Just got your dyno results? Key in the numbers and get a sanity check: implied crank power, trap speed cross-check, and whether the sheet adds up.
A dyno is a comparison tool, not an absolute one. Different dynos (Dynojet, Mustang, hub dynos), different correction standards, and different operator choices can move the same car 10% or more. These checks catch the common games:
Trap speed cannot be faked by dyno calibration. If the dyno-implied crank power and the trap-speed-implied power agree within ~7%, believe the sheet. If the dyno reads far higher, suspect an optimistic calibration or STD correction being quoted as SAE (run the numbers through the Dyno Correction Calculator - STD reads ~4% higher by itself).
The two classic moves: quoting a huge loss percentage to inflate the crank estimate ("450 whp is really 600 at the crank!"), or quoting wheel figures from a happy dyno. Typical real-world losses: RWD 12-18%, AWD 15-20%, FWD 10-15%. Outside those ranges, ask questions.
The headline peak is the number everyone quotes, but it rarely describes how a car drives. What you feel on the road is the area under the torque curve across the RPM you actually use. A tune that adds a flat 30 lb-ft from 3,000 RPM onward will out-accelerate one that spikes a bigger peak right before redline, even if their peak horsepower is identical. When two sheets show the same peak, overlay the curves: the one that comes up earlier and holds longer is the faster car. Be especially skeptical of a sheet that brags about a peak figure but crops or omits the low-RPM portion of the curve.
Dyno software applies smoothing to turn noisy raw samples into a clean line. A little is normal; a lot hides problems. Heavy smoothing erases the small dips and wiggles that reveal a flat spot, a boost-control hiccup, or a fueling glitch - it can also round a jagged peak up into a higher, prettier number. If a curve looks like it was drawn with a ruler, the smoothing is turned up. Ask for a lower smoothing setting, or for the raw overlay, when a sheet looks too perfect to be real.
A shop sends a sheet: 450 whp, 440 lb-ft, RWD, 3.0L, on a 3,700 lb car that trapped 119 mph. At a typical 15% RWD loss, 450 whp implies about 529 crank hp. The trap-speed check says 3,700 × (119 ÷ 234)³ ≈ 487 crank hp - the dyno reads about 9% high, just outside the believable window, so the correction standard and weather data are worth a second look. Meanwhile 529 crank ÷ 3.0L is roughly 176 hp/L, squarely in healthy turbo territory, so the displacement-to-power story is plausible. Two independent checks, one pass, one fail: the engine is real, but the dyno number is probably a touch optimistic.
The only comparison that settles anything is back-to-back pulls on the same dyno in the same session. Everything else - including this page - is sanity checking, not gospel.
Cross-check against trap speed: HP = Weight x (Trap MPH / 234)^3 at the crank. If the dyno's implied crank power and the trap-speed power agree within about 7%, the sheet is believable. Big disagreement usually means an optimistic dyno calibration, STD vs SAE correction games, or a poorly driven pass.
RWD cars typically show 12-18% loss between crank and wheels, AWD 15-20%, FWD 10-15%. A shop claiming 25% loss to inflate a crank estimate, or 5% to inflate a wheel figure, is massaging the numbers.
A Dynojet is an inertia dyno that calculates power from how fast the car accelerates a known drum mass, while a Mustang is a load-bearing dyno that applies resistance and can hold the car at a steady RPM. The load and drag models differ, so the same car commonly reads several percent higher on a Dynojet than on a Mustang. Neither is wrong - they measure power differently. This is exactly why cross-brand comparisons mean little and only same-dyno, same-day pulls truly settle anything.
Look for missing weather and correction data, a correction standard quietly switched to STD to gain about 4%, curves that are impossibly smooth from heavy filtering, a peak number with no torque curve to back it up, or fonts and pixels that do not match the dyno software's own output. Then sanity-check the headline against trap speed and against typical horsepower per liter for the engine. A number that fails both physical cross-checks is the strongest sign the sheet was massaged.
For a street car, area under the curve matters more than the peak. A tune that adds 15 hp at the top but fills in a flat torque curve from 3,000 RPM will feel and run faster than one that only spikes a bigger peak number right before redline. Two sheets with the same peak can drive completely differently. Read the whole curve, especially where you actually spend time on the road, not just the single biggest number.
Trap speed - the speed at the end of a quarter mile - depends on power-to-weight and cannot be inflated by dyno calibration. Using Crank HP is approximately Weight times (Trap MPH / 234) cubed gives an independent power estimate from a real-world pass. If the dyno-implied crank power and the trap-speed power agree within roughly 7%, the sheet is believable. A large gap points to an optimistic dyno, an STD-versus-SAE mismatch, or a poorly driven pass rather than honest power.
Related Calculators