Why this ratio predicts acceleration and raw power does not
Newton's second law says acceleration equals force divided by mass. In a car, the force available at the wheels comes from engine power delivered through the gearbox, so the acceleration you can generate scales with power divided by mass. That is why a 250 hp lightweight and a 400 hp saloon can be neck and neck: 2,600 lb over 250 hp is 10.4 lb/hp, and 4,200 lb over 400 hp is 10.5.
The ratio is also the only fair way to compare vehicles across categories. A litre bike making 200 hp and weighing 450 lb with rider is at 2.3 lb/hp, which no road car reaches. A loaded tow rig at 20,000 lb and 400 hp is at 50 lb/hp. Both are described completely by the same number.
Racing bodies use it for exactly this reason. Rather than police engine specifications in detail, many series simply set a minimum weight for a given power output, or a maximum power for a given weight class. Enforcing a ratio is enforcement of performance, and it is easy to check with a set of scales and a dyno sheet.
What the ratio does not capture is traction, gearing, aerodynamic drag and driver skill. It predicts the acceleration a car could deliver if it could put its power down, which is why it correlates far better with quarter-mile trap speed — a speed measurement, set by power and drag — than with elapsed time, which is dominated by launch traction.
The three conventions and how to move between them
All three common expressions are the same physical quantity with different units, so converting between them needs no new information.
Pounds per horsepower divides race weight by power. Lower numbers are quicker. This is the American convention, and it is the one racing class rules are usually written in, because it makes a class minimum easy to state: “no lighter than 6.00 pounds per certified horsepower”.
Horsepower per short ton inverts it and scales by 2,000 lb, so higher numbers are quicker. Because a short ton is 2,000 lb, the conversion is simply hp/ton = 2000 ÷ (lb/hp). Note that a US short ton is not a metric tonne, and it is not a British long ton of 2,240 lb either — a difference of about 10% between the first and the last.
Kilowatts per tonne is the European and homologation convention. One horsepower is 0.7456999 kW and one metric tonne is 2204.6226 lb, so kW/t = 1643.99 ÷ (lb/hp). Somebody quoting 200 kW per tonne is at 8.22 lb per hp.
The subtlety that trips people up is not the arithmetic but the definitions of the two inputs. Weight can mean kerb weight, dry weight, or weight as raced with driver and fuel, and those can differ by 300 lb or more. Power can mean flywheel or wheel horsepower, and those differ by roughly the drivetrain loss. Mixing a dry weight with a flywheel figure gives a flattering ratio that no scrutineer will accept, and comparing your wheel-hp ratio to a manufacturer's flywheel-hp ratio is not a comparison at all. Pick one convention for both terms and say which you used. If you need to check the power side, the torque-to-horsepower converter will keep hp, kW and PS straight.
Worked example: 3,500 lb car, 200 lb driver, 500 hp
A track-day car scales at 3,500 lb with a full tank, and the driver in gear is 200 lb. The engine made 500 hp at the flywheel.
- Race weight. 3,500 + 200 = 3,700 lb.
- Pounds per horsepower. 3,700 ÷ 500 = 7.400 lb/hp.
- Horsepower per short ton. 3,700 lb is 1.85 short tons, so 500 ÷ 1.85 = 270.27 hp/ton. Cross-check with the shortcut: 2,000 ÷ 7.400 = 270.27. They agree.
- Kilowatts per tonne. 500 hp × 0.7456999 = 372.85 kW. 3,700 lb ÷ 2204.6226 = 1.6783 tonnes. 372.85 ÷ 1.6783 = 222.16 kW/t. Cross-check: 1643.99 ÷ 7.400 = 222.16.
Now the question that actually matters. The class you want to run has a 6.00 lb/hp minimum. At 3,700 lb you need 3,700 ÷ 6 = 616.7 hp — another 117 hp. Alternatively, keep the 500 hp and get the race weight down to 500 × 6 = 3,000 lb, which means shedding 700 lb.
Neither route is obviously cheaper, but the arithmetic tells you the exchange rate: at this operating point, one horsepower is worth exactly 6 lb, because that is the target ratio. Below the target ratio the same trade is worth fewer pounds per horsepower; above it, more. That exchange rate is the reason lightweight cars respond so well to power and heavy cars respond so well to diets.
