Skip to Content
 

How to balance units in a RTS like board game? Hierarchy VS Rock-Paper-Scissors

19 replies [Last post]
larienna
larienna's picture
Offline
Joined: 07/28/2008

It's for my Rats Craft solo game idea, based on Star Craft the board game and Forbidden Star. My objective is to design the game's units which is the core of the game's strategy. It's very complex as it needs to have a good rock-paper-scisor relationship between units. I'll try to get to the point of my problem.

The Situation

The units will be organized as 4 factions which each have 8 units. Divided into 2 groups: 4 infantry and 4 vehicles. It is expected that players unlock between 4-6 of those units during a game. There will be some dice roll to determine attack opportunity, but I'll skip that portion. Infantry target infantry and vehicle target vehicles (like in Star Wars Rebellion).

What I am looking for is to build a matrix of combat results between units, the resolution is deterministic. For example, I could have unit A that can kill unit C, E, F, H alone, it can also kill unit B, G, J if it has support from another unit. Technology card could also allow unit A to temporarily kill other units when a card is played.

So I need to determine the stats of each unit in order to build up that matrix. Originally, I was thinking a simple system like in magic the gathering where you compare the attacker's strength(ST) vs the defender's health(HT) to know if the kill can be done. Supporting unit would give +1 ST allowing using 2 units to kill a unit. Those ST and HT value would actually remain very low. There would be of course other abilities and stats, but that is the core of the system.

One of the problem is Star Craft board game is that the units were organized as a tech tree. Which means that the humans for example always started with the marines, but the battle cruiser was at the end of the tech tree. Yes they were powerful units, but you unlocked then late game and have very few turns to use them. I wanted to avoid that issue and allow players to unlock any unit at the start of the game to allow a diversity of unit composition to encourage replay value and diversify the strategy and counter strategies.

The problem

Since units are not organized as a tech tree anymore, there is no more unit hierarchy, therefore all should be equal and not necessarily stronger or weaker than other units. This has the consequence of making the ST vs HT system obsolete as units will all have almost the same value. Now I would need to have almost a true Rock-Paper-Scissor system but that seems to conflict with many other game mechanics.

Which makes me wonder if I should get back to unit hierarchy again? Should I use a flatter hierarchy?

One thing I though is to have no hierarchical dependencies, but have some units that are very expansive to purchase on turn 1 making it unlikely to use this unit at game start. You could have for example, 2 cheap ad 2 expensive infantry and vehicle.

Another idea is that each of the 8 units could have an upgraded version which is more expensive and has better stats. The problem is that I must make sure the possibility space is big enough to handle unit upgrades and technology cards at the same time. I could make unit upgrade only change stats and while technology cards gives new capabilities. Another issue with upgraded unit if is that I need additional unit tokens for those upgraded version.

By think about it, stronger units in Star Craft and forbidden star seems important because since there is a unit limit on each game space, stronger units allow putting more firepower into the same amount of space.

Else is there a board game that has an RPS mechanism where all units are almost equal?

X3M
X3M's picture
Offline
Joined: 10/28/2013
I don't know the board game

But i have seen somewhere similar with +1 st effects...
Maybe, it is smarter to have multipliers instead.

I think that if you use infantry and vehicles as attributes, you already can have 4 different units.
Then, if you want support units, you can easily have these extra st to the frontline unit who does the fighting.

Inf. Anti inf.
Inf. Anti inf. Support
Inf. Anti veh.
Inf. Anti veh. Support
Veh. Anti inf.
Veh. Anti inf. Support
Veh. Anti veh.
Veh. Anti veh. Support

If you want the support to give a bigger bonus than the actual fighting unit. You can break stalemates by changing the support.

X3M
X3M's picture
Offline
Joined: 10/28/2013
I don't know what combat mechanic you have in mind exactly

Because then I could give 1 or several options.

I am also curious if you had different factions in mind.

larienna
larienna's picture
Offline
Joined: 07/28/2008
Like I said, there are 4

Like I said, there are 4 faction with 8 units each.

You gave me an idea to explore, where the strategy could be about using unit combination and situation to stash +1 modifiers in order to exceed the target's defense. But I think it would work better as a hex game with terrain underneath.

Combat is very simple. Once units has moved into the same area, combat begins. You roll a die for each unit. You might have 2 pool of dice, one for infantry and one for vehicle then you assign your dices to the units you want.

  • 4-6 = Attack
  • 2-3 = Support
  • 1 = Control You can optionally, reduce the value of any die.

