The pacing platform for coaches. Your athlete's physiology, the real course and race-day weather become a watt-by-watt plan they can actually hold. Physics, not guesswork, in minutes instead of a spreadsheet afternoon.
You built pacing on watts, wind and stopwatch. We wired up the peer-reviewed models that describe the same thing, and made them run in your browser. No proprietary black box. Just the physics your PhD colleagues cite.
External work from wheels, gravity, air. The equation every serious pacing tool starts from.
CP, W′ and Pmax. Not just FTP. Physiology the way it actually behaves at threshold and above.
Continuous integral of the reserve. Not a lap average. Not a guess. Second by second, until the line.
Hourly wind, temperature and air density at the athlete's ETA on race day. Not the day-of average. The segment-of.
The rider you know: weight, CdA, Crr and the 3-parameter CP/W′ profile you've been measuring for months. One profile, every race.
Drop the GPX or pick the velodrome. We resolve gradient, surface and turn radius, then layer race-day wind and air density on top.
You call the finish state: 5%, 10%, empty. Dynamic programming over W′ spends exactly enough to hit the line where you want. Your call, resolved to the watt.
Every rider carries one number that decides sprint performance: the cadence where their power peaks. We fit the force-velocity curve from ergo tests or in-session power-cadence data, and read the peak straight off the parabola. That cadence, multiplied by the right gear, is the plan.
You picked 54 × 14 because it felt right on the last block. We rank the whole matrix by expected time to the line, so you can defend the call in front of the athlete, the federation, and yourself.
Gear ratio is a joint optimum of terminal velocity, target cadence and the rider's efficiency curve. Guessing costs seconds. Trusting one gut call over another costs medals. We evaluate every legal combination and show you where your call sits.
Track events use the Spicer-Kidd legal matrix. Road TTs use the athlete's installed drivetrain. Either way, the number under your gear is the seconds it costs (or the seconds it saves) versus the optimum.
Spicer-Kidd rank · cadence-constrained · terminal-velocity solvedYou know the athlete. You know the course. Air density is the third input. It moves ±5% between a hot lowland race and a cool altitude one. That's 15–30 W. We resolve it hour by hour, so the plan matches the air your athlete will actually push through.
Aerodynamic power scales linearly with ρ. A rider holding 320 W at 1.22 kg/m³ (sea level, 25°C) holds the same physiological output at 290 W in 0.95 kg/m³ (2500 m, 15°C) for the same terminal velocity. Miss it, and you either burn W′ you didn't need to, or under-pace a race you had in the bag.
We pull hourly ρ from Open-Meteo for the exact race latitude, longitude and elevation profile, then feed it straight into the pacing solver. The plan you hand your athlete matches the air on race day, not the annual average.
ρ = f(altitude, temperature, humidity, pressure) · Open-Meteo · segment-of ETAYou already know how she'll ride it: the first lap, the negative split, the reserve she'll bring to the line. We turn that call into the exact tools you'll want on race day: race-day brief, lap-time card, turn schedule. Watt by watt, lap by lap.
Founder · Sport physiologist · UCI Level 3 coach
Velometrics isn't a startup pretending to know pacing. It's the tool Borja Alfaraz (sports physiologist, biomechanist, UCI Level 3 coach) built for himself while running Venezuela's national cycling team and consulting on China's LA2028 Olympic program. The models he uses in the pit, now on your clipboard. Without the PhD to run them yourself.
5-day free trial. No commitment. The exact same physics I use with Olympians, on your clipboard.
