The MCP Sessions Vol. 1: Sports Analytics
2026/01/13TL;DR: Watch Claude analyze NFL playoff simulations and PGA Tour statistics in real-time using the MotherDuck MCP connector—demonstrating how natural language queries can replace complex SQL for sports analytics.
What is the MotherDuck MCP?
The MotherDuck MCP (Model Context Protocol) connector enables Claude to:
- Query MotherDuck databases directly
- Write and iterate on SQL automatically
- Self-correct when queries fail
- Generate visualizations from results
The connector is now on Anthropic's approved marketplace—add it to Claude Desktop in Settings → Capabilities.
Demo 1: NFL Playoff Analysis
Jacob runs Monte Carlo simulations for NFL seasons. The demo compares two database snapshots:
mds_box_copy: End of regular seasoncoffee_wc: After Wild Card weekend
Natural Language Prompting
The analyst simply describes what they want to analyze in plain English, and Claude automatically:
- Lists tables in both databases
- Identifies 6 new games added
- Calculates Super Bowl odds changes
- Attributes probability shifts to specific game outcomes
Key Findings
- Seahawks dropped from 27% → 25% Super Bowl odds (bracket got harder as favorites won)
- Texans gained odds after beating the Steelers
- When a team loses, their Super Bowl odds "go back to the pool" for redistribution
SQL Self-Correction
When Claude writes incorrect SQL and gets an error, it automatically checks the schema, identifies the issue, and rewrites the query correctly.
Demo 2: PGA Tour Statistics
Logan analyzes golf statistics to find undervalued players and validate Scottie Scheffler's dominance.
Finding Correlations to Earnings
Asking "Which stat categories correlate most to money list rankings?" reveals that Shots Gained Tee-to-Green vastly outweighs Shots Gained Putting for predicting earnings.
Identifying Dark Horse Players
Asking "Who played well statistically in 2025 but underperformed on money list?" surfaces players like Pearson Cudi and Pureborn Olison:
- Ranked 124th and 107th in earnings
- But ranked 28th and 20th in Shots Gained
- Watch list candidates for breakout seasons
Validating Historic Dominance
Comparing Scheffler's 2025 to the best seasons from 2017-2022 shows his season was statistically better than Rory's 2019 and Morikawa's 2021. Even accounting for the weaker field post-LIV, his gap to #2 was unprecedented.
Visualization with Invis
The demo uses a custom Claude skill called Invis for charting:
- Generates HTML artifacts from markdown + JSON specs
- Creates interactive dashboards
- Supports iterative refinement ("make this chart taller")
Tips for MCP Analytics
- Hint at tools: Include "MotherDuck MCP" in prompts to guide Claude
- SQL is ideal for AI: Verbose but well-validated, with great error messages
- Iterate visually: Ask for different visualization styles (Tufte-inspired, etc.)
- Compare time periods: Load historical snapshots to validate trends
Related Videos
2026-01-27
Preparing Your Data Warehouse for AI: Let Your Agents Cook
Jacob and Jerel from MotherDuck showcase practical ways to optimize your data warehouse for AI-powered SQL generation. Through rigorous testing with the Bird benchmark, they demonstrate that text-to-SQL accuracy can jump from 30% to 74% by enriching your database with the right metadata.
AI, ML and LLMs
SQL
MotherDuck Features
Stream
Tutorial
2026-01-21
The MCP Sessions - Vol 2: Supply Chain Analytics
Jacob and Alex from MotherDuck query data using the MotherDuck MCP. Watch as they analyze 180,000 rows of shipment data through conversational AI, uncovering late delivery patterns, profitability insights, and operational trends with no SQL required!
Stream
AI, ML and LLMs
MotherDuck Features
SQL
BI & Visualization
Tutorial
