After sitting down and thinking over Bill's comments for a while, it occurs to me that what he identifies as a problem with the proportional nature of the AEC system may actually stem from a much deeper problem with current manual military simulations and has to do with quantification of combat ability.

To reduce the calculations necessary prior to combat resolution, we (all of us involved with game design) have accepted the odds ratio system or something very similar to it. The result of this is a type of behavior among gainers that III call "factor fussing" Factor fussing occurs when you're one factor short of a 3:1 attack and where the difference between a 2:1 and a 3:1 attack is immense.

At that point you fuss around until you can get an extra factor into the battle and get your 3:1. 1 would contend that the difficulty that Bill identifies in the AEC system does not stem from any irrationality in the system but rather stems from the inconsistency inherent in a discrete odds readout CRT which mandates factor fussing for optimal play.

I would argue that some of the examples that Bill gives are more examples of rational game play than rational military tactics. For example, assume that an armored corps of two armored divisions and one mechanized division is launching an attack. An infantry division is available to support the attack. Will it necessarily be to the advantage of the attacker to use it?

No, because the use of the infantry will detract from the ability of the mobile formations to conduct a mobile attack. Either they must slow their rate of advance (thus losing much of the benefit of their mobility) or they can forget about the infantry and operate separately (thus reducing the contribution of the infantry to nil). Either you use the strength of the infantry or the mobility of the armor; it's hard to get both.

As to Bill's suggested modification, I see some problems. Primary among these is the fact that this system rewards armor twice in large numbers and underrates it when alone. For example, suppose that an armored division were assigned a modifier of +1. One armored division attacking alone would receive a modification of +1, while three divisions attacking together would receive a modification of +3, even though both attacks have the same ratio of armor to the total force. Such an incremental system could as easily be handled by increasing the combat factors of armored units; furthermore, it ignores the fact that the AEC rationale is fundamentaly proportional rather than incremental in nature.

In my mind, the real problem that needs to be tackled is how to remove the factor fussing inherent in a straight odds table. I have heard only one suggestion along those lines which makes sense to me, from a friend of mine named Alan Monson. It goes something like this:

Suppose the defender has six factors. You are attacking him with ten factors. This gives you a 1:1 attack, but you're really closer to 2:1.

As a matter of fact, you're four-sixths of the way there. Therefore, roll one die before the attack is resolved: if the result is a 1 or 2, read the final die roll on the 1:1 table. If a 3 through 6 is rolled, use the 2:1 table. (Notice that two separate die rolls are involved.) Thus, your position between the odds levels is accurately reflected, and adding two more factors will not have results completely out of proportion to their incremental addition to the attack.

Obviously, this will only work with a progressive CRT, but any game which uses die roll modification will generally have one. Although this system will require a little more math immediately before combat resolution, the play of the game often speeds up as less time is spent in endless counting and recounting of factors. The end result is a more common-sense approach, which I think both Bill and I are aiming at.

More DNO/UNT Armor Effects

Armor Effects in DNO/UNT
My Turn: Designer Rebuttal
The Last Word: Rebuttal's Rebuttal