NBA Prop Lab MCP Server
This server exposes NBA/WNBA player-prop analytics, historical performance, backtesting, player research, and system health checks.
Fetch PropLab's track record with league, tier, stat, and date filters.
Get daily dashboards with games, spreads/totals, top scored picks, and rating distribution.
View NBA/WNBA games for a date.
Search players and pull deep player research (logs, matchups, DvP, trends, injuries).
Inspect individual picks by ID.
Run authenticated backtests by summary, day, rating tier, or stat market.
Score a custom prop pick on demand with the 7-block engine (auth required).
Check system status, pipeline runs, data freshness, and errors.
Provides tools for NBA and WNBA player-prop analysis via the nbaproplab.com API: a daily picks board and dashboard, model track record by league/tier/stat/date range, the day's games, player search, plus authenticated backtest summaries (daily, by rating, by stat), pick details, hypothetical pick evaluation, and per-player research.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@NBA Prop Lab MCP Server show me today's NBA player prop picks"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
PropLab MCP Server
Model Context Protocol server for PropLab: NBA and WNBA player-prop analytics. You get the daily board the model publishes before tip-off, its public record graded against the official box score, defense vs position, usage, referees and player research.
There are two ways to use it:
Remote server: works now, nothing to install, no key.
https://nbaproplab.com/mcpLocal stdio package, built from this repository. It adds backtests and on-demand pick evaluation, which need an API key. It is not yet published on npm.
The data is informational sports analysis, not betting advice. PropLab does not take wagers, link to sportsbooks or guarantee outcomes, and past results do not predict future ones.
Remote server (recommended)
Streamable HTTP, stateless, no authentication. Every tool is read-only and returns the same public data shown on nbaproplab.com.
Claude (claude.ai, desktop, mobile): Settings → Connectors → Add custom connector. Name:
PropLab. URL: https://nbaproplab.com/mcp. Authentication: none.
Claude Code:
claude mcp add --transport http proplab https://nbaproplab.com/mcpCodex:
codex mcp add proplab --url https://nbaproplab.com/mcpor in ~/.codex/config.toml:
[mcp_servers.proplab]
url = "https://nbaproplab.com/mcp"Any other client that supports streamable HTTP can point at the same URL.
Tools
Tool | Parameters | Returns |
|
| The published player-prop board for the next slate (player, stat, line, direction, tier, model score) and how the previous board settled |
|
| Settled record by tier and stat. The published record is the Premium tier ( |
|
| Points, rebounds and assists each team allows to guards, forwards and centers relative to the league average, with defensive rating, pace and ranks |
|
| Players ranked by usage rate this season, with last-10-games usage, minutes and points |
| — | Today's NBA games with tip-off times, yesterday's scores and the standings |
|
| NBA and WNBA games on a date (default: today, US Eastern) |
| — | The crew of each NBA game today, once the league publishes it the morning of game day |
| — | NBA referees this season with games worked and fouls per game against the league average |
|
| NBA and WNBA players by name, accents optional ("Jokic" finds Jokić) |
|
| Recent game logs, current prop lines across books with hit rates, injury status, next-game context and the model's picks on that player |
Example prompts:
"Which NBA teams allow the most rebounds to centers, and do any of them play today?"
"How has PropLab's published Premium tier done in the WNBA since April, by stat?"
"Search for Jokic and show his last 10 games and injury status."
Invalid arguments, for example league: "nfl", return a tool error that explains the valid values.
Related MCP server: NBA MCP Server
Local stdio package
Not yet published on npm, so build it from this repository:
git clone https://github.com/Khavel/proplab-mcp.git
cd proplab-mcp
npm install
npm run build # compiles src/index.ts → dist/index.jsThen register it in your MCP client with the absolute path to the build:
{
"mcpServers": {
"proplab": {
"command": "node",
"args": ["/absolute/path/to/proplab-mcp/dist/index.js"],
"env": {
"PROPLAB_API_KEY": "…"
}
}
}
}PROPLAB_API_KEY is optional: without it the public tools still work. For local development
with hot reload, run npm run dev.
Environment variables
Variable | Required for | Description |
| No | API base URL (default: |
| Backtest, pick and research tools | Bearer API key |
The key is minted at nbaproplab.com → Account → API keys. It is shown once at creation and can be revoked from the same screen.
Tools
Public, no key:
Tool | Endpoint | Key params |
|
|
|
|
|
|
|
|
|
|
|
|
|
| — |
With PROPLAB_API_KEY:
Tool | Endpoint | Key params |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Verify locally
npx @modelcontextprotocol/inspector node dist/index.jsThen call proplab_system_status, which needs no key and should return the live system status.
Data and privacy
Remote server: it keeps no state and stores nothing from your conversations. It only receives the arguments of each tool call (a league, a date, a player name or id), and the web server keeps its standard access logs.
Local package: it calls the public nbaproplab.com API over HTTPS and reads the key only from the environment, never logging it. No tool reads database credentials or app settings, or opens a shell.
Inputs are validated with zod and responses are truncated at 40,000 characters. Every tool is read-only except
proplab_evaluate_pick, which scores a hypothetical pick and stores nothing.Privacy policy: nbaproplab.com/legal/privacy. Responsible gambling: nbaproplab.com/responsible-gambling.
Support
Open an issue at github.com/Khavel/proplab-mcp/issues.
