If you consider any given battle to be “game-breaking”, then you are taking excessive risks and using the ability to concede the game and start over as a justification for those risks. Simply make more conservative attacks. It is an assumption we make that Germany can attack 3 sea fleets, take paris (and even attack Russia) all on turn 1. That isn’t conservative, but its always possible. Same with the Tobruk/Taranto complex…
I don’t know of any T1 or T2 battle where I just throw up my hands if the luck goes completely against me. If Japan loses all of its ground troops in the Yunnan battle, then well, it can make up for that with the FIC factory within 1 turn (if it took and secured it). I also don’t always counterattack there. Sometimes, its better for china to run like hell.
You’re right, India will fall. It always falls. That’s not about luck or even Allied choices (barring bringing all your air there) its more about how the game plays out. As a result, I have undercomitted to building the front line transports before and only shown up on J3 with not enough to win…so I have to wait until J4. I cant simply roll the dice and hope to get lucky, then quit because I didn’t wait until J4 or preplan better.
Maddog is dead on; I want my opponent to have to take risks and thereby lose. Usually making a certain, risky choice buys you 1 turn. You can do a battle on G4 with luck, or build up more and on G5 take it for certain.