I was thinking about it and, I'd REALLY like to play a single-player strategy campaign where I breed units in between RTS battles. But I'm not sure how the breeding could fit into a multiplayer version (it's like the problem with balancing pokemon duels when players haven't completed the single player game, including maxing the level on their dueling team). As for putting the breeding right into the RTS battle, I can't see how breeding, which is a though-requiring activity, would fit into a session where one is playing as fast as possible.

Give your players a daily opportunity to recruit randomly generated offspring from the pool of fighters. Your warriors have a collective gene pool. No micromanaging. Permadeath ensures only the strongest live to pass on their genes.

I'm going with an additional gene pool for the ruling class where you match up mates and marry off daughters to other players. Creates a more hands on breeding, but is limited in scope so as not to overwhelm

Assuming I ever get away from EVE my own project will have one game in the series based on breeding. I call my games exploratory strategy games because people can't get over the fact that RTS doesn't mean 30 minute multiplayer matches with 3 asymmetric races.

My current game has traits for units but its more like a deeper Majesty than a focused breeding game. Population randomly gains stats based on their socio-economic class as defined by housing levels and available food and services. My breeding game will sort of work like dragon warrior monsters/monster rancher but a little more involved. Most of my games involve playtimes similar to Paradox rather than starcraft to account for increased complexity.

Did you play the 1996 Bullfrog game "Genewars" ?

http://en.wikipedia.org/wiki/Genewars

It had some similar ideas.

If it's still feasible, get a copy :)

If you're going to make a game focused on taking care of your units you may want to consider replacing the fighting/mining with accepting missions and sending units out to complete them, and turn the game in a sort of military-base simulator.

I don't see how this is any different than a normal RTS.

The way you build units is the same: out of a building. It is still a queue construction unit type mechanic. Homes? Call them barracks Attributes? Call it a personalizer random generator and bam, same thing. Food count? Supply count. Homes with more powerful breeders? Factories. Units getting more powerful? Promotion.

Nothing is unique about this concept except maybe your personalizer random generator.

Cheap units vs expensive units? Marines vs Tanks? This isn't any different than most RTS out there. All it is, is dressed up in different aesthetics.

What happens is this design will merely create a locked in system where people will crunch numbers and create the most efficient unit type and how to perfectly balance the economy with it. So in the end, your game mechanic idea is merely reduced to a science of numbers.

I'd be more concerning with making a fun RTS with good units, controls, and pace before I go that deep. RTS is hard enough following the proven formula.

What happens is this design will merely create a locked in system where people will crunch numbers and create the most efficient unit type and how to perfectly balance the economy with it. So in the end, your game mechanic idea is merely reduced to a science of numbers.

This also goes for every RTS out there though.

But i agree with the rest of Canyouseeitnow's post, this is why i recommended removing the enemy from the map and focus on a (complicated) base,

making it a sort of Theme Hospital meets Age of Wonders.

enemies can still appear on the map and test the base's defenses/security, but have them appear in the form of thieves, infiltrators, bandits/pillagers,

maybe someone looking for revenge on a particular unit etc.