Formula 1 Data Analysis with AI: Ask F1 Questions in Plain English

F1 produces more data per race than almost any other sport on the planet. Lap times measured to the millisecond. Tyre compounds and stint lengths. Sector splits. GPS coordinates sampled ten times a second. Weather logged throughout every session.

The data is everywhere. The friction is in the questions — the ones that require joining telemetry to lap times to weather data just to get a single answer. Until now that meant Python, SQL, and enough patience to wrangle nested DataFrames before you could even start. We thought there was a better way — so we built F1 Analyst: an open-source Formula 1 analytics tool that lets anyone — fan, analyst, or developer — explore F1 data through plain-English questions, no SQL required.

What Is F1 Analyst?

F1 Analyst is an open-source Formula 1 data analysis tool that connects your AI agent to real F1 timing data so you can ask questions in plain English and get answers instantly — no SQL, no Python, no data wrangling.

It connects FastF1 — the community library that pulls official F1 timing data — to PlyDB, so your AI agent can query laps, results, telemetry, tyre strategy, and weather data in real time, in response to plain-English questions.

The idea is simple: you ask, the AI figures out the SQL, PlyDB runs it, you get an answer. No pipelines. No cloud. No data wrangling.

It covers every race, qualifying session, and practice session from 2018 onwards.

Who is this for? F1 fans who want deeper analysis than a race summary, data-curious engineers who want a working example of AI-powered sports analytics, and anyone who has ever wanted to answer a telemetry question without writing a script.

What Makes This Fun

The real joy is the questions you’d never bother with if answering them required writing code.

Who’s on championship pace right now? Find every season where the eventual champion sat in their current points position after the same number of rounds. History doesn’t repeat exactly, but the early-season signal is stronger than most people think — and the exceptions are just as interesting as the pattern.

New regulations, new pecking order? The 2026 season brought the most sweeping rule changes in years. Ask the AI to rank every previous regulation-change season by how dramatically the constructor standings shifted from the prior year, then track whether the gaps between teams narrowed or widened as those seasons progressed. Which team profile tends to close the gap fastest once development gets going?

How is your driver settling in? Several drivers switched teams over the winter. Ask the AI to pull their lap time deltas and qualifying gaps at circuits they’ve visited before, across their previous teams versus their current one. Already at their historical baseline? Still finding their feet? The data will tell you.

Did the fastest car always win the championship? Compare each constructor’s average qualifying position — a clean proxy for raw pace — against their final points standing, season by season. In some years the gap is stark. In others, strategy, reliability, and driver talent tell a completely different story.

Which circuits punish qualifying pace the most? Look at the delta between a driver’s qualifying position and their finishing position, across every race at every track. Some venues consistently shuffle the order; others don’t. Which ones, and why?

What’s the real cost of a safety car? Pit stop windows compress. Strategy calls get forced. Ask the AI to find every safety car period in a season and compare the finishing order to where drivers would have ended up on pure pace.

Is there a “tyre cliff” in the data? Ask the AI to plot lap time degradation by compound and stint length across different circuits. When does the cliff actually arrive, and does it show up in the numbers before teams react to it?

Rain and chaos: Which sessions had rainfall, and by how much did lap times swing? Which drivers consistently perform above their season average in mixed conditions?

Telemetry deep dives: At Monaco, what percentage of the lap are drivers at full throttle? Compare that to Monza. Ask your agent to find the corner where VER carries the most minimum speed — then ask the same question about Hamilton at his peak.

The data supports all of it. The AI does the legwork.

Try It Yourself

The repo has everything you need: a data download script, a pre-configured PlyDB setup, and a semantic overlay so your agent understands F1-specific encodings and table relationships out of the box.

Setup takes a few minutes. The rabbit holes take considerably longer.

Frequently Asked Questions

Do I need to know SQL to use F1 Analyst? No. Your AI agent handles all the SQL. You ask questions in plain English; the agent translates them into queries, runs them via PlyDB, and returns the answer.

What F1 data is included? F1 Analyst pulls from FastF1, which provides official F1 timing and telemetry data: lap times, sector splits, tyre compounds and stint lengths, pit stop windows, car telemetry (speed, throttle, brake, gear, RPM, DRS at ~240 Hz), circuit position data (X/Y/Z coordinates), weather conditions sampled throughout each session, session results (finishing positions, qualifying times, championship points, retirement status), and the race calendar including sprint weekend formats. It covers every race, qualifying session, and practice session from 2018 onwards.

What AI agent do I use with it? Any PlyDB-compatible agent works. The repo is set up for Claude Code, but the PlyDB config and semantic overlay are compatible with any agent that supports tool use.

Does it work without a cloud database or data warehouse? Yes. Data is queried locally via PlyDB — no cloud account, no ETL pipeline, no ongoing infrastructure required.

Is there a similar project for other sports? Yes — check out Baseball Analyst for MLB data, Olympics Analyst for the 2026 Winter Games, and Oscars Analyst for 97 years of Academy Awards data.

Did you know? PlyDB can connect your AI to boring data too!

Whether it’s business data in a dusty Excel sheet or a complex DevOps log in S3, AI can be surprisingly good at making sense of a mess. PlyDB acts as the bridge, letting your AI query across Postgres, MySQL, CSV, Excel, Parquet, Google Sheets, and more — locally or in the cloud.

Open source and free. Give PlyDB a spin!