# Find where the time went.

**Racing compares two runs and shows where the time changed, with the evidence beside every explanation.** Today it runs on one synthetic Jerez session in your private workspace.

## What Racing does today

Five steps, all shipped, all running on the synthetic Jerez scenario. None of them needs a telemetry upload.

1. Compare the runs: a median over the laps you keep for each run, plus one representative lap per run for the traces.
2. Exclude a lap you judge unrepresentative, such as B03 through traffic, and the run medians update. The representative traces stay fixed.
3. Inspect speed, throttle and brake through the sector 2 corner window, labelled so an explanation can cite it.
4. Read competing driving and setup hypotheses, then correct the context and request a revised analysis. Earlier results keep their source revision.
5. Save the next test: one change, what to hold constant, the measurement, the success criterion and when to stop.

## Every explanation cites its evidence.

Hypotheses link back to the exact window in the traces. Missing measurements, such as hot tyre pressures in the synthetic session, are listed as missing and never filled with a zero.

Engineering judgment remains yours. A test distinguishes explanations; it does not promise a gain.

## The method behind the comparison

Racing applies the rules a careful engineer applies by hand. [How to compare two stints fairly](https://apexscout.racing/guides/compare-two-stints) covers lap selection and medians. [Driving or setup: separating causes](https://apexscout.racing/guides/driving-or-setup) covers reading the traces.

## Design-partner roadmap, 2027 season

Planned with design-partner teams for the 2027 season. Not available today.

- MoTeC and AiM CSV import, so your own logged sessions enter this same comparison.
- A per-team isolated model trained only on that team's data. No cross-team training and no shared model.
- Team data stays the team's, with export and deletion when a team leaves.