Olympus Bets Analytics MCP server
{
"mcpServers": {
"olympus-bets": {
"type": "http",
"url": "https://app.olympus-bets.com/mcp"
}
}
}{
"mcpServers": {
"olympus-bets": {
"type": "http",
"url": "https://app.olympus-bets.com/mcp"
}
}
}{
"mcpServers": {
"olympus-bets": {
"type": "http",
"url": "https://app.olympus-bets.com/mcp"
}
}
}{
"servers": {
"olympus-bets": {
"type": "http",
"url": "https://app.olympus-bets.com/mcp"
}
}
}{
"mcpServers": {
"olympus-bets": {
"type": "http",
"url": "https://app.olympus-bets.com/mcp"
}
}
}file .mcp.json (in your project root)file claude_desktop_config.jsonfile ~/.cursor/mcp.jsonfile .vscode/mcp.jsonfile ~/.codeium/windsurf/mcp_config.jsonversion 1.3.0
What it does
Read-only MCP server for Olympus Bets Analytics, a quantitative sports platform covering 12 leagues. Exposes 19 tools for accessing projections, track records, methodology, player/team profiles, and model performance metrics. Intended for developers building AI agents that need sports analytics data.
- Retrieve today's published picks and edge ratings across 12 sports leagues to inform betting decisions or research.
- Query historical pick performance and model accuracy metrics to evaluate prediction reliability over time.
- Access player and team analytics profiles, league schedules, and model vs. market comparison data for sports analysis.
Tools this server exposes
| Tool | Does |
|---|---|
| get_todays_projections | Returns published free slate with picks, edge ratings, and unit recommendations. |
| get_performance_summary | Returns track record aggregated by tier and league. |
| get_track_record | Returns filtered history of resolved picks. |
| get_methodology | Returns pipeline description, formulas, and research findings. |
| get_engine_versions | Returns per-league simulation engine version information. |
| get_model_vs_market | Returns model skill metrics (Brier score, CLV-style) compared to market. |
| get_league_schedule | Returns league slate and public model probabilities. |
| get_game_recommendation | Returns public model projection for a specific game, with premium edges masked. |
| get_pick_history | Returns legacy history view (deprecated; use get_track_record). |
| get_brand_card | Returns canonical brand and citation metadata. |
| get_subscription_options | Returns subscription plans and checkout links. |
| get_data_status | Returns per-league data freshness and coverage status. |
| search_entities | Resolves team and player names. |
| get_team_profile | Returns public team analytics. |
| get_player_profile | Returns public player analytics. |
| get_premium_slate | Returns full premium projection slate for today (requires Connect or Pro tier). |
| get_premium_game_recommendation | Returns premium matchup detail and analysis (requires Connect or Pro tier). |
| get_premium_history | Returns 90-day resolved premium pick history (requires Pro tier). |
| get_projection_history | Returns normalized projection archive including model-only projections (requires Pro tier). |
Before you enable it
An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.