My EDU balancing system

[fallen851]

I hope the title of this thread adquately describes the system I came up with to balance units. It changes how the game is played quite a bit, especially making important differences between civilized (heavily armored, low charge, resistant to missiles, low attack) and barbarian units (light armored, powerful charge, vunerable to missiles, high attack).

A few quick comments about the below: #1 Fatigue is a factor I developed to help reduce the charge of heavy armored units, and reduce their ability to attack, as well as balancing out defensive skill (I hope the reasons for this are self-explanatory). #2 Level is a factor I developed to provide a base for each unit, so you need to select a "unit level" (which is supposed to be a representation of unit ability, willingness to fight, and training/experiance) before compling the statistics (ie peasants would be a level 1 unit, and Spartans a level 10 unit, although this depends on your opinion of their ability).

Equations:

Select Level 1-10
Morale = Level + 2
Charge Bonus = (Level + Weapon (* 2 for cavalry spear, * 1.5 for other cavalry weapons) - Fatigue * 2) * 3
Defensive skill = Level - Fatigue
Melee Attack = Level + Weapon - 1/2 of Fatigue
Missile Attack = Level + Weapon
Heat = fatigue * 4 (minimum of 2)

Reduce Moral and Defensive skill by 1 for Mercs

Always Round Up

No Charge Bonus for Phalanx Spears.
Pike fatigue is not added when caculating charge bonus or attack of secondary weapons of Pikemen.

Weapons:

Pila: +5, AP
Bow: +6,
Sling: +6
Javelins: +9

Pike: +2, +1 fatigue
Spear: +3
Long Spear: +3

Mace: +4
Axe: +4
Two-Handed Axe: +6, AP, +1 fatigue

Dagger: +1
Short Sword: +2
Gladius: +3
Long Sword: +4
Falx: +5, AP

All lethality is set at .55
Cavalry Spear Lethality set at .47

Armor:

Light helmet: +1
Medium helmet: +2, +1 to fatigue
Heavy Helmet: +3, +1 to fatigue

Heavy Cloth: +1
Leather: +2
Breast Plate : +2
Mail: +5, +1 to fatigue
Heavy Mail: +6, +1 to fatigue
Plate Armor: +7, +2 to fatigue
Heavy Plate Armor: +8, +2 to fatigue

Armor is reduced -2 for unarmoured mount with armoured rider (shield and leather/cloth does not qualify a rider as an "armoured rider").
Armored cavalry recieve +1 fatigue.
Units with light cloth or no shirt recieve minus 5 heat.

Shields and Directional Armor:

Greaves: +1, +1 to fatigue
Directional Breast Plate: +2
Pikeman Small Shield: +3
Peltast Shield: +4
Barbarian Shield: +6, +1 to fatigue
Hoplon: +7, +2 to fatigue
Large Barbarian Shield: +8, + 2 to fatigue
Tower Shield: +9, +2 to fatigue

Based on my EDU research, when adding shields or directional armor, add half the points to the shield value, and then half the points to the normal armor value. If the shield gives an odd number of armor points, add the greater value to to the shield value. The +1 greaves gives is simply added to shield value.

For instance if you have a tower shield, you would add it like this:

stat_pri_armour 4, 0, 5, metal

Why is this? Shield values do not provide the same protection against missile weapons as normal armor (they provide about 3/4 as much protection I think). This helps balance that out, though since AP values aren't calculated when going against shields, it makes shields less effective vs AP, and also increases rear and side defense. I'm not sure yet if this is a sacrifice I'm willing to make, further testing will follow tommorow.

Here is the research:
https://forums.totalwar.org/vb/showthread.php?t=62744

[Monkwarrior]

I'm always interested in this type of equations, just to put some order and rationale in the units.

For the moment I have some questions:

- Is Level an arbitrary value you put from your idea of the unit? My guess is that it is arbitrary and it is the "base value".

- What are the limits of Fatigue? Also 1-12?

- When do you apply the weapon/armour/shields values? Before or after the calculation?

- Could you give one or two examples of calculations and the final entries in EDU? For example one typical roman unit (let's say princeps) and one barbarian unit, such as chosen swordsmen.

I hope this idea will make easier the balance of the units.:2thumbsup:

[fallen851]

No limit to fatigue. That said, even a unit with heavy plate, a big helmet, greaves, and big shield would only develop 6 fatigue (the weapons that develop fatigue are two-handed axe, and pikes that are only used with a medium shield at best). Below I have an example of a unit with high fatigue (Armored Hoplites).

