Measuring horsepower from a run

A timeslip formula estimates power from a result. Measuring is different: if you know how fast a car of known mass accelerated at every instant, you know the force that produced it — and force times speed is power. That gives a wheel-power curve across the whole run rather than a single inferred figure.

Estimating versus measuring

The familiar trap-speed and ET formulas are estimates. They were fitted to a large number of real timeslips decades ago, and they work well on average, but they only ever see a result: a time, a speed, a weight. They cannot tell you where in the rev range your engine is strong or what a gearchange costs you. Try the estimate →

Measuring works the other way round. If you record how the car's speed changed throughout the run, you know its acceleration at every instant. Multiply by mass and you have the net force. Add back what was spent pushing air aside and rolling the tyres, and you have the force the engine actually delivered — and force multiplied by road speed is power.

What you get out of it

• A curve, not a number. Power across the whole pull rather than one headline figure.
• Gearchanges become visible. Each shift shows as a dip where torque briefly stops reaching the wheels — which also lets a run be split into gears.
• Wheel power, honestly labelled. This is power at the road, not at the crank, and it should never be quoted as if it were a crank rating.
• It responds to reality. Bad air, a headwind, or a slight uphill all show up, because they genuinely changed what the car had to do.

Where the uncertainty lives

Being straight about this matters more than the number itself. Two inputs are not measured but assumed: vehicle mass and aerodynamic drag. Mass you can improve by weighing the car with you and your fuel in it. Drag is a genuine estimate — it depends on frontal area and body shape, and at high speed it is a large share of the total work.

That is why a measured figure deserves a range rather than three decimal places, and why the honest quantity from a speed interval is an average across the pull rather than an invented peak. A peak needs a power curve and gearing knowledge that a stopwatch does not have.

Why not just use a dyno?

A calibrated chassis dyno is more precise and remains the reference. But it is expensive, it is a scheduled appointment, and it measures the car strapped to a machine rather than moving through real air. A measured run costs nothing, happens wherever you already are, and answers the question most people actually have — did that change make my car quicker? — with a before-and-after you can repeat any time.

The two are complements, not rivals. The dyno tells you what the engine makes. A measured run tells you what the whole car does with it.