# Yet Another combat system post

16 replies [Last post]
Jpwoo
Offline
Joined: 03/26/2009

Working on a game you know the drill.

Two main things.

First I'm using opposed 2d10 rolls. Both players roll 2d10 Add bonuses and compare. High roll wins, the margin they win by is compared to a CRT (I think thats the right terminology)

10+ Unit killed: 1%
8-9 Unit Hit:10 %
6-7 Forced retreat 15%
4-5 Hit or Retreat 48%
0-1 No effect 28%

The game covers ancient rome around 100 bc. So I wanted relatively low casualties, and lots of retreats and such.

The percentages given are the general spread for two units of even strength. and apply equally to attacker and defender. (so the attacker has a 5% chance of getting an 8-9 result, as does the defender.) I got the percentages by brute force. I would like to know if there is formula that could help me get percentage chance of results for 2d10+2 Vs 2d10 and such. My probability could be better :)

I like the fat bell curve that 4d10 gives, but is it too cumbersome?

Second: combat set up

Chits have a value on them running from 0-4, with 0 being green concripts up to 4 being Romes professional legionares. All units have two values, a healthy and a wounded. If a unit takes a hit, it flips over to the wounded side. This is generally a 1 point penalty. Another hit kills the unit. So a unit may be a 3 combat value when healthy a 2 when wounded.

When a stack of chits moves into an enemy stack that initates a combat, the person who moved is considered the attacker, though both sides can take damage.

First the defender lays out one of the engaged units in the 'center' If this unit retreat or dies the attackers win, and any other troops remaining retreat.

Then the attacker lays a chit directly against the defenders Center chit. similarly if the attackers center dies or retreats the attack is broken.

Then if the defender has another unit, he puts it to either side of his center, making a Left or Right flank.

Next the attacker lays a flank chit if he has one available.

Then the defender puts down the other flank, followed by the attacker again.

The front is only three chits long. LEFT CENTER AND RIGHT. If there are still more Chits available they may be placed behind any of these positions to replace them if they are killed or retreat.

If there is no opponent on a flank to face, you push that unit up and it is considered 'flanking' the center battle, giving a +1 in your favor.

So a typical battle looks like this. Numidians in Blue, Romans in red.

0,2,1
1,3,2

Combat is resolved for each front, in this case, on the left Rome rolls 2d10+1 agains Numidia's 2d10. Center 2d10+3 against 2d10+2, right 2d10+2 against 2d10+1.

Lets say Rome kills the left flank unit. The second of three rounds looks like this:

1,2,1
X,3,2

The X being an empty space. This next turn Rome gets a +1 flanking bonus for the center battle. The flanking bonus is always +1 regaurdless of unit strength.

All battles must fight the first round of combat. After that either side may retreat, with the attackers declareing first if they are retreating or not.

I guess my questions on this are, is a 3 round limit really needed?
I haven't played many hex and counter wargames, so I don't know lots about them. Am I making any big mistakes? Am I reinventing the wheel when there is an elegant tried and true solution?

Just looking for thoughts comments etc... Apologies on spelling and gramar errors.

TheReluctantGeneral
Offline
Joined: 12/31/1969
Re: Yet Another combat system post

Jpwoo wrote:
Am I making any big mistakes? Am I reinventing the wheel when there is an elegant tried and true solution?

Well, since there's almost as many CRT combat systems as there are wargames, and there's alot of those as you know, obviously you're re-inventing the wheel.

Personally, I think that there are some elegant tried and true solutions out there which you ought to look at first.

However, the real question is what your design goals are. Are you looking for quick play, accurate historical simulation or a carefully balanced game with lots of tactical and strategic depth? Once we know what you're aiming at, I'm sure people can point you in the direction of 'best in class' examples. What is your key goal in creating this game?

Jpwoo
Offline
Joined: 03/26/2009
Yet Another combat system post

Quote:
However, the real question is what your design goals are. Are you looking for quick play, accurate historical simulation or a carefully balanced game with lots of tactical and strategic depth?

I am looking for a middle ground. There aren't a lot of units in the game, so the combat should have some choice point involved in it. But not more than one or two. (in the current system the big decision point is whether to take a hit or retreat.)

