In a historical RTS game, I have different units with strengths and weaknesses which mimics rock paper scissor.

• Warriors kill pikemen
• Pikemen kill cavalry
• Cavalry kills archers, may also be effective against warriors
• Archers kill foot soldiers

It just feels a little too simple. Just choose the right counter for the right target to win. As I'm sure there's more to real warfare than rock paper scissors, how would one deepen the tactical aspect of a RTS game?

Situational advantage's and combined arms effects can go a long way to adding depth and strategy to a game. IE:

-Cavalry can rule open fields against everything, but require flanking advantage to win against pikes. Heavy lancers rule against light cavalry, but are more costly and not as effective at running down broken infantry or at scouting. 

-Pikes can hold a position effectively, but are less effective at advancing. 

-Heavy infantry are resistant to ranged weapons, but can't close on enemies as effectively as lighter troops. 

-etc.

Things like these, rather than straight up Rock-Papers-Scissors win/loss mechanics, add up to more emergent game play. Pulling an enemy's heavy cavalry away with a unit of light cavalry, and drawing them into a wooded trap of pikes and archers, but they may be vulnerable to a push by enemy light infantry driving into the woods. (Where both pike and ranged units have negative debuffs to their effectiveness.)

Well, RTS has some important aspects.

Unit speed, unit facing, unit viewrange.  I also don't know if you want to model morale as a separate mechanic.  (In some ways, Morale can be rolled into HP, as they function similarly.

Also terrain bonuses, and what units can move through what terrain at differing speeds.

A lot also depends on how much of a focus of your RTS do you want to be on the combat/tactics over the economic stuff.   If one can't build units during a battle, that's pretty big.

And, my personal fav for RTS tactical battles is Kohan.  Even the basic units, the combat can get pretty deep, with morale, flanking, entrenchment, combat stances (forced march, full attack, etc) and terrain bonuses.

It sounds like you have them all acting independently, like most RTS games do.

A very simple thing you could do is have them get into a simple line formation 3 or 4 ranks deep.  Not a set thing, but whenever you lasso a group they automatically form into a line/box and "act as one", as Russel Crowe would say.  This way it also becomes about the composition of the force, so it doesn't feel as though it is entirely the individual units.

When the fight happens it will be a literal "melee" with your AI telling your units an order of preference of which units they should engage if more than one are next too them.  And it will just work out to be the kind of "jumbled mess" you see in Braveheart.

You have listed pretty much how it works in total war games, which i think is fairly accurate on how these medieval battles could go down.

If you have production of units in your game, you can play with factors like costs, production time, tech needed and specialization vs generalists:

Pikemen (specialist): very strong vs cavalry, vs weak against archers
Spearmen (middle ground): strong vs cavalry, ok vs archers since they have shields.
Armoured spears (generalists): expensive, so only ok vs cavalry  but better against other infantrymen and archers.

This depends on how detailed you want to make it. Remember more choices can muddy up the actual strategy a player can use/be aware of. Too many choices and players start to choose "middle ground options" since it's too messy to keep up with everything.

Homeworld (a fantastic game) had a weak point: to many unit choices. Something like 4 unit classes and each had a counter against another class (so 16 different "unit groups"). It lessens the tactical control a player can have.

On 21/11/2017 at 6:46 PM, Luckless said:

Situational advantage's and combined arms effects can go a long way to adding depth and strategy to a game.

This makes a lot of sense. I could add some kind of artificial bonuses to certain combinations of arms so that it's easier for the AI.

On 21/11/2017 at 6:50 PM, ferrous said:

I also don't know if you want to model morale as a separate mechanic.  (In some ways, Morale can be rolled into HP, as they function similarly.

I want to simulate morale separately from hp. It will be similar to total war, but instead of the whole unit running away, I want to have individual soldiers break from the unit.

On 21/11/2017 at 6:50 PM, ferrous said:

A lot also depends on how much of a focus of your RTS do you want to be on the combat/tactics over the economic stuff.   If one can't build units during a battle, that's pretty big.

There will be some kind of logistics management in battle, but nothing like a full economy. You can spawn more units during a battle, but only if those were built before hand (in a campaign mode).

On 21/11/2017 at 9:09 PM, Kavik Kang said:

A very simple thing you could do is have them get into a simple line formation 3 or 4 ranks deep.  Not a set thing, but whenever you lasso a group they automatically form into a line/box and "act as one", as Russel Crowe would say.  This way it also becomes about the composition of the force, so it doesn't feel as though it is entirely the individual units.

Should grouped units improve their battle effectiveness? I understand putting them in ranks and group so that they make a better melee, but how do you make it so that it affects the tactical play? Games like Cossacks make organized groups worse because they have a lesser DPS than a huge mass of unorganized units. How do you make that organised units have the advantage?

On 28/11/2017 at 6:30 AM, suliman said:

This depends on how detailed you want to make it. Remember more choices can muddy up the actual strategy a player can use/be aware of. Too many choices and players start to choose "middle ground options" since it's too messy to keep up with everything.

That's a very interesting point. As much as I want to add choices, I hate when units don't have specific enough roles that I can't tell which one to use.

I was only thinking in terms of a very simple indie game.  You had been comparing individual units and wondering how to make it more interesting than "A beats B and B beats C".  A very simple way of doing that is to just group the units into a mob so that "unit composition" becomes another factor.  That's all I was saying.  Obviously, if you are thinking in bigger terms than a prototype/small indie game there is much more you can do.

5 hours ago, Kavik Kang said:

A very simple way of doing that is to just group the units into a mob so that "unit composition" becomes another factor. 

Sorry, I didn't wrote my question very well. When you say that unit composition becomes another factor, how do you implement this factor? Is it an extra statistic that is added into the combat simulation?

5 hours ago, Kavik Kang said:

Obviously, if you are thinking in bigger terms than a prototype/small indie game there is much more you can do.

Ideally, I'd like to go further than just a prototype/small indie game.

Ok, I was thinking in very simple terms because the original post was comparing single units too each other.  In a very simple way, just forming rock, paper, scissors units into mods too fight would just naturally wind up creating random matchups between them.  At a very simple level, each unit could have an order of precedence of what other types of units it will attack when more than one are next too it so that they would tend to attack the units they are meant to kill.

As I recently mentioned in another thread, learning Avalon Hill's Advanced Squad Leader really is the best thing to do if you want ideas for how to make an RTS game.  A lot of WWII veterans and real-world military men spent 50 years arriving at ASL.  It is second only to SFB in complexity and how massive it is.  D&D has a larger amount of material than both combined, but it's almost entirely story.  A good example of something I think would translate from ASL into RTS games very well is the AI. 

There is an AI that is far more sophisticated than anything you've ever seen in an RTS game before sitting within ASL.  The phased turns and sequence of play alone will teach you a lot about how, tactically, things should be working.  The order everything happens in, when units have an opportunity to fire and why, things like that.  Advanced Squad Leader will stun modern gamers as much as Star Fleet Battles will.  The games of the "Big Three" era are from a unique time in the gaming world, they are massive beyond the imagination of modern gamers and were in development for half-of-a-century.  They are also both extremely detailed, and were both made by real-world military soldiers and in many cases veterans of actual wars.

Their is a whole lot of amazing stuff in ASL for anyone who makes RTS games.