The level is arbitrary like you said, and furthermore it is relative to other units. It essentially effects everything. I think it is safe to say that some units are clearly level 1 (peasant) while some could be level 10 (Spartans, Beserkers...). It gets rather hard when you have a unit like Hastati though, but I think most would agree they are somewhere between 4-6.

Here are the two examples you requested:

Chosen Swordsmen - Heavy Infantry - 80 men

Level 8

Melee attack = 10 from 8 (level) + 4 (longsword weapon) - 2 (fatigue halved and rounded) = 10
Charge Bonus = 6 from 8 (level) * 1.5 - 3 (fatigue) * 2 = 6
Fatigue = 3 (totaled from the pieces of armor)
Armor Factor = 6, +1 to fatigue
Helmet Factor = 2, +1 to fatigue
Shield Factor = 6, +1 to fatigue
Defensive Skill = 5 from 8 (level) - 3 (fatigue) = 5
Morale = 12 (level * 1.5)
Heat = 12 from 3 (fatigue) * 4 = 12

Please note follow My (Multiplication first) Dear (Division second) Aunt (Addition Third) Sally (Subtraction last).

Princepes - Heavy Infantry - 80 men

Level 7

Pila Attack = 7+5 = 12, AP
Melee Attack = 7+3-2 = 8
Charge Bonus = 3
Fatigue = 4
Armor Factor = 6, 1 to fatigue
Helmet Factor = 2, 1 to fatigue
Shield Factor = 9, 2 to fatigue
Defensive Skill = 3
Morale = 11
Heat = 16

So helmet and armor factor are added, and compose the first number of "stat_pri_armour". Also shields and directional armor are totaled, and compose the third number of that same line. The rest of the numbers are simple plug-ins.

And in the EDU:

type barb warguard briton
dictionary barb_warguard_briton ; Chosen Swordsmen
category infantry
class heavy
voice_type Heavy_1
soldier barb_warguard, 40, 0, 1.5
officer barb_standard
attributes sea_faring, hide_improved_forest, hardy, warcry
formation 1.2, 1.2, 2.4, 2.4, 3, square
stat_health 1, 0
stat_pri 10, 6, no, 0, 0, melee, simple, piercing, sword, 25 ,0.75
stat_pri_attr no
stat_sec 0, 0, no, 0, 0, no, no, no, none, 25 ,1
stat_sec_attr no
stat_pri_armour 8, 5, 6, metal
stat_sec_armour 0, 0, flesh
stat_heat 12
stat_ground 0, 0, 0, 0
stat_mental 12, impetuous, untrained
stat_charge_dist 40
stat_fire_delay 0
stat_food 60, 300
stat_cost 1, 680, 210, 70, 110, 680
ownership britons


type roman princeps
dictionary roman_principes ; Principes
category infantry
class heavy
voice_type Heavy_1
soldier roman_principes, 40, 0, 1.3
officer roman_early_standard
attributes sea_faring, hide_forest, can_sap
formation 1, 2, 2, 3, 4, square
stat_health 1, 0
stat_pri 12, 3, pilum, 65, 2, thrown, blade, piercing, spear, 25 ,1
stat_pri_attr prec, thrown ap
stat_sec 8, 3, no, 0, 0, melee, simple, piercing, sword, 25 ,0.75
stat_sec_attr no
stat_pri_armour 8, 3, 9, metal
stat_sec_armour 0, 0, flesh
stat_heat 16
stat_ground 0, 0, 0, 0
stat_mental 11, disciplined, highly_trained
stat_charge_dist 30
stat_fire_delay 0
stat_food 60, 300
stat_cost 1, 490, 170, 50, 80, 490
ownership romans_julii,romans_brutii,romans_scipii,romans_senate

You'll note I blanked everyone under "stats_ground" just because I haven't decided what the bonuses should be, if any. Also pilum range is increased in this example (I prefer the longer range).

Fatigue's main purpose to reduce the charge of units wearing lots of armor, and to also to reduce their defensive skill, as they will be less mobile. It also helps balance out melee combat with unarmored units.

Lastly, here is an example of unit with lots of fatigue, and one with barely none for comparison.

