
Visit Server

README

NegativeEV MCP server

Grade a sports bet against thousands of play-by-play game simulations, from inside any MCP client.

claude mcp add --transport http negativeev https://negativeev.com/mcp

No API key, no signup, no OAuth dance. The server is public and free, and the command above is the whole install.

  • Endpoint: https://negativeev.com/mcp
  • Transport: Streamable HTTP (stateless)
  • Auth: optional bearer token
  • Registry: com.negativeev/bet-checker
  • Server card: https://negativeev.com/.well-known/mcp/server-card.json

This repository holds the public documentation and the Claude Code plugin for that server. The server itself runs at negativeev.com and its source is not published here.

What it does

NegativeEV takes a bet, simulates the underlying game or event thousands of times play-by-play with a machine-learning simulation engine, and compares the simulated win rate to the probability implied by the sportsbook's price. What comes back is the simulated probability, the implied probability, and the edge between them.

It is a checker, not a picks service. It never sells picks, never places bets, and never handles money. Users bring a bet they were already thinking about and find out what the price is actually worth. Most of the time the answer is that the price is worse than it looks, which is where the name comes from.

Sports covered today: MLB, WNBA, PGA, and ATP, for events on the current slate that have not started yet. Markets include moneylines, run lines and spreads, totals, player props, and parlays built from them.

Tools

Tool What it does Cost
check_bet Grades a bet written as plain text. Returns each leg's simulated probability, the implied probability from the price, and the edge. Spends one check, and only when the bet fully grades
get_slate Reports whether there is a checkable slate right now, per sport. Free
get_usage Reports checks remaining and when the next one frees up. Free

Every check_bet result leads with a status field, so a caller never has to infer what happened from the numbers:

  • graded — every leg scored. The only status that spends a check.
  • partial — some legs scored.
  • ungraded — the bet was understood, but no leg is covered by the sims.
  • no_bet_found — no bet was recognized in the text.
  • no_slate, unsupported_market, unsupported_sport — self-explanatory.

An ungraded bet was understood correctly. Rewording it and sending it again returns the same answer. Call get_slate instead to see what is checkable.

Install

Claude Code

claude mcp add --transport http negativeev https://negativeev.com/mcp

Claude Desktop

Settings → Connectors → Add custom connector, then paste https://negativeev.com/mcp.

Cursor, VS Code, and other MCP clients

The server is plain streamable HTTP with no auth requirement, so any client that accepts a remote MCP URL works. Point it at https://negativeev.com/mcp.

An install page with per-client instructions is served at https://negativeev.com/mcp in a browser.

Claude Code plugin

The plugin in this repo bundles the server config plus a check-a-bet skill that teaches Claude when to reach for it and how to read the result.

/plugin marketplace add jessejohnsohn/negativeev-mcp
/plugin install negativeev@negativeev

Usage

Ask in plain language. Some prompts that exercise each tool:

  • "Is Shohei Ohtani over 1.5 total bases at -120 a good bet tonight?"
  • "Here's my parlay: Ohtani over 0.5 home runs +280 and Dodgers moneyline -150. Worth it?"
  • "Is there a checkable MLB slate tonight?"
  • "How many bet checks do I have left today?"

Authentication and limits

Anonymous callers share a per-device quota of 100 checks a day. Sending a signed-in user's Supabase access token as Authorization: Bearer raises that to 1000 an hour.

There is no agent self-registration: no dynamic client registration, no client_credentials grant, and no API keys. An agent acts on behalf of a signed-in user. An invalid or expired token is rejected with a 401 rather than silently downgraded to anonymous, so re-authenticate instead of retrying.

Full details: https://negativeev.com/auth.md

About

NegativeEV is built in Portland by Jesse. The simulator is the product — the betting side is just the scoreboard that measures whether the simulations match reality better than the market does.

  • Site: https://negativeev.com
  • How it works: https://negativeev.com/about/how-it-works
  • For AI agents: https://negativeev.com/about/for-agents
  • X: https://x.com/negative__EV

Verdicts are estimates from simulations, not guarantees. Analysis for information, not betting advice. 21+ where required.

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured