Then you see my issue.
Power = Force x Time
You have X amount of power available to you per the fuel/air available as set by the throttle. In level steady state flight you have to have enough power to maintain 1G of acceleration against gravity plus the torque to overcome the drag of your speed.
If you now want to climb, you need to accelerate against more than 1G. If you don't change the power available, that has to come from somewhere, so in order to maintain RPM, it comes from the torque that maintains your forward speed. That is achieved by producing a shallower blade angle that has less resistance freeing up torque to turn into HP for the climb taking it from airspeed. So when you pitch up to climb, the load on your engine increases causing RPM to decrease, causing the governor flyweights to retract causing the spool valve to open releasing oil and allowing the blade to flatten pitch and maintain RPM.
Umm, most of us climb at 1G. The exception would be if someone were flying a loop or a parabolic trajectory.
And power = force * velocity. Not time.
The cylinders produce pressure, closely related to force because the pistons are constant cross section. MP is measuring input pressure, which is related to the output pressure we want. It's proportional for correct mixture and normal burn.
The engine power is the force times the velocity of the pistons. That's (up to a constant factor) RPM.
Some portion of that power -- actually quite a lot of it -- into throwing air around into fancy shapes like wingtip vortices. What's left over goes into changing the total energy of your system. At constant airspeed, that's climb rate. And 2/3 of the power (roughly) goes into waste heat, about half of that out the exhaust, for a gasoline engine.
If you want a lot left over, you want the engine screaming as fast as it can go, with as much absolute pressure as it can handle.
That goes for your car, your airplane, your lawnmower, your generator, whatever. Subject to limitations that might damage it, of course. And it even works with forced air. With an airplane, we can pick RPM and hold it constant, within limits, which is convenient. There are some other applications that do that to, particularly with "continuously variable transmissions." As no one has figured out how to do that in a car without using belts, there aren't any cars. But there are snowmobiles, and a few light duty scooters. Believe me, people would like cars to behave that way, as it's a whole lot easier to design engines that work over narrow RPM ranges.
So, how does this work for CS props?
Within limitations, you want it as fast as possible (prop forward) and as hard as possible (full throttle) for takeoff. The ONLY function of the governor in this configuration is to prevent overspeed. When stopped on the runway, full throttle will likely not be enough to pull the blades off the fine stops. As the aircraft speeds up, they will.
Later, you can trade engine "oomph" (MP) for engine speed (RPM), as the power produced is their product. The cylinder temperatures, however, are much more closely related to the pressure, so MP is going to have a big effect. Similarly, RPM determines cooling, as oil volume is in proportion (the oil pump is engine driven). So, your engine runs the coolest for a given power at high RPM, low throttle.