Published HP figures are almost always misleading because they're typically calculated from exaggerated RPMs.

While 2-strokes like to run at high RPMs, four-strokes are typically happier at lower RPMs where they have much more torque than 2-strokes.

HP = torque (ft/lbs) * RPM / 5252

So it's easy to see that the 2-strokes, while producing more horsepower than the 4-strokes, do so because although they produce less torque than the 4-strokes, they operate at much higher RPMs. On the other hand, a four-stroke with more torque swinging a larger prop will perform similarly.

Is this as clear as mud now?