@larienna
If you like. I could focus entirely on one of your games in progress. That you wish to use this mechanic on.
I could work on an excel sheet that calculates the balance.
Make it more functional as well. So that changing numbers would be easier.
I am thinking of a pre-matrix in which you enter the number of dice used against a target. Then another pre-matrix in which you enter the number of dice used in a counter attack. (2 pre-matrixes since this is simpler to work with)
Heck, I can even throw in an extra column for a basic set of dice.
Edit:
The only issue is that I cannot calculate a winchance. But I could have the excel sheet look it up from a separate tab.
One mechanic I am considering is 1 card vs 1 card, where each card could be associated to 2 different unit. It's a bit more complicated to calculate the contribution of a unit, but I think its still possible.
For example, each card could be something like:
If unit A and B is not present: Do effect W
if unit A is present: Do effect X
if unit B is present: Do effect Y
If unit A and B is present: Do effect Z
Or something similar. The inspiration is Starcraft the board game, In that game, you had stats if unit A was present or missing. I though I could put 2 unit combos on the card to promote natural synergies within the game.
Else it must be group battles and its much more complicated to evaluate the contribution of a unit. So I guess the main restriction criteria for this study is the one vs one dueling. Systems like Axis and allies makes it AI unfriendly because you need to manage targeting of damage.
The objective is to have different unit configuration, with strength and weakness. In order for this to work, you need a certain amount of complexity. Rolling the dice is not the slowest portion of the combat system. Matching units is the slowest part, especially if cards are chosen by players (instead of random with my variant)
I am not sure if there are better "duelist selection" mechanism that could be faster. The War card game could be a solution, flip the top card, roll and resolve. But that would work great only if you have many copy of the same card. It also makes it more AI friendly, he just draws the top card.
Which makes me realize that many multi unit system could be converted this way. For example, in my Dracula conquers Europe game, I wanted different types of unit, but was complicated to handle on the board. Now I could use a deck of card for the unit for your entire army, and simply use colored cube for the quantity of units.
As for why my variant, there are various reason. The action system was weird, some action were shared like in Puerto Rico, others was specific to the player. I decided to make them all player specific because many actions would make no sense to be shared. I also improved how trading works as a 2nd way to collect resources than gather.
The combat rules where improved to avoid the RPS mind game, allow retreat to reduce army annihilation which are hard to rebuild. The victory system has also been improved, victory cubes are assigned by the game, The only flaw is the wonder point reward, I could not figure out how to calculate this because its very sensible to the number of players. So a quick fix, is you win if you make the wonder. Else I made new buildings to power up certain actions ignored by the buildings.
So its really a melting pot of improvements selected mostly from community suggestions.