In short: The player must assign its forces to defend one of 8 flanks of the city square as a priority.

The idea is to allow a player to use his defensive units in a more strategic way, by setting them in different "guarding posts".

In reality, a city defense can be enhanced using its natural barriers. While playing, these natural barriers take only two of its forms: movement restrictions, and defensive bonus. However, in reality, natural barriers have a large influence in combat, and yet not only at the time of battle, but also when generals predict their strategy and the opponent´s.

Armies, civilizations, cities and people have limited resources. This reflects in battle, obviously because armies have different sizes, abilities, and weapons. Nonetheless, it is true also that smaller, weaker and more lightly armed armies have defeated their seemingly superior invaders, all because of prediction and strategic preparation.

Every city square is surrounded by another eight. Logically, if I`m surrounded by enemies allover, then it`s reasonable for me to divide whatever force I have in every single direction from which I may expect an attack. Otherwise, doing that would be rather naive.

If I own a coastal city, and four of my surrounding squares are water, why would I waste part of my strength to defend a post that´s not going to be attacked (assuming there aren´t marines yet). On the contrary, if for some reason I expect the enemy to drive his attack from one square only, then I should comfortably risk being attacked from the rearguard to try and protect that single square, with all of my resources. This would imply that my predicting ability pays off.

This can be achieved in the same way that the defensive bonus works. If I have 8 surrounding squares, a valid prediction adds a given percentage to my defensive strength, while a wrong one does the opposite (by the time the reacting army crosses the city, the enemy has already set it in flames... too late). Also, a not so accurate prediction could be somewhere in the middle. The player must assign its forces to defend one of 8 flanks as a priority. Here´s what I propose:

Perfect prediction (the exact square): adds 4/8 (50%) defensive strength.
Wrong by an eighth: adds 2/8 (25%)
Wrong by a quarter: No addition, no subtraction
No prediction (which is the current state): No addition, no subtraction
Wrong by three quarters: subtracts 2/8 (25%)
Completely misguided army: subtracts 4/8 (50%)

What do you say?

Quite an impressive first post, Reality Related, and welcome to CFC! [party]

While this idea is simple and intuitive, the primary concern is how the interface will implement the player's required actions without significantly increasing micromanagement. Done incorrectly, a player would have to constantly adjust city defenses and end up spending even more time on managing the military, when most of the game time is already spent on moving units.

So, the next step would be to consider some type of interface that would minimize micromanagement. My first suggestion would be some easily-distinguishable graphic that allows players to see at a glance all the current positions/postures.

That sounds awfully good to me. If you think about it, there could even be a sign above each city square that states easily where is the defense posted, and the player could move it without even going into the city screen.

Second, this issue could also be managed by the city governor (if the player considers it convenient).

Third, this is most useful when you can expect somebody to attack you (after all, we´re talking about prediction). Thus, if you`re enjoying a peaceful time, there could be an option on the F1 screen that restores every army posted inside a city squere back into the center position (which is the default state for defending armies, where no addition nor subtraction takes place).

Still, your concern is most valuable. War is a huge part of the game; i`d like to see some more Economics take place in the future.

Well, the idea of having a better defence strategy sound cool to me, but I am not very happy with the method proposed. But before I continue, welcome to CFC!! [party] :dance: :band: :banana:

The problem with the method is that it is usually difficult to figure out the square from which the siege shall be carried out, and units can change and attack the city in the same turn, in some cases.

Indeed it´s difficult to predict where the attack will come from; however, you can be entirely wrong or just partially, with proportional comsequences.

Also, because of some units´enhanced mobility, there should be some armies posted at different spots, being that the effect of the enemy applying the "Divide and conquer" principle. You will always have the option of stationing your armies in the center of the city, with no effect whatsoever, but if you have a strong intuition, know your enemy and its possibilities, you can also take advantage from it.

Still, what is most appealing for me is the idea of allowing a player to improve his defense by strategy; the method is secondary.

The best method would probably be a penalty to a units defense if there are enemy units on multiple sides of it. Because you assume the generals can(and will) re position their armies within the space of a turn, although not in the course of a battle.

If forces are in only one square in that turn, then the defenders are as normal, if forces are in two adjacent squares ~10% penalty, if forces are in two nonadjacent squares ~20% penalty, if forces are in three or more squares that are non adjacent to each other, then ~40% penalty

so using the number pad

87,89,84,86,896=10% (partially flanked)
81,83,123,841=20% (fully flanked)
813,7913,1236987=40% (surrounded)

In other words you assume the generals are doing what is best.. your business as a player is to ensure those units don't get into flanking positions.

Krikkitone's idea would actually be well-suited to a macromanagement role of the Civ player that would minimize micromanagement. But this system only has penalties for strategic errors (such as placing units in flanking positions) and no reward for strategic planning/intuition (such as "predicting" that an attack will come from a certain direction, as in Reality Related's original idea). Might there be a way to balance that out?