What your number means in practice
Read the ratio as a rough band rather than a precise prediction, because gearing, tyres, aerodynamics and drivetrain type all shift real-world results around a given ratio.
Above about 20 lb/hp you have an economy car or a laden work vehicle; acceleration is adequate rather than brisk and overtaking needs planning. Between roughly 10 and 15 lb/hp sits the bulk of modern family cars and light trucks. From about 7 to 10 lb/hp you are in warm-hatch and sports-saloon territory. Below 5 lb/hp is supercar and serious track-car ground, and below 3 lb/hp is superbike and purpose-built race car. These bands describe what the ratio buys you; they are not a claim about any specific model.
Two adjustments matter when comparing. First, all-wheel drive costs a little ratio through extra mass and drivetrain drag but converts far more of it into launch acceleration, so an AWD car will usually beat a rear-drive car of identical ratio off the line and lose ground at high speed. Second, aerodynamic drag rises with the cube of speed in terms of power required, so above roughly 100 mph the ratio stops governing and frontal area and drag coefficient take over. That is why the ratio predicts a standing quarter-mile trap speed well and a top speed poorly.
If you are using this to plan a build, decide which side of the ratio is cheaper for you. Power costs money and usually costs reliability; weight costs money and usually costs comfort or crash structure. Removing weight also improves braking and cornering, which power does not, so on a circuit the same ratio reached by lightness is worth more than the same ratio reached by power.
The three ratio conventions side by side
| Pounds per hp | Horsepower per short ton | Kilowatts per tonne |
|---|---|---|
| 3.0 | 666.7 | 548.0 |
| 4.0 | 500.0 | 411.0 |
| 5.0 | 400.0 | 328.8 |
| 6.0 | 333.3 | 274.0 |
| 8.0 | 250.0 | 205.5 |
| 10.0 | 200.0 | 164.4 |
| 12.0 | 166.7 | 137.0 |
| 15.0 | 133.3 | 109.6 |
| 20.0 | 100.0 | 82.2 |
| 30.0 | 66.7 | 54.8 |
A US short ton is 2,000 lb; a metric tonne is 2204.6226 lb; one horsepower is 0.7456999 kW.
Assumptions and traps
- Weight definitions differ by hundreds of pounds. Dry, kerb, and as-raced-with-driver are three different numbers. Class rules almost always mean the last one, measured straight after the run.
- Flywheel and wheel horsepower are not interchangeable. Using a chassis dyno figure against a rival's manufacturer claim understates your car by roughly the drivetrain loss.
- A short ton is not a tonne. 2,000 lb versus 2,204.6 lb is a 10% error if you mix them, which is more than most modifications are worth.
- The ratio ignores traction. Two cars at 4 lb/hp with different tyres and drive layouts will not run the same elapsed time, though they will trap at similar speeds.
- It ignores aerodynamics. Above about 100 mph the power needed rises with the cube of speed, and drag area governs rather than mass.
- Fuel load changes it during a run. Twenty gallons of gasoline is roughly 120 lb, which moves a 3,700 lb car's ratio by about 3% between the start and end of a stint.
Where the ratio fits with the rest of the numbers
Power to weight is the top of a chain. Below it sit the things that actually create the power: displacement and compression set the airflow the engine can use, boost multiplies it, and gearing decides how much of the resulting torque reaches the road at any given speed. If you are trying to move the ratio, the practical order is usually to verify the power first, then attack whichever side of the ratio is cheaper.
To verify the power side without a dyno, use the trap-speed method: quarter-mile terminal speed is a direct function of power and weight, so it back-calculates flywheel horsepower with useful accuracy and is very hard to game. If the trap-speed estimate and the dyno sheet disagree by more than about 10%, one of them is wrong, and it is usually the dyno sheet.
To move the power side, the airflow calculators are the place to start: boost horsepower for forced induction, carburettor CFM for induction sizing, and injector sizing so the fuel system can support what you plan to make. To make the power usable once you have it, axle ratio and tyre diameter decide where in the rev range the engine sits at any road speed.
One convention note for anyone writing or reading class rules: state whether power is certified at the flywheel or the wheel, whether weight is with or without driver, and whether it is measured before or after the run. Rules that leave any of those three ambiguous get protested. The ratio itself is unambiguous arithmetic; the definitions of its two terms are where every dispute lives.