First, don't take OEM output numbers as anywhere near fact. And most dyno output numbers are even further from fact.
If you know what the injectors flow at specific pressures and duty cycles, as well as the head/cam flow capeability, or even the total weight and mph in the 1/4 mile, you have three fairly accurate ways to determine actual output at the crankshaft. This rarely comes close to what OEMs claim for output.
Next add in losses for gears (submerged and spray lubrication differ), chains, belts, etc. Despite what some dyno manufacturers claim, a coast down does not load the drivetrain anywhere close to an acceleration run.
Tires differ greatly in the way they absorb and transfer output to the ground though their sidewall design, and compound.
The way output is measured via static loading, controlled acceleration sweep, or inertia sweep (whatever rate it spins the dyno drum), all change the output of the engine, as well as how much rotational/inertial mass of components alters the output.
So now, you may start to see that what we think the bike actually makes, and what it actually produces, can be far from the same. The only way to know, is to test on the road or track. And if you are on a dyno, you can compare before and after differences in percentage, but even this has flaws, as some dynos vary more and more as output increases.
Some interesting reading on the subject, for those who haven't been in the industry since the 1980's, or spoken to Mark Dobeck in person:
The Story Behind the Dynojet Chassis Dyno - The Truth Meter - Hot Rod Magazine
I hope this helps.