Do you know a link to possible/probable re-entry dates of the larger pieces? i.e. the Centaur stage that lifter the military GPS satellite yesterday
Predicting orbital decays is a bit of a black art. Or, to use my own favorite analogy, it's like predicting the outcome of a roulette wheel. As it gets closer you can begin to predict when, but exactly
where is almost always a guess.
You can find all the objects associated with a given launch by looking up its international designator, which is of the form
2016-001A
where 2016 is the launch year, 001 is the launch sequence within the year, and A is the piece of the launch. Usually, but not always, A is the primary payload and B is the upper stage of the launch vehicle. C, D, and so forth are any secondary payloads. Then look for the other objects from the same launch and examine their orbital elements. Unless the perigee is lower than 300-400 km and the eccentricity is near zero, decay is not imminent.
Lots of things affect decay rate, from the mass, size, shape and attitude of the object (which controls its drag) and solar activity (solar maxima heats the outer atmosphere and increases drag at a given altitude). For decaying objects the keplerian orbital elements frequently include the time derivative of the mean motion, another indicator of decay rate. The mean motion is the number of orbits per day, i.e., 24 hours divided by the period. This increases as the orbit decays. When there's no decay and no perturbations, the mean motion derivative is zero.
In recent years it has become customary for launch vehicles to save some fuel to deorbit themselves after their missions are over. They may deorbit very quickly (like within an orbit or two) or they may simply maneuver into (or remain in) an orbit that will decay within some specified number of years.
Edited to add: This might be of interest:
http://www.satview.org/spacejunk.php