Available Tools
12 toolsproplab_backtest_by_ratingGet Backtest By RatingARead-onlyIdempotentInspect
Hit rate and profit broken down by pick rating tier (Elite, Good, Marginal, Weak, Avoid). Requires auth. Use to compare tier quality across leagues.
Inputs: to, from, league.
proplab_backtest_by_statGet Backtest By Stat TypeARead-onlyIdempotentInspect
Hit rate and profit broken down by stat market (Points, Rebounds, Assists, PtsRebAst, etc.). Requires auth. Use to find which markets the model performs best on.
Inputs: to, from, league.
proplab_backtest_dailyGet Backtest DailyARead-onlyIdempotentInspect
Day-by-day backtest results with hits, misses, daily profit, and cumulative profit. Requires auth (PROPLAB_API_KEY). Good for charting profit curves.
Inputs: to, from, league.
proplab_backtest_summaryGet Backtest SummaryARead-onlyIdempotentInspect
Aggregate backtest over PREMIUM (Good+) picks only: total picks, hits, misses, pushes, hit rate, profit, and ROI for a date range. Requires auth (PROPLAB_API_KEY). Supports league filter.
Inputs: to, from, league.
proplab_dashboardGet Daily DashboardARead-onlyIdempotentInspect
Get today's (or a specific date's) dashboard: games with spreads/totals, top 10 scored picks, and rating distribution. Public — no auth required.
Inputs: date, league.
proplab_evaluate_pickEvaluate a PickBInspect
Score a custom pick on-demand through PropLab's 7-block engine. Returns a confidence score (0-100), rating, and block-by-block breakdown. Requires auth.
Inputs: line (required), statType (required), direction (required), playerName (required), +1 more.
proplab_gamesGet GamesARead-onlyIdempotentInspect
Get NBA/WNBA games for a date: teams, spreads, totals, game status. Defaults to today.
Inputs: date.
proplab_pick_detailsGet Pick DetailsARead-onlyIdempotentInspect
Get full details for a specific pick by ID: player, stat, line, direction, score, 7-block breakdown, spider chart data, and settlement result.
Inputs: id (required).
proplab_player_researchGet Player ResearchARead-onlyIdempotentInspect
Deep research data for a player: season averages, recent game logs, matchup history, DvP, trends, and injury status. Use proplab_search_players first to get the playerId.
Inputs: playerId (required).
proplab_search_playersSearch PlayersARead-onlyIdempotentInspect
Search for NBA/WNBA players by name. Returns player ID, full name, team, and position. Use the returned ID with proplab_player_research or proplab_evaluate_pick.
Inputs: q (required).
proplab_system_statusGet System StatusARead-onlyIdempotentInspect
Check PropLab system health: API health status, recent pipeline runs, data freshness, and any errors.
Inputs: no named parameters.
proplab_track_recordGet Track RecordARead-onlyIdempotentInspect
Get PropLab's historical pick performance: overall hit rate, profit, rating breakdown (Elite/Good/Marginal/Weak), daily results, and recent Good+ picks. Without a tier filter the overall/total fields cover ALL tiers (including internal Weak/Avoid picks that are never published) and the premium* fields cover Good+; with a tier filter EVERY headline field describes that tier only. Supports league filter (nba/wnba), tier filter, stat type, and date range. Public — no auth required.
Inputs: to, from, tier, league, +1 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 12 tools
While most tools have distinct purposes, there is notable overlap between proplab_track_record, proplab_backtest_summary, and proplab_backtest_by_rating, which all report historical hit rates and profit, and both track_record and backtest_by_rating provide rating-tier breakdowns. This could confuse agents selecting the right tool, though the descriptions help differentiate scope and auth requirements.
All tools use the consistent `proplab_` prefix and snake_case with clear verb_noun or noun_noun patterns (e.g., proplab_search_players, proplab_backtest_summary). No deviations in convention.
With 12 tools, the set is well-scoped for an NBA prop analytics server, covering games, picks, backtesting, and player research without excessive fragmentation or missing core areas.
The surface covers core workflows: game data, pick details, historical performance, backtesting, custom pick evaluation, and player research. Minor gap: no explicit tool to list all picks for a date (dashboard only shows top 10), but agents can work around via track_record or backtest tools.
Maintenance
Related MCP Connectors
Flash Props API: player-prop analysis, projections, evidence, and line movement over REST/MCP.
Live odds, cross-book +EV and graded player-prop results across 37 books. Hosted endpoint included.
NFL/NBA/MLB/NHL/PGA + DFS and prediction-market data. Browse free; query with a free API key.
NBA MCP — player, team, and game data via the BallDontLie API
Related MCP Servers
- AlicenseNot gradedqualityBmaintenanceEnables querying basketball data including players, games, season averages, and stats from the balldontlie.io API.85 npmMIT
- AlicenseAqualityDmaintenanceEnables MCP-compatible LLM clients to access live NBA data, including stats, standings, rosters, scores, and box scores via callable tools.9MIT
- AlicenseNot gradedqualityBmaintenanceEnables AI agents to query NBA, WNBA, and men's and women's college basketball data through tools for play-by-play, shot charts, stats, odds, props, injuries, and ratings.MIT
- AlicenseAqualityCmaintenanceEnables looking up players, season and career statistics, game logs, schedules, live box scores, division standings and league leaders through the public statsapi.mlb.com API, with no API key required. All ten tools accept natural team and player references, and responses are cached and retried automatically.10MIT