How a dyno number gets inflated — five methods, and how to check

1. The correction factor nobody mentions

Every dyno corrects its raw reading to a reference atmosphere, and there is more than one standard to correct to. SAE J1349, SAE J607, STD and DIN all describe different reference conditions, and the same pull reported under two of them can differ by around four per cent. On a five-hundred-horsepower car that is twenty horsepower that exists only in the arithmetic.

This is not fraud by itself — every standard is legitimate. It becomes a method when a shop quotes its own numbers in the standard that flatters them and compares against a baseline in the one that does not, without saying so.

Check: the correction standard is printed on the sheet. If it is not, the sheet is not evidence.

2. The weather station that took the day off

Correction only works if the dyno knows the day’s actual temperature and pressure. That data comes from a weather station, and a weather station can be unplugged, mis-mounted next to something hot, or simply overridden. Feed the correction a colder, denser day than the one outside and every number on the sheet rises together.

The tell is a sheet with no ambient conditions on it at all — or a shop whose published pulls carry identical conditions in January and July.

Check: ambient temperature and pressure, printed on the sheet, varying the way real weather varies.

3. The variables a roller adds

A roller dyno measures power after it has passed through the tire. Tire compound, pressure, temperature and strap-down tension all change how much of the engine’s output survives that trip — which means all of them are levers. Lower the pressure, cinch the straps differently, run the pull on a warmer tire, and the number moves with nothing about the engine changed.

A hub dyno bolts to the wheel hubs. There is no tire in the measurement chain, which is why the same car measures the same on it next month. That repeatability is the entire reason a before-and-after comparison means anything.

Check: what kind of dyno, and whether the before and after came off the same one.

4. The crop

The most common method is the simplest: publish the peak and nothing else. A peak number without its curve hides where the power lives, hides the run-to-run scatter, and hides the pulls that did not cooperate. A shop confident in its work publishes the whole sheet, including the runs it is not proud of, because the shape of the curve is the work.

Check: ask for the full sheet, every run. A refusal is an answer.

5. Moving the goalposts between runs

The last method compares two numbers that were never measured under the same rules: a baseline on pump fuel against a final on race fuel or E85, quietly; a baseline from another shop’s roller against a final from this one’s; a stock figure from a magazine against a measured tuned figure. The delta is the headline, and the delta is a fabrication even when both endpoints are real.

When we publish a gain we publish the whole session it came from. Our most-shown example is a C8 Stingray measured three times in one afternoon on the same hubs under the same correction: 427.9 hp stock, 449.6 on a 91-octane calibration, 474.7 on E85. The fuel is named on each run because the fuel is most of the story.

Check: same day, same dyno, same correction, fuel named per run. Anything less is two unrelated numbers.

Why we publish the method

Every check above takes seconds once you know to make it, which is exactly why we put our methodology on its own page and our conditions on every sheet. A shop that shows its working cannot be ambushed on it — and a customer who knows how numbers get moved is our customer, because measured is the only thing we sell.