Without any special powers, the default behavior is:

Attack: You select a valid target (inf vs inf, veh vs veh), if ST > HT, destroyed. If a unit has no valid target, it's ST goes into splash damage pool.

Support: You give +1 ST to another attacking unit

Control: You double the amount of control points of the unit.

Once all the targeting is done. The opponent must spend your splash damage pool to destroy units of his choice.

Finally, sum up the control points of all surviving non-routed units. The player with the most control points wins the territory.

Some special abilities will rout units, which disables the units from further battle and ignore their control points. They can be rallied later. I am not sure if routing could be in the core rules like in "Battlemist". Maybe splash damage could rout units depending how you split the damage.


As you can see, the combat system around is not that complicated and seems pretty interesting. Still I need first the combat matrix. One bug I could foresee is that it could be more advantageous to place unit ST into the splash damage pool than giving a +1 ST. But you don't get to choose your target and I was thinking of having a separate HT for applying splash damage which could be higher.


Still, how things are going, I think I would need to have strong and weak units because I think it's the whole idea used in RTS.

Even in Age of Mythology the board game, you have a strong RPS relationship, but you also have strength increase over time.

I could try to flatten the hierarchy and remove dependencies to allow diversifying various unit combinations. But there will still be stronger and weaker units with maybe a light RPS bonus modifier. I think this is the only way for that type of game.

X3M
X3M's picture
Offline
Joined: 10/28/2013
Not sure if it helps

Well, most RTS use a RPS mechanic too. Mechanically or natural.
Do I understand correctly that it doesn't matter what type the unit is. They offer 1 die that is either anti infantry, or anti vehicle?

Thus you have I-aI, I-aV, V-aI and V-aV?

In that case. What if some units offer 2 dice?

Also, what if the ST produced in an attack is equal to the HT the enemy offers? But only when it is the correct target?
I think, the attack can be considered to be a hit in this regard.
If it is not the correct target, then +ST from the support becomes an option. If there is a difference there as well. You can simply assign the correct support like this.

ONE important difference with your idea is.... the wrong targets will not go to splash just yet!

A binairy system is what I think of here.

Wrong Splash 1 ST
Right Splash 2 ST
Wrong support +2 ST
Right support +4 ST
Wrong target 4 ST
Right target 8 ST
HT is always 8.

***

I don't know if you can get an idea of this though.
Anyway, here are 3 example rolls:

Each example has 12 dice. 6 roll an attack, 4 roll an support and 2 roll a double control point.
Control point only doubles regardless of what the die would target. In other words, the die doesn't take part in the attacks.

All 12 are linked to the wrong targets?
Wrong attack is 4 ST each. They need support. There are 6 of them.
Wrong support is +2 ST each. There are 4 of them.
And 2 extra control points are rolled.

Player A wrote:
4 + 2 + 2 = 8
4 + 2 + 2 = 8
4...1 wrong goes into splash
4...1 wrong goes into splash
4...1 wrong goes into splash
4...1 wrong goes into splash
4 wrong splash; 4x 1 is not enought for a kill.
2 kills
10 survivors? Any player has 12 control points.

All 12 are linked to the right targets?
Right target is 8 ST each. There are 6 of them. They don't need support.
There are 4 right supports...but they can't support this time.
And... 2 extra control points are rolled.

Player B wrote:
8
8
8
8
8
8
+ 4...1 right goes into splash
+ 4...1 right goes into splash
+ 4...1 right goes into splash
+ 4...1 right goes into splash
4 right splash; 4x 2 is enought for a kill.
7 kills
5 survivors? Any player has 7 control points.

All 12 are divided in 6 right and 6 wrong targets?
We have 3 right targets and 3 wrong targets in the attack.
The 3 wrong targets need support.
We have 2 right support, +4 each and 2 wrong support, +2 each.
And.... once again.... 2 extra control points are rolled.

Player C wrote:
8
8
8
4 + 4 = 8
4 + 4 = 8
4 + 2 + 2 = 8
6 kills
6 survivors? Any player has 8 control points.

If you want different HT, then this is possible. But... the correct weight value's are a bit harder.

X3M
X3M's picture
Offline
Joined: 10/28/2013
Only 2 types?

If you really are planning on having infantry being effective against only infantry and vehicles being effective against only vehicles.

I find it strange. But the system still can work.

I also thought about the dice you use and how to track.
I think, it can still be done that some units offer 2 dice instead of 1. Maybe even 3.
Of course, HT can still differ. And once you are set on a combat mechanic. The balance can be tested.

