These are two different kinds of quantity, which is why arguing about which one "matters more" rarely goes anywhere. Torque is a moment, and NASA's own beginner's guide defines it as "the product of the force and the perpendicular distance" to the pivot. Power is a rate: HyperPhysics puts it as "the rate of doing work or the rate of using energy". One tells you how hard, the other how fast.
Revs are what join them
In an engine they are joined by revs. The same twisting force delivered more times a second does more work per second, so a given torque figure produces more power the faster the crankshaft is turning. That is the whole reason the two curves on a dyno graph peak in different places, and why an engine can be past its torque peak and still climbing in power. The join has an exact form in imperial units: horsepower equals torque in pound-feet multiplied by revs, divided by 5,252. The constant is not arbitrary. A horsepower is 33,000 foot-pounds a minute, and a full turn of the crank moves through 2π radians, so 33,000 divided by 2π gives 5,252. It also explains a detail every dyno plot shows and few articles account for: because the two are equal when revs are 5,252, the torque and power curves always cross at exactly that point on an imperial graph.
What it means from the driver's seat
What that means from the driver's seat is fairly intuitive. Torque is most of what you notice pulling away and in gear at low revs, while power is what keeps a car accelerating once it is already moving. Neither number describes a car on its own, which is why both get quoted. The units themselves are on the horsepower and torque pages, and where the power figure gets measured is on BHP.
Written by Craig Fearn, Petrol & Ink. Last updated .