Kosovo is also disputed even in terms of actual casualties, let alone projected casualties, and is probably not a good case for intervention.
I fully agree there is not a good answer either way (syria included)... Just pointing out that your examples don't really predict the past or future either. Prediction in any case seems like a waste of time. (Unless you're The Doctor that is.)
A group of armed Tutsi's is ultimately what ended the Rwandan genocide. So yes, that likely would have worked well.
http://en.wikipedia.org/wiki/Rwandan_Genocide#Rwandan_Patrio...
Genocide tipping in exactly the opposite direction is another (obviously unintended) possibility. Oh yeah... The rest of the wiki you link talks about that... Oops!
You can't predict a future that didn't happen. (Again unless you're The Doctor).
Also it isn't so convincing to link to a wiki section that supposedly proves your point about reducing bloodshed, while the very next section on the page describes 5 million people killed in the same area in the following decade.
I don't pretend to have the answer in the case of Syria, and I'm not advocating anything one way or another. I'm just pointing out that your model of the world doesn't appear to predict the past, let alone the future.