Balancing combat encounters

[zedturtle]

One of the challenges that I always have is figuring out how many opponents to throw at my characters. If I give too easy a challenge, it becomes a mechanical exercise that serves as nothing more than a speed-bump. If the opponents overwhelm the heroes then either I have to pull a deus ex machine out of my hat or I end up with overwhelmed or frustrated players.

So, balancing combat encounters would be something I would be interested in. I bet others would find it useful as well. So three things to figure out:

1. What is balanced? I'm going to go ahead and define this as the ideal encounter: a combat where the characters sustain more damage than they can recover by a short rest, but not much more than that. I want each combat to feel like it matters (some of the characters should be permanently hurt, at least a little) but not overwhelming (e.g. no more than 25% of the heroes wounded, no more than 25% knocked out). The other point with this is if we can get this working then LMs can make an easy encounter by making an encounter with a 75% budget or a super tough encounter with a 150% budget.

2. How do we calculate the heroes value? Originally I was thinking something along the lines of "(Highest Valour + Highest Wisdom + Highest Weapon Skill) x Number of Heroes" but maybe total Experience Points (which you'd have to use to get all of the above) would account for more things... magic items, for example.

3. How do we calculate the adversaries value? Something similar like "(highest Hate + highest Attribute Level + highest Weapon skill) x Number of Adversaries"... That doesn't account for Endurance though.

- - -

How to figure this out? I'm thinking lots of Monte Carlo simulations is the only way. Elf's combat simulator does good for single opponents but I'm thinking I need to do a company (of variable numbers... there are Fellowships of 2 or 3 out there and Fellowships of 7,8 or 9 as well) and look a newbie heroes, experienced heroes and maxed out heroes.

Input or ideas would be most appreciated. Also, tweaked-out character builds (baseline, 50 XP and 150 XP) like James made in the "biting dart" thread would be awesome. I'm not too good at doing super-efficient builds like that.

- - - - -

Or, y'all can tell me that I'm missing something that already exists, or that I should just continue to wing it.

[zedturtle]

Y'all can also tell me that I've got too much free time, as well.

Adding up the adversaries in the core rulebook according to formula (AL + Hate+Parry+Armour + 1 for each favored, +1 for special abilities, -1 for negative SA, + End/10, round up) gets this:

Great Orc 36
Messenger of Lugbúrz 24
Snaga Tracker 13
Black Uruk 25
Orc-Chieftain 28
Orc Soldier 12
Goblin Archer 9
Orc Guard 20

Attercops 18
Great Spider 24

Cave-Troll 37
Hill-Troll 35
Hill-Troll Chief 45
Mountain-Troll 44
Stone-Troll 32

Wild Wolf 16
Wolf Leader 22
Werewolf of Mirkwood 52
Hound of Sauron 30

Great Bat 18
Secret Shadow 33

Of course, Endurance is slighted here, but it tends to be the easiest damage type to inflict, so I think that's ok.

Now, we either need to mod these numbers or figure out a way to calculate the heroic value appropriately.

[zedturtle]

Alright, so how about this?

Heroes are ranked by the highest of their Weapon Skill, Valour or Wisdom. For each of the two non-highest values, if they are four or more this gives a +1 to the ranking. This means every hero has a number anywhere from 2 (starting hero) to 8 (maxed out or nearly so). Simply add up the hero ranks to find a Fellowship rank total.

The above values for enemies look decent... An orc chieftain is as dangerous as four goblin archers. Also, an orc chieftain seems to be a good baseline for starting heroes. If we make him equal to 6, we're saying that a starting Fellowship of four should be able to take care of him with no trouble and a Fellowship of two would think he's a major threat.

