Yeah, but would that be so bad? In fact I think it is how it should be.
If you attack with much greater numbers, you should win quite easily.
No problem there, we should still get enough "even" fights, so that you wouldn't be tempted just to use autoresolve all the time (which usually gives me better and more sensible results than playing myself, btw.)
It's kind of an odd thing the relationship between the campaign map units and the same units in RTS gameplay. They hold much greater significance in campaign mode but little in RTS mode considering how easily they die.
Part of the problem right now is the maps. Most conflicts are 1vs1 on a 4 player map. Not a problem on an individual basis but it prevents small scale conflicts. It seems silly when you move in 5 troopers against an enemies 3 grenadiers on the campaign map, that the battle could potentially turn in to a full on war with devastators, balloons, ironclads, and imps.
Obviously it's meant to be representative but I think the map should scale based on the size of the initial conflict to some degree. Excluding capitals or heavily fortified map locations.
So if you are attacking with that trooper vs grenadier example. Since that's a relatively small conflict that should bear out in the RTS mode just as it did the campaign mode. It should throw you on a smaller map with less resource points and building sites. If you bring the house or a truck load of card support then the game selects a larger map with more sites.
As the game goes on those larger battles will be more prevalent. Just tie it to the cost of the troops initiating conflict. 2-20 gold worth of units, very small map few points. 21-50, medium. 51+ large. Or something to that effect.
Just an idea anyway.
The populations were dropped a couple betas ago, which effectively does make units more expensive.
Only over the long haul if I understand how population works. It doesn't affect opportunity cost in terms of how quickly you build more advanced troops or how quickly you can get huge armies on the field.