Change one thing.
Prove it worked.
Measure wheel power from GPS. Find unused grip from RaceBox. Two tools on one account, for the drivers and riders who change something and then want to know whether it actually worked.
- Peak wheel power
- 210kW
- At 6050 RPM, 4th gear
- Peak wheel torque
- 403Nm
- Averaged in 100 RPM bins
- Session grip score
- 111
- 100 is one g all the way round
- Peak combined demand
- 1.30g
- Largest vector of the session
Synthetic GPS trace and a synthetic RaceBox session, computed by the shipping analysis pipeline (v4) as this page rendered. Nothing here is a screenshot, and no vehicle was driven to produce it.
Calibrate once.
Then just ride.
Drive a single gear and the phone does the rest. One steady hold teaches it your RPM from speed alone, and from then on every pull comes back as a wheel power curve you can lay over the last one. A cleaner curve tells you more than a peak number, and a noisy trace stays visibly noisy instead of being averaged into something reassuring.
- Calibration
- One steady hold at a known RPM captures rollout in metres per rev: tyre, gearbox and final drive in a single number.
- Capture
- Drivers watch the screen and confirm. Riders pocket the phone, ride, and pick the pulls once they have stopped.
- Output
- Wheel power and torque in 100 RPM bins, plus accel times and the raw trace behind them.
Solid: wheel power. Drawn by the shipping pipeline (v4) from a synthetic 10 Hz GPS trace, in your browser, as this page rendered.
Find the grip you left on track.
Drop in a RaceBox CSV and see the traction circle you actually used, corner by corner, lap after lap. Every headline figure is an absolute score, measured g demand times one hundred, so it stays comparable across laps, sessions, bikes and riders instead of flattering your own best lap back at you.
- Envelope
- The traction boundary you actually reached, per angular bin, from your timed laps only. It describes what you did; it is never used as a divisor.
- Turns
- Apexes are matched to a track turn across every lap, so T4 means the same bend on lap 1 and on lap 9.
- Compare
- Up to six laps on one spatial axis, with the stretch outside their common section hatched rather than guessed.
Dots: one sample of demand. Outline: the envelope you reached.
Session score
111
100 is one g in every direction
Peak combined
1.30g
Largest vector on the session
Per turn
| Turn | Brake g | Apex lat g | Drive g | Score |
|---|---|---|---|---|
| T1 | 1.12 | 1.06 | 0.68 | 113 |
| T2 | 0.63 | 1.30 | 0.82 | 130 |
| T3 | 1.07 | 1.21 | 0.66 | 121 |
| T4 | 0.83 | 1.28 | 0.77 | 128 |
| T5 | 1.21 | 0.93 | 0.59 | 122 |
| T6 | 0.67 | 1.24 | 0.85 | 124 |
| T7 | 0.97 | 1.06 | 0.76 | 110 |
Record. Compare. Decide.
Record
Capture a clean GPS pull, or import a RaceBox session. Riders record the whole ride and choose afterwards, because nothing worth choosing happens while the vehicle is moving.
Compare
Put runs on one RPM axis and laps on one spatial axis. A cleaner curve tells you more than a peak number, and a lap delta says where the time went.
Decide
Keep the change that improved the data. Revisit the one that did not. The comparison is the product; the number on its own is not.
What these measurements are worth.
Stating the limit next to the claim is part of the product, not a disclaimer bolted onto it. Both of these belong on the sheet.
Wheel power
Estimated from GPS acceleration, vehicle mass, gearing and road-load assumptions. It is not a replacement for a calibrated rolling-road dyno. Read it as a comparative measurement: the same vehicle, the same road, before and after one change.
Grip
Reflects GPS and IMU quality, tyre condition, rider inputs and the settings you choose. It does not measure the tyre's absolute limit. The traction envelope describes the boundary you reached, and is never used as a divisor.
Make the next change count.
Change one variable, keep the evidence, and make the next decision against a baseline you can defend. Signing up takes an email address. The demo takes nothing at all.