I think the idea of rewarding fighting with specific troops has some potential.
The way it was implemented doesn't work for me.
To make it work the limit should not be on which barracks we can build during an incursion, but on how many different barracks we can have
at the same time in the settlmenet.
The problem is that most troop and troop combination cannot be used without some KIA, even with moderate bonuses. And this is not susteinable when a node require specific troops we do not produce.
So the luck of the draw is affecting the experience.
In IA (but this is valid for almost ages) there are some suggested troop combination that work well (8 archers and 8 ballista). So if a player choose one of those he can do all battles with little problem.
But since the random generator is faulty, it's easy to get incursion with higher or lower percentage of nodes allowing those combination.
This can be very frustrating if we get a lot of nodes like this:
BTW in the picture above you chouldn't tell what type of node I was on even if I incuded the whole window. The informaiton is NOT in the window.
Is it a defense or an attak army? Not visible in node, the informaiton is in the colour of the node in the map.
The increased action cost didn't help. But the more frustrating is not the action to do something, just the WASTE of action required to move around. That discorages exploring, less fun, with no added benefit in actually limiting progress. It should be changed to something more like 200 (still a small drain, but doesn't affect playing in incursion).
I still believe that allowing players to take multiple nodes at the same time, even without branching would be
more fun (like before the bug fix in first incursion), as long as all mandatory nodes are taken before boss can be defeated.
Not knowing the medal cost for the expansion is harming this feature. Data frombeta should not be used to set those costs! Beta data on player activity is suspect and skewed.