Kusari-Gama work on entirely different principles than flails. Also, that is a poor example of a Kusari-Gama.
The point irmo is trying to make is that while *HUMANS* can (in theory, with proper knowledge) create weapon designs; a computer that lacks a detailed physics simulator that takes fighting styles, the physical forms of both the attacker and defender, and the historically prevalent attacks and defenses of the setting into account.
Procedurally generated weapons would be fine; the problem is that we already have procedurally generated weapons in real life--lo and behold, they're remarkably similar across different geographic and cultural boundaries, with allowances made for materials and culturally grounded styles of fighting.
There are always *reasons* for things.
An axe for example, seeks to focus power onto a relatively small and sharp area, while providing mass behind the strike to enhance momentum.
A flail, on the other hand, uses the length of rope or chain as an enhancer to the lever force, and the benefit to be gained from having an axe-head shape would work counter to the efficacy of the weapon.
Your example, a variant Kusari-Gama, is not just "an axe blade on the end of a chain"--it has some very specific purposes as well. First, the weight can be used in much the same way as a flail. The blade (note that it is clearly designed for stabbing (i.e. can be thrown a short distance to enhance reach) but the hook portion is designed such that the bladed section can also be held and used as a stabbing and cutting weapon at close range while leaving the weight free. The chain also has a purpose, it can be used to trap weapons and limbs. The entire point of a kusari-gama is that a skilled user can continuously manipulate the battle situation in unpredictable ways. The weapon itself is actually less effective, mechanically speaking, than a spear thrust or a flail strike. A traditional Kusari-Gama also had the advantages of being easily concealable and easily constructed with commonly available materials.
Also, a simple axe blade on the end of a chain would lack the follow-through that the haft would normally provide, which would further weaken its efficacy as a weapon. The reason flails work is that they don't NEED follow through, they just get the head of the flail up to a very high speed and strike sharply. Enough of that though.
So the real trick is for the combat simulation to be sophisticated enough to recognize and exploit these sorts of things. As it is, reach is ignored, skill is mainly a matter of one or two skills vs one or two skills, whereas in real life it would be more accurate to have 40+ vs 40+ fluctuating constantly as to which were used at any given time. Once the combat simulation can do that, then it also needs not only to be able to evaluate the relative merits of altering a basic weapon design but also to do it within the context of a specific tradition of warfare and perhaps even up to the next level, considering more than one at a time.
Nobody disagrees with you that it would be awesome Anty--it's a question of figuring out ways to do it properly so it does not become a ridiculous, exploitable, kludgy mess.