Soldiers with low HT and 3 dice would be targetted first. But there is an easy way (for me) to determine their value, compared to the more normal soldiers.

larienna
larienna's picture
Offline
Joined: 07/28/2008
I like the idea that if you

I like the idea that if you cannot follow the RPS, you need support. Either it's something I force, or you just lose a +1 opportunity forcing you to get it from somewhere else. Most of the time in SC the BG, there is 2 ways to get something, and that could be an option. It would also makes the RPS less intense to conpute. You could go with more power or taking advantage of the RPS.

I think the inf vs inf & veh vs veh could be an important mechanism. It avoids turning infantry into cannon fodder and force having a balance between both. Infantry would yield more control than vehicles, and flying units have little control value. Still the unit limit could complicate this mechanism (see below). Another option if to make inf and veh only part of the RPS. So if inf fights inf, it gets a +1.

There is no official Anti-infantry or anti-vehicle units. Because that would make splash damage becomes useless. But there could be some technology card that gives you new attacks which could cross over the other type of unit. Still I am considering that maybe support could cross over, like infantry supporting a vehicle.

One thing I forgot to mention is that the unit limit is either 2 or 4 units on each side. The attacker can bring 1 or 2 extra. The defender can have a base as an extra unit. I could modify the value of those limits if I want, because it should have impact the the targetting and plash damage.

Another thing I though is that to create a fake tech tree, I could make resources available for research determines which unit could get unlocked. According to my calculation, if there is 2 strength of units where the low units need 1 resources and the high unit needs 2 resources to unlock. The following results should be:

1 resource = 1 unit
2 resource = 2-3 unit
3 resource = 3-6 units
4 resource = all 8 units.

I consider that during a game, players should have access to 2 or 3 resources. If things goes very badly, you might get stuck with only 1 resource. Maybe I could put a resource off board to ensure you always get access to a resource.

I am considering the possibility to have an upgraded version of each unit with better stats and more expensive. But that would make 16 unit. Still I like the alternative way to powerup your army by focussing on the same few units if you are stuck with little research resource.

I guess I'll have to improvise a combat matrix for now and test the above.

X3M
X3M's picture
Offline
Joined: 10/28/2013
Always a chance with 1 vs 8 resources

I like the idea of 1 unit always having a chance to kill an enemy unit. While the enemy has a chance of returning 0 damage.
The chance should be reasonable. In fact, if the player has more than 1 resource, there should be more targets to hit. And thus the chance improves, but little.
Is this possible in you game?

If the 1 resource always dies. Players tend to give up very soon once the weight is shifted to one of the players.

larienna
larienna's picture
Offline
Joined: 07/28/2008
There is an idea I came with

There is an idea I came with today to create a kind of soft RPS and soft edge between infantry and vehicles. Each unit would have 2 properties with 2 values each. For the example, let just say:

Infantry VS Vehicle
Biological vs Mechanical

(I will probably use something else, but it's just an example)

Now let say that the average ST is 2 and the average HT is 3. Each attack will give a +1 VS Infantry OR vehicle and another +1 VS Biological OR Mechanical units.

So that if you match at least 1 property you can make the kill, else you'll need support from another unit. If you have both properties, then you can get a +2 allowing a weaker ST 1 unit to make the kill.

Using un-even ST and HT prevent the need to handle -1 penalty and count everything as +1 bonus.

I could extend the idea further and have multiple binary properties and make attacks give +1 bonus on 1 or more properties.

X3M
X3M's picture
Offline
Joined: 10/28/2013
That is a good idea

Indeed. And you can start mapping out some basics already.
Let's assume that the factions have a very decent mix. So every mix of attributes is being used in the game. You have 2x2 attributes, so there are 4 options here.

1 ST can be 1, 2, again 2, or 3 ST after the bonus.

4+ HT has 0% of the 1 ST as direct enemy.
3 HT has 25% of the 1 ST as direct enemy.
2 HT has 75% of the 1 ST as direct enemy.
1 HT has 100% of the 1 ST as direct enemy.

The whole table moves up 1 spot with a +1 support.
In case of the 2 ST and 3 HT. The 25% goes up to 75%.
These aren't chances, but parts of the enemies that are able to do this.

You mentioned uneven ST and HT. Are you talking about balance value's? Or are you talking about alternating between ST and HT when designing the units?

larienna
larienna's picture
Offline
Joined: 07/28/2008
By uneven, I mean that the

By uneven, I mean that the values will not be symmetric.