So in the combat system, I am looking for light, relatively quick (1-5 min) with 1-2 descision points per combat.

TheReluctantGeneral
Offline
Joined: 12/31/1969
Yet Another combat system post

Well, from what you said above I'd have to say that my personal opinion as someone who has played a few such games in his time, is that the 2d10 system and the menu of combat outcomes is too complex for nearly all the criteria you mentioned.

Speed: I think that there is an unecessary amount of math involved. It may seem silly but even for people used to doing mental arithmetic, adding up two digit numbers repeatedly, followed by a subtraction, repeated many times over is not going to give the very short playing time you're looking for. It's just too much hard work to be fun, I think. Adding up the numbers produced by d6's along with a few very carefully chosen modifies is a lot easier.

Historical Accuracy: The system you described does not seem to offer much in this regard, IMO. Nothing wrong with a quick playing abstraction of course, but this aspect of your system does not seem to chime well with the complexity of the resolution mechanism.

I am also having difficulty interpreting what you mean by 'choice points'.

You may find this article which deals with the aspect of chance in such games useful reading. It deals with crts vs various alternatives.

http://www.thedicetower.com/musings/musings18.htm

Hope some of that helps. I'm not being deliberately negative, just taking my cue from your statement that you had not played many games of this type yourself.

If you like I can suggest a variety of references you may find useful, but it would help to know what the wider context of your game is. (I am presuming that this combat system is part of some wider game).

Jpwoo
Offline
Joined: 03/26/2009
Yet Another combat system post

Quote:
adding up two digit numbers repeatedly, followed by a subtraction, repeated many times over is not going to give the very short playing time you're looking for. It's just too much hard work to be fun, I think.

I certainly don't disgree with you here. It is my main concern with using 2d10.

Quote:
Adding up the numbers produced by d6's along with a few very carefully chosen modifies is a lot easier.

Easier and faster yes. But the curve isn't as long.

Basically there are 4 kinds of units. The values here are Healthy/wounded.

1/0 Green troops
2/1 Calvalry, some legions
3/2 Most legions
4/3 A couple of elite units.

These combat values become more brutal with a 2d6 system. Now granted I could make the elites 3/2, and downgrade the rest of the units and probably get about the same effect. I will work on this tonight and check it out.

Quote:
Historical Accuracy: The system you described does not seem to offer much in this regard, IMO. Nothing wrong with a quick playing abstraction of course, but this aspect of your system does not seem to chime well with the complexity of the resolution mechanism.

What do you consider the key aspects of combat in this time? What would you choose to model?

My reasoning for it.

A large stack can still only engage up to 3 units at a time. So while a large army is useful, it is still something that can be faced.

Both sides can pull a retreat, Allowing for short delaying engagements. Which the Numidian player should do often.

Losing the center loses a battle. Which in this case is an abstraction of loosing a critical place, city etc will break an army.

Forced retreats allow an army to be defeated without taking damage. Battles in these days were frequently low casualty and resulted in retreats.

TheReluctantGeneral
Offline
Joined: 12/31/1969
Yet Another combat system post

Quote:
Easier and faster yes. But the curve isn't as long.

You keep saying that, without stating what the advantage of a long curve is.

Basically there are 4 kinds of units. The values here are Healthy/wounded.

Quote:
These combat values become more brutal with a 2d6 system. Now granted I could make the elites 3/2, and downgrade the rest of the units and probably get about the same effect. I will work on this tonight and check it out.

That is not to say that a d6 based system with a different CRT or resolution mechanism such as a dice bucket will not achieve the level of brutality you require while being easier to play, which is I think my central point. I do not understand how you have arrived at this resolution scheme as being the best fit for your game. You need to explain this.