Rejiggering our numbers from above (divide by 6, round standard), we get:

Great Orc 6
Messenger of Lugbúrz 4
Snaga Tracker 2
Black Uruk 4
Orc-Chieftain 5
Orc Soldier 2
Goblin Archer 2
Orc Guard 3

Attercops 3
Great Spider 4

Cave-Troll 6
Hill-Troll 6
Hill-Troll Chief 9
Mountain-Troll 9
Stone-Troll 6

Wild Wolf 3
Wolf Leader 4
Werewolf of Mirkwood 9
Hound of Sauron 5

Great Bat 3
Secret Shadow 6

[Glorelendil]

I'm not sure of the usefulness of my simulator for balancing combat, even if I modify it to allow for group vs. group. The sim is good for examining the relative difference between variables (e.g., between types of armor, or Fell vs. Keen, etc.) in specific circumstances, but it doesn't do a good job of modeling situation tactics, such as when heroes spend Hope or use Combat Tasks, or when adversaries spend Hate.

But even if the simulator could model these things, what would we mean by 'balanced'? "The heroes win x% of the time?" TOR combat relies on relatively few dice rolls, and while I like the variability smoothing effect of the dice system, it's still very easy to have streaks, good or bad. So let's say we determine a "balanced" encounter is one that the heroes win 90% of the time. For the 10% that lose, it's going to feel like it was 'too hard'.

If you try to define 'balanced' as having to do with averages ("On average, the heroes win but lose 75% of their endurance") there is still going to be some % that results in TPK.

I guess my definition would be something like "They win less than 50% of the time without spending Hope, and 90% of the time if they do spend Hope." If that sort of 'balance' could be achieved, the fights would tend to feel hard, and I'd only have to resort to deus ex machina 10% of the time.

[zedturtle]

Well, true balance is probably not possible (or maybe desirable), but I just wanted some way of being able to say "if you put this adversary up against this group, a TPK is probable."

I did try to make there be a sensible numbers (more Endurance damage than Heart score, no more than 25% of company wounded) but I agree, without a comprehensive way of simulating full combat, it's a guessing game at best.

[Glorelendil]

Thinking about this more last night, if I did modify my sim to handle many v. many combats, I would want to link it to a character sheet database so that you don't have to spend 30 minutes setting it up.

Further incentive to get started on just such a database...

[Hermes Serpent]

Can't you pull from azrapse's character tool? The output is pretty straightforward and the 'free' un-password protected character data has a fair number of characters listed.

[Glorelendil]

Can't you pull from azrapse's character tool? The output is pretty straightforward and the 'free' un-password protected character data has a fair number of characters listed.

That gets me part of the way there, but there are some features I'd want that it doesn't have.

Besides, the coder's motto is "why modify somebody else's code when you can start over?"

[zedturtle]

Can't you pull from azrapse's character tool? The output is pretty straightforward and the 'free' un-password protected character data has a fair number of characters listed.

That gets me part of the way there, but there are some features I'd want that it doesn't have.

Besides, the coder's motto is "why modify somebody else's code when you can start over?"

As a self-taught hobbyist programmer, this is what has always gotten me in trouble when I've programmed for pay. I'm so used to reinventing the wheel in order to understand it that I have a really hard time incorporating or understanding other people's code.

[Indur Dawndeath]

Hi Zedturtle,

I think that it is a great project. I also have difficulties scaling the opposition. My players was nearly defeated by 6 Orc Soldiers at one occation, and easily killed a Hill-Troll Chief and 3 other Trolls at another, not at the same time, but without rest between the fights.

So what I think is missing from the formula is an "outnumber value". If players outnumber the opposition, then they are going to win most of the time, but if they are outnumbered, then they will have a difficult time surviving.

Another, very important issue, is that it is such a powerful combination to have favoured armour and a high attribute, in combination with Great Size and possibly favoured weapon. Such a foe will not be defeated until it is wounded, so it all depends on the players ability to wound. A Great Orc is so much more dangerous than any of the other creatures... Snow Trolls are the only creature, that is equally insane...

And combinations of enemies can also be very deadly. One hero facing 3 orc soldiers may be victorious, but alone against 1 orc soldier, 1 wild wolf and 1 great bat... He will be defeated.

This may be too difficult to fit into a nice spread sheet, although it would be nice to have one available when you plan the next combat scene.