Gross versus net horsepower

Start with the test, not the biggest number

SAE's April 1971 J245 engine-rating code identifies engine configuration and test conditions as requirements for gross and net designations. “Horsepower” alone leaves that context unresolved.

Equipment can change the result. As a concrete modern example, Briggs & Stratton's rating explanation distinguishes gross tests without the exhaust and air cleaner from net tests with them. That small-engine example illustrates the principle; it does not establish which equipment was used for a particular classic-car rating.

The modern J1349 net-rating summary describes repeatable measurements representative of an engine in service. That later standard is useful context, not evidence that its current procedure governed a 1971 rating. These links provide SAE's public scope statements; the full standards are not reproduced here.

A factory example: four 1971 Corvette engines

The September 1970 AMA specification sheet, printed page 4 (PDF page 139), lists these paired ratings. They describe the named 1971 Corvette applications; the LS6 row is not a 1970 Chevelle specification.

Calculation: (gross − net) ÷ gross × 100, rounded to one decimal place. These gaps compare published peak ratings, sometimes at different RPM. They are not measured accessory losses, drivetrain losses or conversion factors.

Source discrepancy: the kit's Power Trains–3 sheet (PDF page 118) has an LS6 heading of 460 HP but a gross-rating row of 425. The table above follows the explicit gross/net entries on the AMA sheet. The scan remains available so readers can check that choice.

Why “subtract 20 percent” is unreliable

Even this small, same-year example yields different gaps. Applying the LT1's percentage to another engine would replace that engine's evidence with an assumption. If a source supplies only gross power, report it as gross; leave the net value unknown until there is matching evidence.

A useful comparison needs more than equal units. Match the engine application, model year, rating basis and RPM. For a new engine test, also record the tested equipment and correction method. For a road test, record the car's transmission, gearing, tires and modifications. Do not infer acceleration from the horsepower column alone.

Why a year-to-year drop needs two explanations

A rating-method change can coincide with an engine change. GM's 1972 Corvette history describes both lower compression and a move from gross to net when discussing the LT1's earlier 370-hp rating and its 1972 255-hp rating. The numerical difference alone cannot separate those effects.

Keep the historical numbers and their labels. Avoid rewriting an older rating as a supposedly equivalent modern figure, or describing every change as either a testing artifact or a physical loss of power.

Sources, method and corrections

This guide transcribes four pairs from one factory sheet and calculates their percentage differences. The figures were checked visually against the scan. This is a limited documentary comparison, not a representative sample of all engines or an independent engine test.

Prepared September 13, 2026, by Classic Muscle / Data Relativity with AI-assisted research and drafting. No independent specialist review is claimed. See how we check the record for attribution and evidence practices, or contribute a source that clarifies a disputed rating.