ST could range from 1-3 while HT range from 2-4. The width of the range is the same, but HT is slided up to force the player to get at least a +1 advantage in average. I could make the splash damage HT range between 1-3. Or splash damage just reduce cost by 1. I'll see with gameplay.

I think sticking to a fixed amount of 2 properties should be enough and more convenient for the player to calculate and plan ahead. It's also easier to visualize, it's a 2x2 grid. Other property split could be:

Armored VS ?mobile?: Armored unit are more resistant to explosions and area of effect based attacks, while mobile units can dodge more easily projectiles. Melee attacks would be considered explosion.

Range VS melee: Not sure about this property. Can a weapon be more efficient against range or melee units? I think I'll use it for bombardment and initiative instead.

X3M
X3M's picture
Offline
Joined: 10/28/2013
larienna wrote:Range VS

larienna wrote:
Range VS melee: Not sure about this property. Can a weapon be more efficient against range or melee units? I think I'll use it for bombardment and initiative instead.

To me, melee and ranged always made part of a 3-RPS.
And a middleman added to them.

larienna
larienna's picture
Offline
Joined: 07/28/2008
Just o keep the thread

Just to keep the thread updated:

I decided to go with a weaker RPS for now.

  • Having the right weapon can give you +1 (50% chance) or +2 (25% chance) against your target. That is the RPS portion.
  • Having support from another unit gives +1.
  • You must have in average at least +1 to kill the opposing unit.

The idea is to have multiple ways to get your +1. Or you can just have a stronger unit and not bother about the RPS. Or you can have a weaker unit and try to stack the modifiers to your advantage.

I'll make some dummy units with a combat simulation to see how things goes.

Also as the Hierarchy VS RPS, I think I am opting for a flat hierarchy with 2 levels of strength. I am thinking about a structure like this for 8 units:

  • Infantry, offense, weak
  • Intantry, offense, strong
  • Infantry, defense, weak
  • Infantry, defense, strong
  • Vehicle, offense, weak
  • Vehicle, offense, strong
  • Vehicle, defense, weak
  • Vehicle, defense, strong

Infantry and vehicle could be renamed small and large units.

I am also keeping a door open to make each unit improvable to get a few bonus stats as a elite variant unit. that would make 16 different kind of units. This is to allow an option to focus on very little unit variety if stuck.

I thought of value ranges. I was thinking that:

  • Weak unit ST goes within 1-3, with optional 4 if elite units are used
  • Strong unit ST goes within 2-4, with optional 5 if elite units are used

HT would have the same range but 1 point higher to force getting a +1 bonus.

X3M
X3M's picture
Offline
Joined: 10/28/2013
You made a nice and clean overview

The defensive/offensive are attributes here, right?

Maybe find something else for the weak/strong attribute? Unless you were actually talking about ST and HT here.

larienna
larienna's picture
Offline
Joined: 07/28/2008
Weak vs strong: It's

Weak vs strong: It's basically the number of points distributed in stats like ST and HT. I am considering 4 points for weak, 8 points for strong, maybe +2 points for elite units.

As for offensive/defensive is more the general purpose of the unit. Maybe defensive units will have move conservative strength, a good defense and or defensive special abilities. You should use those units to camp places to defend. While offensive units would be stronger in attack, have special offensive abilities and be used to punch through defensive positions.

For the ST & HT stats, I am considering subdividing both stats between:

  • Direct Attack: Value used when a unit is directly targeted by another unit.
  • Collateral/splash damage: Value used when distributing damage at the end of the turn.

Else, there is not that much other numeric stats to spend points into. Many binary stats like melee vs range, ground vs flying, projectile vs splash all seems to have their pro and cons. So having one or the other should not be an advantage.

larienna
larienna's picture
Offline
Joined: 07/28/2008
Just to keep the thread

Just to keep the thread updated. I made a simple combat test with age of mythology pieces, and I realized 2 things.

  • If I use a random initiative order where destroyed units does not get to act, it drastically change the outcome of the battle as you can kill your opponent before it can reply.
  • I remembered that Starcraft the Board game had variable ST and HT which required the need of combat cards to hold both variable value.

So I came up with an idea that changes how dices are used(I don want to use cards). All unit will be able to attack, the single die roll would determine the initiative order. Early units would get -1 ST, late units would get +1 ST. HT would remain always fixed for simplicity. Other source of damage, like collateral damage would use the fixed value.

