Krikke100

Well, the next battle was one of the smaller ones with the description saying I'm facing Bragg (who fought at Shiloh), so I figure it's a Western battle. Fair enough that survivors regroup and such, but seeing an army composed entirely of three star units still seems a bit weird.

Ending the campaign on a major victory might not be ideal as you might miss out on half the campaign. But why not use some form of scaled scaling wherein the AI army grows with the player's, but the exact scale between the two depends on casualties on both sides. This way succes will lead to an easier time in the next battles, thereby increasing the odds of final victory, while failure will make things harder in the future. That way the result of a battle has effect on the campaign while still keeping some scaling for balance. eg: AI army is scaled to the player's at a ration between 80% and 120%. If the player wins battles with low casualties and heavy damage on the AI the scaling will trend towards 80%, while heavy player casualties or AI victories will trend the scaling towards 120%. Scaling tied to numbers of casualties might also add an extra tactical choice when choosing between large or small armies. Large deployed army means larger deployed AI army, which means higher overall casualties for that battle, which means a bigger swing on the total casualties and therefore the scaling itself. If the player faces a difficult battle he might choose to deploy less troops, thereby reducing the effect of a defeat on the overall campaign. Additionally it might be interesting to keep troop quality in mind too. After completely defeating the Confederates at Shiloh (with everything but their cavalry routed of the map after the first day last stand), in the next battle I find six 2500-man three-star brigades, I wonder where all these high-quality troops suddenly came from. I would expect high enemy casualties to lower the overall quality of their troops in future battles.