TL;DR:
The current Combat Strength formula is flawed, breaks the balance of warfare at its core, invalidates the use of fortifications and terrain, makes nearly all melee units entirely redundant from a minmaxing perspective, makes never updating your units nor researching unit technologies and spamming ancient era archers in contemporary era the one winning strategy, makes it so that ANY tinkering with unit stats to try and change the balance of warfare is pointless because anything outside of a very narrow difference in combat strength has no effect, and other minor concerns in that vein. Therefore, said formula should probably be changed entirely, the damage based on the ratio of combat strength between the units involved, rather than the flat difference in it.

For details, read the wall of text and/or take a quick glance at the tables below (do note that since the recent changes it is slightly outdated for the -30 to -70 range of damage dealt only, but the gist of it remains):

I may or may not have made an account just to mention this, and I'm sure the topic has already been raised at some point in the past. However, the latest patch has mentioned (quote) "adding a notion of attacker anachronism to limit the damage formula.", and I feel like the devs might be on the wrong track on how to fix the issue, or indeed might be misunderstanding why it is one in the first place.

Anachronistic units are not the problem. They are not what is broken, especially not if the civ using them is actually in that era. What is broken is the combat strength formula itself, and until that is changed, any tinkering with combat strength of units or modifiers from terrain, fortifications, empire bonuses and so on is... not going to do a single thing.

It might sound crude to claim so, but it is true. Why?

Okay, so first of all, the impact of each individual point of combat strength is often widly and unpredictably different. For example, currently:

Going from +0 to +1 relative CS will provide you with +12.5% damage dealt and -10% damage taken.
Going from +9 to +10 relative CS will provide you with +6.3% damage dealt and -0% (yes, no difference) damage taken.
Going from +13 to +14 relative CS will provide you with +20% damage dealt and -0% (yup) damage taken.
Going from +16 to +29 relative CS will provide you with... +0% damage dealt and again -0% damage taken. Yes, nothing for an entire 14 points. It is only at ~ +30 that there is a small reduction.

As illustrated by these examples, not only is combat strength currently unintuitive in its actual impact and does not seem to follow any formula known to man, but it also caps out the damage dealt at +16 and establishes the floor of damage taken at a difference as low as 4 points in Combat Strength, which reduces the damage by ~25%, relative to the value at +0. This means even units within the same era commonly hit the damage floor, making any further differences in CS extremely unimpactful - especially since anything from +8 to +15 means requiring two whole hits to kill a unit, therefore every value within that range of seven entire CS points is effectively the exact same for both damage dealt - and damage taken.

This is particularly an issue with ranged units, as they do not usually take retaliation damage, and so they can defeat far stronger melee units through sheer numbers while any bonuses from good terrain, fortifications, veterancy and empire boni might as well not exist for the latter - since they will be taking - and dealing - the exact same damage values regardless - which particularly affects units with a one or two era difference, though it can affect even same-era matchups such as Ancient Era archers vs emblematic units or even generic chariots.

Therefore, very outdated ranged units become extraordinarily powerful, and spamming them is far quicker and cheaper in industry and often even population that producing up to date ones when taking the cost to performance ratios into account - allowing the player fielding them to not only win against more advanced units with equal population expenditure, but to also save a huge amount of industry to invest into districts to snowball, on top of negating any and all industry that the other side might have sacrificed in order to make their location more defensible - after all, all bonuses to Combat Strength of their units become entirely meaningless.

As is, the anachronistic thresholds that randomly start appearing at differences of ~30 CS and above do very little to mitigate this even against ancient era archer spam, until the last two eras. And even if it were tweaked, one could always spam slightly less outdated units (such as english longbows) for much the same or even greater effect. And since CS bonuses are flat and not percentage, and the impact of each point of combat strength is uneven, so long as you can make your units not one-shottable, they will be twice as effective - which will arise from a difference of as little as 1 CS point such as -15 vs -16!

Currently, 20 combat strength, ancient era archers will deal and take the exact same damage to and from ~36 combat strength classical or medieval units, not to mention ~45 combat strength early modern units. The exact same damage dealt and taken. That is to say, the archers will deal 5-25, and take 100 damage.

While the anachronistic formula change does improve the situation on the very high end (such as a difference of 60+ CS), it does little to affect actually common outcomes - +39 CS is a mere 16.7% damage reduction relative to +4 CS, for a whooping grand total of 0.48% per point, and applied in wide thresholds instead of granularly at that. In comparison, at several points on the damage curve, a single point of CS can mean a 20% damage difference - which is more than the gargantuan 35 such points do here.

In other words, functionally speaking, there is zero difference in the efficiency of units of long obsolete eras while fighting something of just an era ahead (or sometimes even the same era), and something two or three eras ahead.

Sadly, all that the recent band-aid change does is it makes you switch to spamming slightly less outdated units past the early modern era (the most efficient strategy is spamming ancient era archers until early modern and then medieval era units like longbows up till and including the contemporary age), since it does not address the core issue - which is that the formula itself is flawed.