Armored Hoplite - Heavy Infantry - 80 men

Level 8

Spear attack = 8+3-3 = 8
Melee attack = 8+2-3 = 7
Charge Bonus = 2
Fatigue = 5
Armor Factor = 8, 2 to fatigue
Greaves* = 1, 1 to fatigue
Helmet Factor = 3, 1 to fatigue
Shield Factor = 7, 1 to fatigue
Defensive Skill = 3
Morale = 12
Heat = 20

*Greaves would be added in the EDU text under shield.

Barbarian Swordsmen - Heavy Infantry - 80 men

Level 6

Melee attack = 6+4-1 = 9
Charge Bonus = 7
Fatigue = 1
Armor Factor = 0
Helmet Factor = 0
Shield Factor = 6, 1 to fatigue
Defensive Skill = 5
Morale = 9
Heat = 0 (minus 5 heat because they are shirtless)

As you can see, fatigue really does a number on attack (despite being two levels lower, the swordsmen has a better attack), heat, defensive skill, and especially charge. But the armored units gain in overall protection, and especially missile protection. Whether or not that is worth it to you depends on your style of play. Lightly armored units would prefer a short battle, in which their high charge breaks the enemy line quickly, and a short time of missile exchange. They don't want to have to melee for a long time with no armor. Heavily armored units prefer quite the opposite. I think I would like to increase the charge bonus further for lightly/unarmored units over heavily armored troops.

[Ciaran]

Very interesting, but have you actually tested the resulting units stats ingame?

The "Level" factor seems a bit arbitrary, now one could deduct the Level from, say, the point of the game where the unit is available, but on the other hand (and that´s what I would think this stat creation system best for) if you´re developing a mod, you tend to make the availability of a unit relative to it´s strength...
If I understand the sysem right, the basis value for fatigure is zero (or one?)?

And why does morale cap out at twelve? As far as I know, Urban and Praetorian Cohorts have a morale of 14. Technically, it could go up to 99 (I´ve tested, it works).

[fallen851]

The results in game work to my liking. My mod is balanced for 1.5, and to remove the archer bug, archers are quite power (base archer has 8 for attack!), but I only give archers 12-18 ammo, and archer units are half sized, so it works out ok.

I increased the size of many barbarian units to 100 also.

The base value for fatigue is 0.

As for morale capping out at 12, well one of things I wanted to with mod was to avoid what Alexander the Pretty Good says, "If you have to use strategy to win, you don't have enough elite units".

In vanilla RTW I feel each unit is exponentially greater than the last. I wanted to make it quite different. Now Regular Hoplites are still better than Militia Hoplites, Armored Hoplites better than Regular Hoplites, and Spartan Hoplites are best, but the quality improves in slowly, in a linear curve, while the price of the units goes up exponentially. While Spartan Hoplites are clearly the best, the difference between them and Armored has been much reduced from vanilla, and the difference between Armored and Regular reduced as well. In fact one unit of Armored and one unit of Militia Hoplites will be able to defeat relatively easy a single unit of Spartans together, and those two units cost roughly the same as one unit of Spartans. In other words it would now be far better to have lots of decent basic units, instead of a few elite units. Of course the kicker is you can only have 20 units in an army, so you want to use those spots as best you can.

To me, with a shorter spread between normal and elite units, the game is a lot more fun. It really gives the underdog a chance against elite units, because now it focuses less on unit quality (which is still important though), and more on strategy. An army based on normal units (Regular Hoplites) with a good commander can now defeat an army based on elite units under poor command (Spartan Hoplites), which I felt was a rarity in vanilla. Now its clear that elite units are still superior, but I didn't want unbreakable Spartans fully surrounded and not routing, that is unrealistic.

Now I've thought maybe the moral cap should be higher, as a level 7 unit gets moral 11, level 8-10 get 12, but I don't want to make elite units overwhelming. I'm open to ideas.

[HalfThere]

Your system is interesting, and certainly more scientific than mine (which is basically think up values as I edit the EDU, then balance in-game through trial/error). However, it's not really going to be very useful to anyone else. After all, we each have our own vision of gameplay and it's not entirely likely that it will overlap with yours enough to go through the trouble of using your system.

Still, if it works for you more power to you.

