Skcos

This isn't really how it works.  Large objects are indeed subject to greater stresses than small objects when in orbit, but the effects are backwards.
 
Any object in orbit around a larger object experiences a certain amount of tidal stress.  Essentially this is the difference in the force of gravity on the near side of the object relative to the force on the far side of the object.  Given a fixed diameter object, such as a moon, the difference between the near and far sides is much greater in a low orbit than it is in a high orbit.  If we increase the diameter of this moon then we also increase this difference.  Earth's moon, for example, experiences enough tidal stress at a relatively high orbit to keep one side facing our planet at all times and create a bulge on the near side.If the Moon were brought closer to Earth then the tidal forces would increase, the bulge would be exaggerated and we could start seeing seismic activity on the Moon (as well as Earth).  If it were brought too close then those tidal forces would rip it apart and, instead of a single moon, we would have rings.  The distance at which this happens is called the Roche limit.  The stress isn't related to "entering" the gravitational field or how high the object is when it "starts," what matters is the object's distance relative to the Roche limit at any given time.
 
On the opposite end we have small objects, such as spacecraft, satellites, most asteroids, meteors, etc.  These objects are so small that the tidal forces they experience are negligible even at extremely low orbits.  If someone were to try to launch a huge object at us, such as Ceres, then it might break up during its approach (though the effects would still be devastating, intact or otherwise).  If, on the other hand, someone picked one of the small asteroids then it would likely remain intact right up until it hit the atmosphere.  Actually redirecting the object towards the planet is the real challenge; in addition to the high energy requirements this kind of attack would probably result in a tug of war between attackers and defenders trying to redirect a specific asteroid months or years in advance of the actual impact.
 
As for game purposes things get a bit more difficult.  I highly doubt this game will have anything resembling orbital mechanics and it sounds like we won't normally have collision damage.  It is possible that the devs could code a special exemption to have objects which fall from space and hit the ground at sufficient velocity to explode, though it would be a lot of extra work on their end to code both this system and a method to push asteroids around.  On the other hand bombarding a planet with conventional weapons could be made as simple as pointing your guns directly at the target and firing.  This wouldn't work with orbital mechanics (unless you use lasers or particle beams) but it might work in DU's physics with the main concern being game balance.
 
EDIT:  One thing I should also mention, if someone does manage to throw an object at a planet that is large enough for the Roche limit to actually be a factor then it probably doesn't matter whether it breaks up before impact or not.  Either way you can pretty reliably count on wiping out all life on the surface.

Megaships would be nice, indeed. However, I would stave my enthusiasm. Water always mean technical difficulties.