As for the rest of your point about the accuracy of the system, I am now confused about exactly what this combat resolution is modelling. Is it a single pitched battle (as implied by the centre/flank mechanism), or is it supposed to represent a more abstract series of running engagements and larger scale campaigns (as implied by your comment about loosing a city)? Until you explain the wider context of your game and/or what you are trying to model, it is going to be impossible for me to give any more feedback. Also I think it would be useful to understand your motivation for making this game - there are lots of hex/counter games simulating ancient roman campaigns. Why are you making another one? Perhaps you are trying to remedy perecived deficiencies in the existing library of games?

zircher
Offline
Joined: 12/31/1969
Yet Another combat system post

Just to chip-in, the advantages of a longer bell curve is that the extremes are more extreme while the middle zones are wider. To get a massive seperation between high rolls required more luck. This also fits in with his desire to force more retreats and have less slaughter.

You'd probably have to fire up a spreadsheet or simple program to get the exact probabilities for the various levels of separation that his chart has. That level of analysis would be required to see if the opposed 2D10 method is overkill or if a tuned opposed 2d6 method is possible with similar probabilities.
--
TAZ

TheReluctantGeneral
Offline
Joined: 12/31/1969
Yet Another combat system post

zircher wrote:
Just to chip-in, the advantages of a longer bell curve is that the extremes are more extreme while the middle zones are wider.

That interpretation sounds like the OP is trying to minimise the effect of luck on the game.

You could easily design a system based on a single d6 in which retreats are far more likely than kills, without requiring the the use of multiple d10s. For example, the DBA ancients rules use a single D6 and in that game, kills are quite rare, although kills are more likely for certain specific match ups. In some DBA match ups, two opposed units cannot kill each other at all unless some additional factor (such as a flank overlap) is brought into play.

So I'm wondering it the ratio of kills/retreats and playability or luck mitigation are most important. It helps to be specific if there are specific goals in mind.

Jpwoo
Offline
Joined: 03/26/2009
Yet Another combat system post

Will respond in somewhat random fashion:

Quote:
For example, the DBA ancients rules use a single D6 and in that game, kills are quite rare, although kills are more likely for certain specific match ups.

From Looking on the Geek I'm guessing you are talking about De Bellis Antiquitatis. I will check those rules.

Quote:
Just to chip-in, the advantages of a longer bell curve is that the extremes are more extreme while the middle zones are wider.

Another effect of a long curve is that is softens the effects of an advantage. I did some more number crunching last night.

In the 2d10vs2d10 version, the odds of rolling a 10+ in your favor is about 5.5 % if there are no modifiers. In the 2d6vs 2d6 the odds of rolling 6+ in your favor is about 5.6%. So those two numbers are roughly equivilent.

Now lets say there is a +4 modifier involved. The odds of rolling a 10+ in 2d10+4 Vs 2d10 goes up to about 23%. The odds of rolling a 6+ in the 2d6+4 vs 2d6 goes up to 32%. Which is a signifigant difference.

That said I am still leaning toward 2d6 just for the ease of math issues you mentioned.

Quote:
As for the rest of your point about the accuracy of the system, I am now confused about exactly what this combat resolution is modelling. Is it a single pitched battle (as implied by the centre/flank mechanism), or is it supposed to represent a more abstract series of running engagements and larger scale campaigns?

This is my fault. The terminology was misleading. The hexes are large, and the scale for a turn is in the months. It could have just as easily been called, Center/Supply left/Supply right.

Quote:
Also I think it would be useful to understand your motivation for making this game - there are lots of hex/counter games simulating ancient roman campaigns. Why are you making another one? Perhaps you are trying to remedy perecived deficiencies in the existing library of games?

Sort of off topic, but if you want to know, who am I to argue?

I am making the game because I am reading a history of rome, and this conflict seemed interesting to me. I also like to mess around making games, so this seemed like a fun project. I am not looking to publish beyond a downloadable pdf.

The Jugerthine war is interesting because Rome is pinched. There is a king in Africa who has openly bribed roman officials, assassinated heads of state in roman protectorates and defied Rome at every turn. However at the same time north of Rome barbarians have invaded Gaul and threaten Rome itself.

The Roman player has powerful military resources, but randomly chosen generals of varying usefulness. He has to put down Jugurtha, before Rome becomes too war weary and calls off the war to deal with the barbarians to the north.

The African player has weaker but more manuverable units, and has to fight a delaying war, working to increase the war weariness of Rome through bribes, defeating Roman elites and passing time.