[professorspatula]

I use similar system myself for determining stats, though simplified: armour is counted as 1 overall element for example, though it still affects attack and defence values, and fatigue and heat is less of a consideration. Equipment is a big factor in determining skills, although I consider the training and professionalism of the unit to be the main factor. I've used several systems to work out stats, and my current one's main purpose is to level the playing field so that the difference between levy and elite isn't the huge gap it is currently. I also re-enforced the Scissor/Paper/Stones element of the original TW games, making spearmen more defensive and stronger against cavalry, but easily defeated by swordsmen, whilst cavalry destroy the sword infantry etc.

I wonder if like me, you end up re-adjusting the stat system often and never quite satisfied? I'm on my 8th version and only just reaching the point where I think it is fine tuned enough to offer a really good balanced game.

[Seasoned Alcoholic]

Some interesting work there fallen851, especially the fatigue system. With regards to this, how frequently do you use the 'hardy' and 'very_hardy' stamina attributes to counterbalance a unit's fatigue levels? Tbh, I've not examined unit fatigue under the microscope, but have attempted a similar 'level' approach as yourself, with each progressive level meaning a general increase in unit ability, and therefore costing more to recruit and upkeep per turn.

With each level comes a collective morale stat, similar to that of other cultures so as to balance the gameplay and not skew the results of a battle in a particular direction. Although sometimes it does depend considerably on the type of battlefield a unit is fighting upon, EG heavily armoured units slugging it out in the simmering heats of the desert, hence the stat_heat attribute.

A long time ago (in a galaxy far, far away ~D), DARTH VADER revealed that the stat_heat attribute was bugged in his opinion. Not sure if he reversed the signs in front of the values or increased / decreased them accordingly. Think this post was located at .COM forums, and knowing their habits, this post will now be long gone (deleted in other words). I haven't looked at the stat_heat attribute and examined how it works, but it would be good to know whether DARTH or CA are the most accurate ~D

[fallen851]

I wonder if like me, you end up re-adjusting the stat system often and never quite satisfied?

Well I just adjusted the system in terms of morale, charge bonus, and lethality (edited above) and I'm really pretty happy how its plays out in the game.

If you guys have any negative feedback about why my system sucks, please feel free to criticize... The system itself is rather arbitrary. Maybe I should ask this, is there any values you would change if this were your system?

[Monkwarrior]

I've just began to apply this system and (as you can imagine) more questions come to me:

- The attack value for archers seems to be rather low. As the bow value is only 2, and the archers are not highly skilled units (I put level 4 for low archers, and 6-7 for high ones), the values are in the range of 6-9 (counting no fatigue penalty). Do you consider to distinguish between short range bows and long range bows?

- Have you any method to calculate the unit cost?

More questions in the near future.:2thumbsup:

[Watchman]

One thing that always kind of bugged me already with the vanilla EDU is the oddly low lethality values given to things like falxes and big two-handed axes. I simply find it very difficult to comprehend how such large, very powerful weapons that (at least in the case of big axes) pulverize shields with one hit and all but cleave even armoured men in half with just one blow could be in game terms considered less "lethal" than, say, short swords...

Not that the non-phalanx spears' low lethality rate made much sense either. By what I know of it the human physique doesn't much differentiate between deep stabs from swords and spears (or arrows, javelins...), it goes out of order quite equally from both (and rather easier than from slashes too). Organ failure and system shock aren't terribly interested in the exact weapon that inflicted the wound, what matters to them is that assorted important squishy bits just got inappropriately perforated.

Just a bit of a peeve of mine. I don't know what (if any) the game-mechanical justifications are, but it just doesn't make sense to me.

[fallen851]

I've just began to apply this system and (as you can imagine) more questions come to me:


I actually didn't expect anyone to start using it so soon. The reason archer attack was so low was because thats the way it is in EB, I updated the above states/equations to what I'm using at the moment, for 1.5, and by using this system and setting base archers at level 2, and ones like Chosen at level 4, with Forester and Cretan being level 5 and the best. This high attack nearly annihilates the archer bug. I also set archer/slinger units to 25 man size (I play on large so it come outs to 50 men) and reduce ammo to 10-18 depending on if it is a base archer or an elite archer. Archers become quite effective, but not overly effective, and quite worthless in melee too. Furthermore, most barbarian units are set at 50 man size.

I spent the day doing some significant tests which I'm going to post soon, and despite what others have said, it does look like my system makes sense.

I don't really have any method to calculate cost yet.