So the die roll creates a speed VS strength balance that can change the outcome of the battle making it less predictable. It also only require 2 stats ST and HT. If a unit cannot make the kill, it can rout the target, or it can be set aside to support other unit or control the area. If a unit is late to play, it strength boost might allow kills otherwise not possible

One of the benefit is that it will make it more friendly for the AI since the initiative order dictates who acts first. Less chance to have ties. I also thought of a few other side mechanics:

  • Maybe melee units force the target to counter attack the attacker instead of another unit. So you could force the enemy to target the units you want. It would also allow the defender to use is combat capability. While range unit suffer no counter attack, but could lose a chance to attack if killed first. Both seems to have their pro and cons.
  • Flying units could behave like a kind of first strike. Not sure yet.
X3M
X3M's picture
Offline
Joined: 10/28/2013
Initiative via first strike

Units that are fast and/or have a long attack range. Often get to fire first.

True initiative could also get +1 from these 2 effects; ranged and fast.
An unit that has both could get +2 on the initiative.
Add [first strike], and the maximum can be +3.

Perhaps you can create a bit of a natural RPS with this. When rolling, these are the totals for the 1 on 1 matches:

Vs =0; 15 initiative, 6 equal, 15 waiting.
Vs +1; 21 initiative, 5 equal, 10 waiting.
Vs +2; 26 initiative, 4 equal, _6 waiting.
Vs +3; 30 initiative, 3 equal, _3 waiting.

You could put equal rolls in favour of the attacker, or defender. Then you got it all sorted out.

Example, attacker gets the initiative with an equal roll.
Then if thebattacker rolls a 3 and 4, yet the defender rolls a 4 as well.
The attacker has the first initiative for 1 of the 2 units. Then the defender. Then the attacker again.
In a sense, attacking is a +0.5 here, if you will.

Determining the weight is a bit more complex.
But the initiative values can be;
21, 26, 30 and 33.
Or if the defender gets the extra mile;
15, 21, 26 and 30.
If you allow equal results anyway;
18, 23.5, 28 and 31.5.

The weight is determined differently.
Because is is a combination of getting the initiative or not.

larienna
larienna's picture
Offline
Joined: 07/28/2008
Since I would be using D6,

Since I would be using D6, yes there could be ties, and like you said the attacker could get the advantage. I currently managed it as everybody act at once. But giving the attacker the advantage could be interesting.

So far, I did not put an initiative stat to keep the amount of computation low. But there could be a special ability that gives bonus, wins tie, allow attacking first, etc.

Else many war games use initiative letters to order the units. It has a more predictable outcome, but they make the combat result unpredictable, I somewhat do the opposite.

X3M
X3M's picture
Offline
Joined: 10/28/2013
larienna wrote:Else many war

larienna wrote:
Else many war games use initiative letters to order the units. It has a more predictable outcome, but they make the combat result unpredictable, I somewhat do the opposite.

This is a bit confusing.
I think that if the outcome is predictable, so will be the combat result. And actually, the combat result determines the outcome. I suspect, you wrote this diagonal...

But yeah, if you roll for initiative. The combat result is not predictable. And thus the outcome is also not predictable. But only in a sense of, the players do not know what they will roll.

It is still possible to put the dice results in a table. With the other statistics. And then see what the possible combat results are. Eventually, the possible outcomes are also a possibility.

The outcome will be harder to determine if you want to look at a long run. And the dice might result in a combat result of 0.

If you are curious about this all. Perhaps you can give me an example, once you are done designing.
I could test, how much the combat results and outcomes are predictable.

I like the fact, that you look at the 2 as different factors.

larienna
larienna's picture
Offline
Joined: 07/28/2008
I made a computer program

I made a computer program that generates the combat matrix.

So far, the modifiers are:

-1,0, +1: Due to initiative
+1: if same size
+1: if right weapon vs defense

I tried a quick demo yesterday, on thing I realized is that barely checked the weapon modifier has the initiative was most of the time enough to make the kill.

I have various solution on the table.

* Increase the stat range from a width of 3 values to 4 or 5 values. For example, from 2-4 to 2 to 6.

* Restrict even more the RPS. Make sure that the size and weapon type must match before comparing strength. That would imply that a unit can only target roughly 25% of the units available. Then the ST HT comparison would reduce even more the possibility of kill. A variant to that could be that one of both properties must match giving a 75% range of valid target. Else like Star wars rebellion, I could just make size match for targetting. Then maybe make armor type give a modifier.

Adding up modifiers and comparing numbers seems a bit complicated for the mind. Maybe just matching types could be a be easier to compute.

Syndicate content


forum | by Dr. Radut