it's an odd concept and counter-intuitive but it seems that two strokes are different than 4-strokes.. Actually things are not so different. Modified 4-strokes with high lift cams and other things experience it too, to some degree.
The objective of ignition timing is to produce a spark that ignites the fuel mix at the "right" time. But what is the right time? Right is whatever causes peak cylinder pressure at some stage in the piston's travel downward past TDC. But that "right" spark time can vary according to more factors than just crankshaft position.
For instance, a lean mix burns faster than a rich mix. A faster burning mixture requires less advance than a slow burn. Peak cylinder pressure would occur too soon if ajustments are not made to accomodate the faster burn.
Another thing is compression ratio. High compression ratios require less advance for similar reasons.. But can compression ratio change while an engine is running? Yes it can. If an engine is more efficient at some RPM than another, the effective compression ratio gets higher. More mixture is being compressed into the same combustion chamber space and the actual, true compression ratio is raised. A retarded spark would be called for.
Evidently, cylinder filling in 2-strokes is very inneficient at low rpms, no doubt due to the lack of valves.. At higher RPMs the inertia of the fuel mix and other factors improve cylinder filling. In contrast, volumetric efficiency in unmodified 4-strokes is more efficient at low RPMs than at high.
Add to this 2-stroke a performance exhaust and the engine suddenly gets very efficient volumetrically in a narrow, high RPM ranvge. .. at the same time the mix is clearer of residual exhaust gasses when that expansion chamber kicks in.. So timing should be retarded from what it is below that power band... because the fuel mix is much quicker burning.