I studied Cavitation professionally when I worked R&D in the Oil Patch. Cavitation happens when the Shear Strength of a fluid is exceeded. In the situation I researched, it occurred when a valve was shut abruptly. The fluid had velocity as it passed through the orifice, and when the valve shut, the inertia of the fluid would pull the molecules apart creating small pockets of very low pressure akin to vacuum bubbles. This by it's self is not the destructive phase of cavitation. The damage occurs when the bubbles collapse. When we started the research, the first thing we did was to collect all known studies and reports on cavitation. We found that the US NAVY had blocked the release of the known research since their submarine propellers produce cavitation as they screw through the water. Sub searchers drop a microphone into the water and listen for the tell-tale clicking noise that the bubbles make when they collapse. Anyways the shock of the bubbles collapsing creates very localized pressures that are in the 1,000,000 PSI range,(yeah, one million PSI) more than enough to peck away at cast iron and even very hard materials like Tungsten Carbide. I suspect that as the cylinder wall on our 7.3's flexes from the shock of combustion, the coolant is forced away from the water jacket side of the cylinder and when the bubbles collapse we lose material to cavitation. When we tried to prevent the cavitation from occurring by raising back pressure on the valve we the found that the cavitation was worse. So to make some assumptions, we might assume that if the Water-less coolant truly does prevent cavitation, it may be due to the fluid having a higher shear strength than water or the ability to run a Zero Pressure system minimizes the effect of the bubble collapse or both.