It isn't that the numbers are made up; it's just that they're probably exaggerated a bit. For example, in industry the standard is to throw out the high and the low, and average the rest. With a motorcycle dyno, you throw out everything but the highest reading and then tweak it a little. This may sound funny, but it's exactly what goes on. In fact it's actually much worse than that; many Dyno's are modified to read high. Operators find that if their Dyno's are accurate their customers complain, cry crocodile tears and sometimes want their money back for the header you just installed. If the Dyno reads high, nothing but praise, and another happy customer. Manufacturers of headers and other performance items also tend to record only the most extreme claims. It's all about sales.
I've always used 12 to 15 % correction for crank to rear wheel hoprsepower. Dyno's can actually test for the exact power drive train losses. It's called a negative power pull.
Last edited by DaveL; 06-05-2007 at 03:24 AM.