Sports Probabilities MCP
io.github.commodus67Published by GitHub account commodus67v0.0.12Updated Oct 24, 2026
Monte Carlo NFL, MLB, NHL, NBA and soccer odds, plus model-vs-market edge from live prices
Overview
Remote MCP server that returns Monte Carlo season probabilities for NFL, MLB, NHL, NBA and 17 soccer leagues, plus model-versus-market edges against live…
- What you can do
- Get one team's full probability set, record, projected finish and ranking
- Rank a whole league by probability rather than points or record
- Compare model probabilities with live Kalshi prices, including edge, expected value after fees and a capped quarter-Kelly stake
- Run a fresh simulation under custom assumptions with movement analysis
- Review how a team's probability moved across the season from archived runs
- When to use it
- Useful when you want computed playoff, title, qualification or relegation odds for a specific team or league instead of a model's guess, or when you want to see where a simulation disagrees with live prediction-market prices.
- Requirements
- A remote streamable HTTP endpoint; no local install. An Apify account: OAuth-capable clients sign in, others send an Apify API token as a bearer token in the Authorization header. Usage is billed through Apify per probability read.
Installation
In SourceWeft
- Open Sports Probabilities MCP in the dashboard and add it to a workspace.
- Connect Authorization token credentials. They are stored privately and never shown on public pages.
- Enable the server for the chats that should use its tools.
Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.
Use with an agent
Paste this prompt into Claude Code, Codex, Cursor or another agent — or into SourceWeft chat. The agent reads this server's install guide, shows you its publisher, endpoint and sign-in, and adds it to your client once you agree. Keys never go into a command.
Read https://sourceweft.com/mcp/io.github.commodus67%2Fsports-probabilities-mcp/install.md and connect the MCP server it describes to my client. Before changing any config, show me its publisher, endpoint and sign-in method and wait for my OK. Never put an API key in a command or in a file I would commit; ask me where it should come from.Other MCP clients
Add this to your client's mcpServers config.
{
"mcpServers": {
"sports-probabilities": {
"type": "http",
"url": "https://commodus67--sports-probabilities-mcp.apify.actor/mcp",
"headers": {
"Authorization": "Bearer <YOUR_TOKEN>"
}
}
}
}README
Sports Probabilities MCP
An MCP server that gives AI assistants Monte Carlo probabilities for the NFL, MLB, NHL, NBA and 17 soccer leagues, plus a model-versus-market comparison against live Kalshi prices.
Ask in plain language from Claude, ChatGPT or Cursor and get a computed number back — not an estimate the model made up.
The server runs on Apify: https://apify.com/commodus67/sports-probabilities-mcp
This repository holds the documentation and the registry manifest (server.json). The implementation lives in the Actor above.
Coverage
Soccer leagues: Premier League, EFL Championship, LaLiga, Serie A, Bundesliga, Ligue 1, Eredivisie, Primeira Liga, Süper Lig, Belgian Pro League, Scottish Premiership, Brasileirão, Liga MX, Liga Profesional (Argentina), Primera A (Colombia), MLS, Austrian Bundesliga. Other ESPN soccer slugs are supported.
Tools
Model vs market
compare_model_vs_market reads live Kalshi playoff contracts for all four North American leagues — KXNFLPLAYOFF, KXMLBPLAYOFFS, KXNHLPLAYOFF and KXNBAPLAYOFF — and returns the gap between model and market, the better side of each contract, expected value after the taker fee and a capped quarter-Kelly stake.
For the NHL and NBA, value calls and stakes stay off until teams have played ten games, as in the underlying Actors; edges are still reported in full. For soccer, or to compare against a bookmaker, pass your own prices in marketProbabilities.
Connect
Streamable HTTP transport, stateless. Endpoint:
Clients that support OAuth (Claude custom connectors, for example) sign in with an Apify account — no token to paste. Other clients send an Apify API token as a bearer token:
How it works
Each answer comes from one of five published Monte Carlo Actors — NFL, MLB, NHL, NBA and soccer — that take live standings and the remaining schedule, play out the rest of the season thousands of times, and count how often each outcome happens. The MCP server calls those Actors and caches the result — it does not reimplement the simulation, so an answer here never disagrees with the same query made directly against the API.
Every response carries the date and the source of the run it came from.
Pricing
Pay per event, billed through Apify. From $50.00 per 1,000 probability reads. Full pricing on the Actor page.
Note
This is statistics and simulation. It is not betting advice, and it does not produce picks.
Source: README.md at commit f6b2e52
Tools
0Version history
1- v0.0.12LatestSep 16, 2026
Score
75% of the applicable checks met
How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.
- Reached in the last 30 daysRequired25 pts
- Auth declaredRequired10 pts
- Tools listed15 pts
Answer tools/list without a credential, or ship a package that starts in a sandbox without one.
- DescriptionRequired15 pts
- Logo10 pts
- Documentation15 pts
- License10 pts
Declare a license in the repository or the registry entry.
