Now, a major world will have dozens of field armies - you don't try to invade them. You bombard them until they surrender - and bring in cadre from the center, and troops from closer than the center... It's the minor worlds nearby that you deny them with your troops. And you take their troops, and expend them on the other side of the empire...
Basic Machiavelli stuff.
I've often pondered this issue.
I don't like the TCS approach to taking a world. "World surrenders when you appear in its sky with the ability to bombard it, for fear that you will" just doesn't do it for me.
A world is a VERY big thing to bombard, and can be given all sorts of planetary defences. Missile bombardment is futile. However many missile racks you can put in the sky above my world, I can put more Repulsors on the ground in all the places that matter. All other weapons on board are point-attack.
Plus, even if a world DID surrender just because it had a few Space Invaders to contend with, I can't see it suddenly becoming a loyal, tax-paying, resource-contributing member of your polity.
To control a landmass, whether it be an island in the pacific, the European mainland, the District of Columbia or Deneb - 3, you will ALWAYS need boots on the ground. Therefore you will ALWAYS need to move armies through space if you want to take somebody else's world.
I think DS9 got this one right, as well, with their Cardassian occupation of Bajor (transparently modeled on the Soviet occupation of Afghanistan): if the population of the world is hostile to your occupation, then your problems don't end when you overwhelm the "official" defence forces. In many cases, they're only just beginning. Therefore, a heavy continuing military presence to support the civilian administration which you impose will often be required.
There are, of course, cases (such as the Japanese incursions into South East Asia) where many of the occupants of the invaded territory considered (initially at least) the invader to be a more welcome occupier than their previous colonial masters; and there will always be a certain number of willing collaborators (as in Vichy France). But I do not see ANY way of taking somebody else's world other than transporting large numbers of troops through space and depositing them on the real estate.
I therefore conceive of the troop transport function of any offensively-organized Navy as being a co-equal with the fighting arm (the crews themselves may view a Cruiser as a better posting than old Beachhead Belinda; but the admirals and the politicians will want their APs every bit as much as they want their BBs).
I also favour this because it means that the taking of a world will invariably take TIME. Time in which you need to bring reinforcements and supplies in and out ... time in which you need to protect those convoys, and keep a protective force of fighting ships in the system to cover these movements of transport ships ... time, in short, in which the defenders can muster a relief effort, and space combat can take place.
In TCS, you've got to predict where the enemy will strike, and have your defenders in place and waiting for him. I don't see it panning out like this at all. I see a typical planetary assault involving an initial assault phase when the attackers are likely to enjoy overwhelming local superiority, followed by a consolidation phase when the invading forces need to dig in and hold their ground against the inevitable counter-attack whilst the Navy attempts to deliver the troop surge necessary for the army to go over onto the offensive. During this phase, the defenders attempt either to retake control of the space around the disputed world, to prevent the fresh troops and supplies arriving; or they focus their efforts on the support structures and logistics chain necessary to deliver the supplies and troops. Disrupt that, and even though the attacker enjoys control of the system space, his troops may not be able to achieve control of the planetary ground, and a pull-out may be necessary.
I think of the war in North Africa, 1940 - 1943, as providing a classic example of this sort of conditions (even though the rival armies were both, initially, present on the same landmass and could both march across each other's borders). The balance of power shifted dramatically as the abilities of each side to support and supply their armies by sea varied with the varying fortunes of the Mediterranean naval war.
The other model for planetary "invasion" in such circumstances would be to provide covert support, assistance, and encouragement to a rebel movement, and then seeking to make them your "puppets" by making them dependent upon your continuing practical assistance - but this requires that there be a rebel movement which is powerful enough to have a reasonable prospect of overthrowing the incumbent government in the first place. In this model, however, the naval battles will take the form of your navy seeking to interdict the incumbent regime's attempts to call in material assistance from elsewhere. So now it is more likely that your forces will be the raiders, and the enemy will be escorting convoys in.
I am currently trying to come up with a set of abstractions which will enable the planetary conflicts to be played out in a suitably simplified form to enable naval campaign games to be played around this concept of interplanetary warfare - but so far everything I have devised is either unsatisfactory in its operation, or too cumbersome to be practical. However, I am working on it.
Once I have those rules figured out, I will be able to back-project to the implications for Navy support requirements, and what that means in terms of the balance between fighting ships, transport ships, communication ships, support and resupply ships and other specialized Navy functions. I already have a pretty shrewd idea of what it is going to mean ... but I am not fully there yet.