Within a realistic range difference, as long as a unit is outside the magical bound of 4 or less combat strength less than the opponent, it will deal 5-25 damage (and this is often the case between units of the exact same era, even before any terrain, empire or fortification bonuses are applied, making them quite pointless in gameplay, since everything will deal the same damage anyway whether you use terrain and fortifications to your advantage or not!)

As long as a unit is outside the magical bound of 16 less combat strength than the opponent or more (again not that wide a range, that's often just one era ahead), it will get one shot. Since units only get one attack per turn, this inevitably results in a situation where making - or indeed even researching higher era units is in fact detrimential - in the current version of HK, you do NOT want to be making anything but archers, maybe longbowmen later, and the more advanced the unit you are making, the worse it is.

Differences in combat strength past a certain point (a wall of -4 to +16 CS, both sides of which are often surpassed by units in the exact same era, especially aided by unit abilities or bonuses) often do exactly nothing. A +29 is no different from +16 in neither damage dealt nor taken, a +8 from +15 as well, though the latter only if one accounts for the fact that minor differences in damage dealt that don't change the "hits to kill" counter don't really change anything.

This is even more of an issue when one considers that melee units often don't particularly care about damage dealt, since they require adjacency and are limited by space and terrain and find it difficult to concentrate their damage output and indeed sometimes even dish it out, therefore damage reduction from higher CS is their only reliable defense against massed and concentrated missile fire - and currently there is realistically nothing as such, as they will always take 5-25 damage or more from remotely era-adjacent ranged units, making them not have a place in the game because of how the formula works.

Effectively speaking, there is a greater chance that any given change in combat strength will do nothing, than that it will do something. And even when it does do something its impact is uneven per point, making it very unintuitive on top of creating balance issues - after all, there is hardly a way to balance unit A against unit B safely when any change to the former's stats will have an entirely disproportionate impact on its performance against units C and D.

And that's far from the only problem even. Because of the current formula, Archers will beat the same industry cost in Chariots on neutral terrain plains without any defensible locations (that they can always use), despite Chariots literally hitting them at 1 CS shy of the damage cap (+15 CS), and getting hit at the damage floor (-5, and floor is at -4 CS). They will beat Warriors 1 to 1, as well, even harder so if massed and allowed to make use of opportune terrain.

As an extreme example of obsolete units beating state of the art ones, I will use the contemporary era main battle tank - it requires a ton of research, takes 16805 industry to make and reserves 90 gold of upkeep, on top of costing 3 population (which is in itself an odd requirement - it is a tank, not an infantry brigade.. but I digress).

Now, due to the mechanics between combat strength, 3 medieval longbowmen will perform just as well if not better on the field of battle, while having the same range and indirect fire - and they will cost a sum total of 600 (200) industry, 45 (15) gold of upkeep, and also 3 (1) population points. Therefore, they will cost the same amount of population to make, but a mere 3.6% of the industry and half of the upkeep, and this is after the anachronicity formula changes. You will be able to produce these longbowmen by the dozens in every city in a single turn - in the same cities that will struggle to produce a tank per turn. And you will win every single war by spamming medieval era units against contemporary ones, just as you will every battle against early modern units by spamming ancient era archers.

Unless you research their improvements. Now, besides these being inferiors due to line of sight mechanics, they will do the exact same damage (5-25), and take the exact same damage (100) from everything they will fight, while costing several times more industry and upkeep than their predecessors. And you will no longer be able to make those because of this "upgrade". It is the same for every ranged unit, there is no need for a gap of more than one era between them - longbows and tanks are just examples. You might as well use an industrial construction team and wonder why they are oneshotting medieval knights in full plate with their... shovels?

Well, because if civilians didn't have so high a combat strength under the current formula, they would just get one shot themselves by anything era adjacent. And one can imagine why all the issues mentioned above might be a problem.

Anything below -4 combat strength is utterly irrelevant for damage taken, and anything above +16 for damage dealt. A difference of 30 combat strength is the exact same as a difference of 16. Anything inbetween +8 and +15 combat strength is also practically wasted, since it will still take two hits to kill an enemy unit all the same.

Players should never be penalized for researching better technologies, or modernizing their army. And the only combat stat that exists should actually have said combat impact, always, and not just in two, three very narrow, edge-case scenarios.

Unit strength should matter, any difference in it. Yet, currently the difference that actually matters at all is kept to a very narrow range (-4 to +16 or a range of 20, which is often 1 era plus some bonuses), and even then the impact is not even the same for each point of combat strength, but wildly different. The combat strength formula being based on absolute - rather than relative - difference in strength also breeds issues by making flat modifiers more potent on weaker units, and is not predictably granular at all, varying greatly in impact per point and using artificial damage floors that do not appear to be a part of the formula at all, but rather something plastered on top of it, a band-aid.