Quote:
I am also having difficulty interpreting what you mean by 'choice points'.

I mean that during the combat players should have to make a choice. In this case the choices are, Take a hit or Retreat, and Push the attack or back off.

TheReluctantGeneral
Offline
Joined: 12/31/1969
Yet Another combat system post

Quote:
From Looking on the Geek I'm guessing you are talking about De Bellis Antiquitatis. I will check those rules.

Yep. Unfortunately they are not online, you have to buy them...

Quote:
Another effect of a long curve is that is softens the effects of an advantage. I did some more number crunching last night.

Well, I think that depends on how tactically important getting a particular kill is. If it is very important, because kills are quite rare (and say you need to get that legion off tha hill, in order reach the enemy camp or whatever), then it might badly skew the results of a given game. The bell curve will only smoothe this out over many games, which is probably not the effect you want.

Have you got enough die rolls and enough potential victims in a given tactical match up to avoid one player being basically stuffed by a single result?

However, it seems to me that if you do have alot of die rolls and plenty of balanced tactical situations, then you don't need to worry quite as much about finessing the probabities to acheieve the right game balance. If you can do this then you may be able to utilise a simpler resolution mechanic which is more fast playing.

Quote:
This is my fault. The terminology was misleading. The hexes are large, and the scale for a turn is in the months. It could have just as easily been called, Center/Supply left/Supply right.

Aha, now I understand :) In which case I guess the centre/flank scheme is OK, but what you seem to be lacking from an historical accuracy perspective is a command and control mechanism. Perhaps you have one worked out already, but you have not mentioned it yet.

For this type of game, IMO it is more fun and and more realistic if you can have choice points which are decided by the troops rather than by the player. Why not have stacks retreat (and advance) according to their own judgement rather than the players? That's not to say you need to remove all choice from the player - just give the troops the ability to disobey. In this case given the scale represented it is more likely the generals of the stacks who will disobey or misinterpret orders though.

Quote:
The Roman player has powerful military resources, but randomly chosen generals of varying usefulness. He has to put down Jugurtha, before Rome becomes too war weary and calls off the war to deal with the barbarians to the north.

The African player has weaker but more manuverable units, and has to fight a delaying war, working to increase the war weariness of Rome through bribes, defeating Roman elites and passing time.

Cool. But I definitely think you need a C&C mechanism to represent this situation correctly. I can provide some examples of such rules if you feel this is something that would add value to your game.

Another game you might like to take a look at if you have not done so already is Command & Colours:Ancients. It has an interesting combat resolution scheme...

Jpwoo
Offline
Joined: 03/26/2009
Yet Another combat system post

Quote:
Have you got enough die rolls and enough potential victims in a given tactical match up to avoid one player being basically stuffed by a single result?

I'm ok with a player getting stuffed on a wild result. But with the current CRT this will happen about 2% of the time on an even fight and 5.6 on a battle at 1 point disadvantage (which seems to be a common match up). Which is acceptable to me.

Quote:
For this type of game, IMO it is more fun and and more realistic if you can have choice points which are decided by the troops rather than by the player. Why not have stacks retreat (and advance) according to their own judgement rather than the players? That's not to say you need to remove all choice from the player - just give the troops the ability to disobey.

In a sense I tried to put this in by making retreat one of the outcomes on the CRT. I would prefer not to add a morale check step. For both time and complexities sake.

Do me a favor and just give the system a try. You can just write on some coins with a sharpie to make chits.

Give the Romans two 3/2 units and a 1/0 infantry. Give the Numidians two 2/1 units and a 1/0 infantry. Alternatively get rid of the Roman 1/0 unit to see the effect of 'flanking, supply cut what have you'

2d6+unit str+1 if opponents supply is cut. vs 2d6+unit str+1 if opponents supply is cut.

Turn order: (not in rulesesse)

1. Defender puts down center unit first. The attacker puts down a unit on Left Center or Right, and players alternate until there are no more units to place. Players may choose to put units behind other units as "backup"

2. going from left to right players resolve conflicts Rolling 2d6 and adding their unit str, and compareing against the CRT. If a center unit has an unengaged enemy on its left or right, supply/communication is considered cut and their opponent gets an additional +1. (Up to +2 if both sides are cut)

A killed unit is removed from game.
A hit unit flips. If it is on its wounded side it is killed.
A retreating unit leaves the battle, but does not flip.

