Cross-season comparison

[joshuachuah]

Would it be good to implement a feature where we're able to compare the performance of the drivers/teams to the previous seasons?

[palomacdev]

Hi @br-g and @joshuachuah!

I'm interested in working on this feature. I've been exploring the OpenF1 API and think cross-season comparison would be a valuable addition.

Proposed Implementation

I'm considering implementing this through new API endpoints that would allow comparing:

1. Driver Performance Across Seasons

   • Endpoint: GET /v1/comparison/driver/{driver_number}?years=2023,2024
   • Returns: Average lap times, points, positions, etc. for each season

2. Team Performance Across Seasons

   • Endpoint: GET /v1/comparison/team/{team_name}?years=2022,2023,2024
   • Returns: Constructor points, wins, podiums per season

3. Circuit Comparison

   • Endpoint: GET /v1/comparison/circuit/{circuit_key}?years=2023,2024
   • Returns: Lap records, race statistics across seasons

Technical Approach

• Create new comparison service module
• Aggregate data from existing endpoints
• Return normalized data for easy comparison
• Include proper error handling for missing seasons
• Add comprehensive tests

Questions

Before I start working on this:

1. Is anyone already working on this feature?
2. Do you prefer this approach or would you rather extend existing endpoints with comparison parameters?
3. Are there specific metrics you'd like to prioritize for comparison?
4. Should this handle partial season data (e.g., mid-season comparisons)?

I'd appreciate any feedback before I dive into the implementation!

Thanks for maintaining this awesome project!

[joshuachuah]

Hi @palomacdev !

This is a really nice idea and I agree that cross-comparison would be a valuable addition to the API.

To your questions:

1. As far as I'm aware, no one is working on this feature right now.
2. The /v1/comparison/* approach makes sense and feels cleaner.
3. For metrics, it's better to start with the core metrics then expanding later.
4. Handling partial season data would be great, as long as its clearly indicated in the response metadata.

I would love to work on this with you!

[soundsessential]

I mentioned this to @br-g last year. You will probably need to solve for driver number changes before you could implement this idea. With Max being driver number 1 last year and Lando now taking that number for 2026, comparisons like this won't work? Drivers should have a unique identifier that never changes, in addition to their driver number attribute. And then most API calls should be using the driver ID. Hope that makes sense.

[palomacdev]

Hey @soundsessential thanks for sharing! @joshuachuah will be a pleasure to work with you on this. Let's talk and discuss this issue.

[br-g]

Hi! Thanks for the thoughtful proposal 👍

That said, this kind of aggregation and comparison logic is currently out of scope for the OpenF1 API, which intentionally focuses on providing raw, unopinionated data rather than analysis or derived metrics. Adding comparison endpoints would significantly increase complexity and maintenance.

This would be a great fit for a separate project built on top of OpenF1, though. If you decide to go that route and build something solid, I’d be happy to reference or link to it as part of the OpenF1 ecosystem.

Thanks again for the interest and for checking before starting the implementation!