Quarter-Mile Times: What Horsepower-to-Weight Really Buys

Search for a "1/4 mile time calculator" and every tool worth using is running the same one-line physics: estimated elapsed time equals a constant multiplied by the cube root of weight divided by horsepower. No hidden simulation underneath — no gear ratios, no tire model, no launch. Two numbers go in: total race weight in pounds (driver, fuel and everything else aboard during the pass) and a horsepower figure. The quarter-mile calculator implements exactly this, but unlike most tools it names its constant — because there isn't one constant. There are three in common use, and at street-car power levels they disagree by nearly a second.

The immediate answer, for a typical combination: a 3,200 lb car with 400 hp carries 3,200 ÷ 400 = 8 lb per horsepower. The cube root of 8 is exactly 2, so the Fox street-car model estimates an ET of 6.269 × 2 = 12.538 seconds, with a trap speed of 230 × 0.5 = 115 mph (0.5 being the cube root of the inverted ratio, 1/8). That is the entire calculation — the rest is knowing where each constant comes from and what the cube root does to your budget.

ET = C × ∛(weight ÷ horsepower) · trap speed = C′ × ∛(horsepower ÷ weight)

One formula, three constants

The constants are empirical — fitted to real cars on real strips, not derived from first principles. Geoffrey Fox published the underlying physics in a 1973 American Journal of Physics paper, "On the Physics of Drag Racing," showing that under reasonable assumptions about aerodynamic and rolling losses, trap speed scales with the cube root of power-to-weight. The named calibrations sit on top of that relationship, and each reflects the cars behind its fit:

Same car, same 8 lb/hp — and the ET estimates span 12.580 − 11.650 = 0.930 seconds. That spread is information, not a flaw: Hale's constants describe race-prepared cars that launch hard on prepared surfaces, so at identical inputs it always predicts quicker and faster than the street-car fits. A full-weight street machine belongs on Fox or Huntington; a strip-built car on Hale. What you must never do is switch calibrations mid-comparison. Swapping Fox for Hale "gains" 12.538 − 11.650 = 0.888 seconds without touching the car — a bigger apparent effect than most genuine modifications produce.

What the cube root actually buys

The cube root is the reason drag racing is expensive. Because ET scales with ∛(weight ÷ power), halving your lb-per-hp figure does not halve your ET — it multiplies it by ∛(1/2) ≈ 0.7937, a gain of about 21 percent.

Concretely: double the 3,200 lb car's power from 400 to 800 hp. The ratio falls to 3,200 ÷ 800 = 4 lb/hp, the cube root of 4 is about 1.5874, and Fox now estimates 6.269 × 1.5874 = 9.951 seconds — 2.587 seconds quicker for twice the power. Trap speed climbs to 230 × 0.6300 = 144.9 mph (0.6300 ≈ ∛0.25). Real gains, but nowhere near proportional to the outlay. Here is the Fox ladder across the ratios most builds occupy, with trap speed computed as 230 ÷ ∛(lb/hp) — the same cube root, inverted:

Weight pulls the same lever from the other end, which is why it is usually the cheaper move. Take 300 lb out of the 3,200 lb / 400 hp car: 2,900 ÷ 400 = 7.25 lb/hp, cube root about 1.9354, so Fox gives 6.269 × 1.9354 = 12.133 seconds — 12.538 − 12.133 = 0.405 seconds found in ballast. Buying that same ratio with power instead would take 3,200 ÷ 7.25 = 441.4 hp, so the 300 lb diet is worth roughly 41 horsepower on this car. And because ET depends only on the ratio, the model cannot distinguish a light, modest car from a heavy, powerful one at the same lb/hp — to first order, neither can the stopwatch.

Running it backwards: horsepower from a time slip

The relationship inverts cleanly, and the inverse is the question behind every "quarter mile hp calculator" search: how much power is that car really making? The horsepower calculator's trap-speed mode implements the drag-strip calibration HP = weight × (trap speed ÷ 234)³ — Fox's cube-root relationship with the 234 speed constant drag racers settled on, numerically the same constant the Hale model uses. A 3,400 lb car that trapped 100 mph works out to 3,400 × (100 ÷ 234)³ = 265.36 hp. No dyno, no teardown, just a time slip. (Its other mode computes horsepower the dynamometer way, torque × RPM ÷ 5252, when you have a spec sheet instead of a slip.)

Why trap speed and not ET? A bogged launch or a missed shift wrecks elapsed time while barely denting the speed carried through the traps, so the finish-line number is the more honest witness to delivered power.

One catch in the round trip: the constants must match. Run Hale's 117 mph prediction for the 3,200 lb / 400 hp car back through the 234-based inverse and you get 3,200 × (117 ÷ 234)³ = 3,200 × 0.125 = 400 hp — exactly the input, because 234 appears on both sides. Run Fox's 115 mph prediction through that same inverse and you get 3,200 × (115 ÷ 234)³ = 379.8 hp: about 20 horsepower evaporates purely because a 230-calibrated output met a 234-calibrated inverse. Neither figure is wrong. They are answers from different fits, and mixing them silently is the most common way these estimates get misquoted.

Where the fits stop fitting

An empirical formula is only as good as the population it was fitted to, and these have edges worth respecting.

They assume traction. The cube-root law describes cars that can use their power. A 900 hp street car on ordinary tires is traction-limited off the line: the formula predicts an ET the tires cannot deliver, the real slip reads slower, yet trap speed lands close to prediction. That asymmetry — ET sensitive to launch, traction, gearing and shift time; trap speed mostly to power, weight and aerodynamics — is the most useful diagnostic the two outputs give you.

They assume their own era and preparation. Fox's street-car physics dates to 1973 and Hale's fit describes race-prepared machines; the further a car sits from either population — a modern all-wheel-drive launch-control machine, say — the less the constant deserves your trust.

They assume honest inputs. Weight means total race weight — car, driver, fuel, gear — not the brochure curb figure, which routinely runs a couple hundred pounds light. Horsepower must keep one basis throughout: crank and wheel figures differ by the drivetrain's losses, and swapping bases mid-comparison manufactures phantom gains exactly the way swapping models does.

None of this makes the formulas useless. It makes them comparison tools rather than prophecy: hold the model, the weight convention and the power basis fixed, and the estimate cleanly isolates the one thing you changed. The distance, at least, is exact where the physics is not — a quarter mile is 1,320 feet, or 402.336 meters.

Before you chase the number

A calibration is a bet about which population of cars yours belongs to, and only a time slip settles the bet. So fix your bookkeeping before you fix the car: one model, one weight convention, one power basis, written down next to every estimate. Then let the strip grade the prediction — a sanctioned drag strip, with its safety inspection and its timing beams, is the one venue these numbers were fitted to, and a public road was never in any of the datasets. When the slip comes back, read it the way the formulas were built to be read: trap speed tells you about power, ET tells you about everything else, and the gap between prediction and pass tells you which problem to spend money on next. That reading habit transfers — Quanta's full catalog is built on the same disclosed-constant discipline as these two tools, and the contact page is there for corrections, calibration questions, or anything a page left unclear.

Sources

• SAE J1349 — Engine Power Test Code (standardized net power ratings)
• National Hot Rod Association — glossary of elapsed time, trap speed and other drag-racing terms