CRT:
7+: Unit is killed
5-6: The unit takes a hit
4: Unit retreats
1-3: the unit may either take a hit or retreat
0: no effect.

3. Attacker may choose to retreat.

4. Defender may choose to retreat.

5. If both neither side retreated, repeat steps 2-4. Combat last a maximum of 3 rounds then the attackers must retreat.

TheReluctantGeneral
Offline
Joined: 12/31/1969
Yet Another combat system post

Quote:
Do me a favor and just give the system a try. You can just write on some coins with a sharpie to make chits.

Ack! Cornered.

OK, special favour then since I gave you a hard time so far. I've done what you asked and here's the result:

Deployment: 2 roman 3/2 units line up square against 2 numidian 2/1 units

Round 1:
Roman left: 7+3 =10 Numid right: 10+2=12
Result : roman takes a hit rather than retreat
Roman centre: 4+3 =7 Numid right: 10+2=12
Result : roman takes a hit

Round 2:
Roman left: 6+2 =8 Numid right: 6+2=8
Result : none
Roman centre: 6+2 =8 Numid right: 11+2=13
Result : roman takes a hit

Round 3:
Roman Right: 3+2=5 Numid right 5+2=7
Result: roman retreats
Roman centre: 11+2 =13 Numid right: 3+2(+1 supply)=6
Result : numid killed

Overall result - attackers repulsed, two romans wounded and one numidian killed. The romans rolled badly but made up for it with one good rool at the end.

Verdict: Not bad, reasonably quick, but these were very small stacks.

Issues:

-when the units are stacked you cannot see which are wounded and which not
- flanking/supply bonus is too small, IMO
- if one ends up with a unit from each side killing or retreating their opponent such that are they allowed to redploy to face each other in the absence of other units?
-unless there are significant other effects brought into play (i.e. terrain, morale, maintainance of supply lines) it's going to feel a bit risk-like.
-I still think the same could be achieved with 1d6 - just make it that units must beat the opponent to get a retreat, or double them to get a kill, and tweak the unit stats a bit. Try it out...

Disclaimer: I'm not in the habit of providing an on the spot CRT evaluation service, just in case anyone else was thinking of sending me theirs.

Jpwoo
Offline
Joined: 03/26/2009
Yet Another combat system post

Quote:
Ack! Cornered.

Haha! I'm sneaky! And thanks, both for your feedback so far and giving it a shot.

Quote:
Round 3:
Roman Right: 3+2=5 Numid right 5+2=7
Result: roman retreats
Roman centre: 11+2 =13 Numid right: 3+2(+1 supply)=6
Result : numid killed

Overall result - attackers repulsed, two romans wounded and one numidian killed. The romans rolled badly but made up for it with one good rool at the end.

A couple of things were slightly off here, but that is probably thanks to my unedited description etc.

The Supply bonus wouldn't occur in the same round that a supply side was broken. The three combats are considered simultainious.

If the Numid Center force was killed, then the Romans won the conflict. Only the center battle matters for determining victory. The Romans would get the hex, and any defenders left over would retreat.

Quote:
Verdict: Not bad, reasonably quick, but these were very small stacks.

Thanks. There is really no incentive to have a stack bigger than 3 or 4.

Quote:
-when the units are stacked you cannot see which are wounded and which not

I'm not sure I understand. How is this different than normal stacking? Is this just a dislike for stacks?

Quote:
- flanking/supply bonus is too small, IMO

Possibly. Maybe I will bump this up after playtesting.

Quote:
- if one ends up with a unit from each side killing or retreating their opponent such that are they allowed to redploy to face each other in the absence of other units?

This can't happen. If a left or right unit has no opponent, moves up to cut supply. If a center unit retreats or is killed then the conflict is over.

It is possible that both sides can cut supply against eachother.

