corogers, I'm a cheap skate. I don't want to waste the CO2 required to fill, release, fill, release. Each time you fill and release, you reduce the O2 level a bit - the amount depends on the pressure you use.
If you're using 15 psi (about 1 atmosphere at sea level): Pushing in 15 psi of CO2 will squeeze the air in the keg to about half its uncompressed volume and fill the remaining space with CO2: air is about 20% O2, it's been squeezed to half its volume, so your keg now contains about 10% O2. The rest of the keg is filled with N2, CO2, and insignificant quantities of other (non-reactive) gases. When you vent the keg, the percentages remain the same. Pressurize with CO2 again at 15psi and you will squeeze the O2 down to half its previous volume - 5%. Repeat the process ad nausium and you'll be reducing the number of O2 molecules by half with each pressurization and release cycle.
I keep a keg of Star San pressurized and ready (yum! Sparkling sanitizer!). When I clean a keg, I do a liquid - to - liquid transfer of the Star San to the cleaned keg and pressurize it. The newly emptied Star San keg is now (almost) oxygen-free and will receive the next beer that's ready to carbonate and lager.
Does my process use less CO2 than several fill-release cycles? I guarantee that I think it does. Or ... maybe not. But the arithmetic is fun to play with; too bad it's all based on estimates and opinion.