If I can venture an opinion, I think that the basic problem you’re perhaps running into is the following one. You’re starting from the quite logical premise that you want to keep the game as recognizable as possible, in order to give your proposed revisions to the game as much broad acceptability as possible. That’s a perfectly reasonable and understandable premise, but it comes with a built-in limitation: by thinking about potential change X to element Y of the game, and then analyzing whether this concept is viable, you sometimes discover that it won’t work because of the constraints imposed by element Z of the game.
A case in point. In the real world, WWII battleships and cruisers were rather similar in design and function: fast, armoured surface-combat vessels armed with a mixture of heavy guns and of intermediate and light weapons. Battleships had more firepower and more protection, but they were a lot more expensive so you could buy fewer of them for the same amount of money. Cruisers had less firepower and less protection, but in some cases they were somewhat faster than battleships; their main advantage was they they were cheaper, so you could buy more of them for the same amount of money. In some situations, having two cruisers might be more advantageous than one battleship, even thought battleships were more capable; for instance, the best battleship in the world, no matter how good it is, can only be in one place at one time, whereas two cruisers can operate either together or separately (potentially even on opposite sides of the planet, which is handy when one is fighting a war on a global scale). The obvious way to model this in A&A, if we take nothing else under consideration, is to give battleships and cruisers the same types of capabilities, to give more of those capabilities to battleships and fewer to cruisers, and to give battleships a higher sticker price and cruisers a lower one. This is simple in principle, but when this sort of thing gets proposed in house rule discussions people often respond with the “element Z” argument: “We can’t do that because it would unbalance such-and-such an element of the game.”
In the case of cruisers, as I recall, the “element Z” argument has tended to be, “No, we can’t fix the cruiser problem by simply making it cheaper because unit price changes have such-and-such an unbalancing effect, and therefore the only acceptable way to fix he cruiser problem is to change the capabilities of cruisers.” Well, that’s one way of looking at the problem…but it’s based on the assumption that element Z has to stay the way it is. It doesn’t. If element Z prevents change X from being done to element Y of the game, then why not change element Z too in order to break out of its limitations?
The answer, of course, is that there’s potentially no end to this chain reaction, and that one could potentially end up changing a hundred different things in the game, by which point it would become unrecognizable, and therefore impossible to market (as a house rule modification) to other players. I see the point of this argument, and I’m not saying it’s wrong. All I’m saying is that, speaking in very general terms, I think that there are three basic approaches to house-ruling A&A (which I’ll describe in a moment), and I think that the middle-of-the-road approach (option 2, in what follows) is perhaps the worst of both worlds. Option 1, the conservative option, is to simply change a few small things here and there. Option 3, the radical approach, is to throw out the entire rulebook and start from scratch: keep the sculpts, keep the map board (or produce a custom version), keep the poker chips and the base markers and the national identification markers and so forth, but tear up the rulebook and start with the proverbial blank sheet of paper (or, in modern terms, a blank word processor document). This gets rid in one shot of all the annoying things about the OOB game that people complain about, all the element Z rule constraints that get in the way of revising other rules, all the arguably ridiculous abstractions and clunky devices that have made their way into the OOB rules over the years in order to “simulate” (often very imperfectly) various aspects of WWII that in real life didn’t at all work in the same way as they do in the game.
Yes, I agree that this is quite extreme. It would only really work for someone who was genuinely prepared to see what kind of game would emerge from the subsequent design process, and who didn’t have preconceived notions of what would or would not be acceptable. So let’s call that one option 3a, and consider a somewhat less radical option 3b. In this option, one would print out (single-sided) a copy of the G40 rulebook (and perhaps a few other A&A rulebooks too), take out a pair of scissors and start reading. Each rule element (or section, or whatever) would be read, carefully cut out, and stacked into one of three piles: rule elements that I love, rule elements that I hate, and rule elements I’m ambivalent about. Elements can even be imported from other A&A games, if one likes. Once the work is done, use the “rule elements that I love” pile as the starting point for your redesign. It’s not as intimidating (or as much work) as starting from scratch, and it provides a familiar base of OOB elements from which to start. See how well these elements work together; if some don’t work well, revise or replace them until the y do. As needed, use your “rule elements I’m ambivalent about” as a source of ideas for plugging holes or providing potential solutions to problems. Use your “rule elements that I hate” pile as a guide to tell you what to avoid in your redesign. As in option 1a, the main advantage here is that you have lots of maneuvering room to redesign the game to your liking, without running into those annoying element Z rule constraints I keep mentioning.
As for option 2? Basically, it involves using the existing OOB rules as your starting point, but deviating more and more from the OOB rules by changing element after element after element on a large scale. And I think it has the danger of being the worst of the three options because in aims to produce a major redesign of the game (unlike option 1, which involve light tinkering) while staying within the constraints of the full set of existing rules (unlike options 3a and 3b, which either toss out the OOB rules entirely or start out with a highly selective subset of those rules).
Just some ideas. Possibly not of any practical use, but if they stimulate some thought then they were worth the typing.
