Think of the following as an analogy. A guy is paddling down a river without realizing that he is headed toward a waterfall. When is the last point that you can warn him in time so that he can paddle to safety? It depends on how fast the river is moving, how fast he can change direction, how strongly he can paddle, and so on.
The sovereign debt crisis trigger point is like this problem, only one level deeper. You don't win by guessing the last point at which the guy can make it to safety. Instead, imagine that you have a bunch of people standing next to the river placing bets on whether or not the guy makes it to safety. Your job is to pick the point at which the folks doing the betting will believe he can no longer make it to safety. That is what the trigger point for the sovereign debt crisis looks like. (For those of you familiar with Keynes' beauty contest analogy, there is something similar going on here.)
In any case, this an issue where academics generally fear to tread. Rogoff and Reinhart have taken a lot of grief for their magic number of 90 percent ratio of debt to GDP. And that is for a less striking prediction. They just want to predict the point at which you will see economic growth slow down, not the point at which you get a sovereign debt crisis. I'm trying to tackle the harder issue.