Repository navigation
2026: measuring energy "super-clipping" from public telemetry, plus data-quality issues found along the way #971
Priyanshi507
started this conversation in
Show and tell
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Hi everyone, and thanks to @theOehrly and all contributors for FastF1.
Since ERS data isn't available in the public feed for 2026 (see #861), I tried to measure energy management indirectly. On each circuit's longest straight, I look for laps where speed falls while throttle is at 100% and the brake is off ("flat-out speed loss"), the visible signature of super-clipping or deployment cut-off.
What I found (7 races, Australia to Belgium):
Data-quality issues that may be useful to others:
Only 2,924 of 6,400 candidate laps passed basic physical checks. The main reasons:
For the Belgian GP race, FastF1 itself warns "Car data is incomplete" for every driver.
Without filtering, these produced misleading results. In China, apparent large differences between teammates disappeared after cleaning.
Code, tests and per-lap results: https://github.com/Priyanshi507/ers-lens
(the checks are in
scripts/derate_analysis.py)Questions:
Feedback on the method is very welcome too.

All reactions