To me it is very complex. I have never ever tried to figure an out of control rock from HEO sped by solar activity to LEO and eventually to stratosphere and slowed by it’s own unknown weight. NASA did a good job of making sure it didn’t hit me on the head. :^)
Saw a tweet UARS, if still in orbit, ought to be visible from Australia in the next few minutes.
Haha...forecasting the landing location is extremely complex; the methods they use are far enough beyond my understanding that I don't even know most of the words involved.
But figuring the odds of a person being hit, given that the landing location was essentially randomly chosen, is much simpler: Figure a person takes up one square meter of space. Multiply that by the total population of the planet and you get the total surface area covered by people. Divide that number by the total surface area of the Earth (or of the portion of the Earth within the orbit's coverage area) and you get the odds that any one piece will hit a person. Multiply those odds by the number of pieces falling and you have your odds.