ONR
NRO

Quote:
-I still think the same could be achieved with 1d6 - just make it that units must beat the opponent to get a retreat, or double them to get a kill, and tweak the unit stats a bit. Try it out...

Tried it out in a short test. 10 combats, only 2 made it to the second round. More appropriate in a tactical battle game I think.

Quote:
Disclaimer: I'm not in the habit of providing an on the spot CRT evaluation service, just in case anyone else was thinking of sending me theirs.

That's because you are reluctant!

filwi
Offline
Joined: 12/31/1969
Yet Another combat system post

Hi JP,

from what I read in the thread your system seems to work and you're only concerned with the speed/complexity, right?

If so, would it be possible to decrease time by increasing the uncertainty of play? What I mean is, say that you let both players see what they face (the put the counters openly for the other player's view). Then they set it all up hidden and only reveal their finished army combo. That way you'd decrease the amount of step-by-step decisions and the time it takes to make them but make the outcome more uncertain for the players.

Also, if I understand it correctly you are able to place a unit behind another to replace it if it dies/retreats, right? If so, what would prevent a player with four units in a hex to simply stack them behind each other:

x R x
x R x
x R x
X R x

With a three round limit to the fight, even if the (R)omans got whipped each round they'd have another unit to step in and could never be routed from the hex, or have I missed something?

OutsideLime
Offline
Joined: 12/31/1969
Yet Another combat system post

The formation-creating rules might include a rule that you must fill in an entire rank (left flank, centre, and right) before you can put units into reinforcement slots.

Also, you're forgetting the flanking-bonus. If you lined up your units single-file and I lined mine up abreast, I'd get an auto +2 (+1 flanking bonus against the centre enemy for each unopposed flank unit) on every clash, which would greatly improve my chances of victory.

Sure, with single-file strategy you can hold the centre for more rounds, but if you lose every round, what's the point?

~Josh

EDIT - oops... didn't notice the 3-round limit to the fight... that changes things... well, perhaps that's strategy... line up your troops and take a beating, but at least you hold on to the territory.

Epigone
Offline
Joined: 12/31/1969
Re: Yet Another combat system post

Jpwoo wrote:
I would like to know if there is formula that could help me get percentage chance of results for 2d10+2 Vs 2d10 and such. My probability could be better :)

a) I just learned a lot about Excel, and
b) here's a spreadsheet that will calculate probabilities for rolls like you describe: NdK+B vs. NdK. It's zipped and in CSV format for smallness, but Excel should open .csv without complaining. There are 4 things you can change around, N=, K=, B=, and the ranges you want. The way the range works is if you're trying 2d10+2 vs. 2d10 (for instance), and you want to get the probability for 0-1, 2-3, 4-5, 6-7, 8-9, and 10+, you would enter

-100 -9 -7 -5 -3 -1 2 4 6 8 10 100

Then the spreadsheet will calculate the chance of the difference being in the range [-100,10], [-9,-8], [-7,-6], [-5,-4], [-3,-2], [-1,1], [2,3], [4,5], [6,7], [8,9], [10,100].

Just don't try something like 10d20, since I didn't make the tables big enough for that!

Jpwoo
Offline
Joined: 03/26/2009
Yet Another combat system post

Epigone,

You should consider putting it up in the files section here.

Quote:
If so, would it be possible to decrease time by increasing the uncertainty of play? What I mean is, say that you let both players see what they face (the put the counters openly for the other player's view). Then they set it all up hidden and only reveal their finished army combo. That way you'd decrease the amount of step-by-step decisions and the time it takes to make them but make the outcome more uncertain for the players.

Once I start to test with live people I will probably give this a shot. I am not sure that it will save time, as people will have to analyze more information. But that said I think it could add more fun.

Quote:
Also, if I understand it correctly you are able to place a unit behind another to replace it if it dies/retreats, right? If so, what would prevent a player with four units in a hex to simply stack them behind each other:

As Josh said, you would be giving up +2 to your opponent, which is a pretty big deal. Yes, You could set up so that you definately don't loose a spot.

I have considered eliminating the three round limit, as most fights don't seem to be going 3 rounds anyway, the artificial cap isn't really necessary.

All in all I don't think I mind it the big "protect the center Strat"