From: Iowan in MD/DC
Unfortunately the AI for both sides is pretty bad. If order to have a really good game againist eithr side's AI, you need to play along historical lines as much as possible. For the people played Allied, this means do not try and defend forward at all in '41 and pretty much leave the areas in the DEI and elsewhere in place. I.E. do not reinforce or add extra AC into these areas. As a Japan player, push into China so that a land connection exists between Singapore and Shanghi and capture Sain. Then stop. Also, do not take extra units out of Manchuria or China, even by paying PP. It also helps both sides AIs to play with PDU off to minimize the impact accelerated R&D for Japan can have.
In other words, either side's AI is easy to 'break' if that is what the player wants to do. If you want a more robust Allied AI, then don't invade Pearl or other things you can do as Japan. Also by playing along more historical lines, allows you as Japan to at least have a feel for what will happen as the war goes on in the later stages :).
In my Japan game against the Allied AI (which I used as the basis of this Primer), I did not do the above and have completely overrun China . So when '43 hits I will have an AV (unless something goes horrorably wrong between now and then). I do have PDU off though. If I had wanted I could have not done that so the game would have gone longer. The Allies have made an effective landing at Normea and kicked me out. So the good news is that AI does seem capable of invading properly
So you can have an interesting game as long as you try and play as if the AI is an actual person and don't try to do things just because you know the AI can't handle it.
Somebody said in a thread the other day/week that you can still continue playing even if autovictory is reached, "to see how it turns out" so to speak... I am working to verify this in my AI game, but I waited too long to conquer China and there's 15k AV sitting in Chungking - after I destroyed 6k on the road there! Even by nearly emptying Manchuria I don't have the AV to combat this yet.
I'm reading through this and it's good basic info. If you can insert links in your PDF, perhaps some "Further Reading" links to forum threads/discussions that go into more detail on nuances and trade-offs? Like why Tojo is the answer rather than Tony, etc. Not strictly production-related, but production is symbiotic with your combat strategy so it's relevant.