Even between units of the same era, the damage floor at -4 makes melee units practically useless. Archers will win against warriors on an open field by kiting. They will win in an offensive siege by shooting them to death in their own city. They will win in a defensive siege by doing the exact same thing. They can just pick a defensible spot and force the warrior to come to them, and it will not only be half-dead by the time it does, but will have less combat strength in melee than the archer by then because of damage taken, fortifications and terrain bonuses - which do help if your units are weaker in CS than the enemy, but not if they are stronger - and let us not mention the fact that the warrior's attacks will invoke retaliatory damage while the archers' will not. Or the fact that the archers will shoot over several tiles and focus fire enemy units one by one, whereas melee troops do not have such an option.

In short - the current formula does not work in its present shape. Not only does it currently reach the damage floor at a difference of four points in combat strength, not only does it entirely invalidate fortifications and positioning and makes melee units almost entirely redundant over ranged ones, but worst of all it makes spamming ancient era to medieval archers over basically any unit from any later era the one true winning strategy, and not even because archers are that good - they could even have a Combat Strength of 0, and they would still do & take the exact same damage to and from most era adjacent things that they are fighting.

While the band-aid of anachronicity made it so that there is now a difference between a medieval knight and a tank against an ancient era archer, there is still usually no difference between said knight and a praetorian, halberdier or others such, and if one starts appearing - you just upgrade your units by one era, still keeping them comfortably obsolete by 2-3 eras.

The formula itself is flawed, and without changing it, any stat tweaks on combat strength modifications are quite meaningless and have little to no impact. None of the balance issues here are unavoidable though - far from it - all of them are mere side effects of the combat strength formula being such as it is, at the moment, and not the result of any singular unit (or a group of units or a strategy in making them, or even the +7 combat strength event that was nerfed) being overtuned.

How could one go about addressing these (quite gamebreaking) issues, then?

I believe changing the formula is strictly necessary long-term, because plastering patchwork fixes on a faulty foundation always just opens as many holes in balance as it closes. The easiest way would be to change the formula to weigh the relative combat strength of the two sides and multiply that by a damage base to determine the damage (note that this would probably necessize changing all terrain, fortification etc bonuses to percentage ones, rather than flat, but that would probably be a good thing regardless and would fix many other issues, on top of being easier and more intuitive to understand regardless)

For example, damage could be calculated as:

(20-30) * Combat Strength (A) / Combat Strength (B)

(The above is the example formula from the table atop the post).

Under this kind of formula, a difference of 50% in combat strength means that the stronger unit would always, on average, be worth between three weaker units (if every attack made involves retaliation) and two (in case its own does, while the enemies' does not, such as being pelted with arrows). Such a formula makes each percentage point of CS boost matter exactly as much (whereas in the current formula going from, for example, +0 to +13 CS means going from ~20 to ~63 damage (+43 or +3.3 per point), and then 3 more CS points somehow give another 37 damage (12.3 per point).

Unlike a flat value difference formula like the current, a ratio based one also has no artificial damage floors (or cellings) which makes every single point of combat strength matter, whether from the unit itself, fortifications, clever use of terrain, special unit abilities or empire bonuses - which in turn, encourages actually upgrading units and researching better ones. All of this, and more, is sadly simply not the case with the current formula, hence the thread.

I would also propose that each unit should come with an Armor Class or Armor Rating value, which would reduce the damage they receive by half if the Combat Strength of the enemy unit is below the unit's Armor Class or by 75% if it is below half of its value. That way you can differentiate tankiness between units of the same era without making their relative strengths to one another too out of whack, without making the system too complicated or unintuitive. This could serve to differentiate melee units from ranged ones in terms of tankiness, and would also extend to things like tanks, battleships and helicopters, protecting them from outdated weaponry (like arrows or melee weapons) without making them unbalanced to era-adjacent units.

Yutterh wrote:

Nice, I hope they implement this. Though they sis have weaker era units do less damage during the beta or so I'mtold. but people felt it was unfair if they get behind or something like that. Hopefully they revert it to what you have as it makes kore sense. But then again the Soviets would become ridiculously OP with their weapons factory lol

But that's the thing, even soviet weapons really wouldn't necessarily be overtuned with such a formula :)

Currently (besides what I wrote above), the formula also has the issue of lacking in possible granularity. You can at the very least add +1 Combat Strength per source of a buff, and it is applied to every unit equally, even though it does not have an equal impact, making different units very much not equal recipients of it.

The Soviets' weapons under the current system are broken because you can stack up ~40 of them and have your archers go from Combat Strength, say 30 with other layered buffs, to 70, which will allow them to fight everyone else's contemporary units on an even ground, (not that that matters overmuch due to how the current formula is structured). Or you can recruit actual contemporary units with 100+ strength (which is at the moment not much of a concern because they will steal get beat by three archers each all the same, but you can).

On the other hand, if you base it all on the relative, rather than flat, difference, you can then shift all modifiers to percentages safely. Not only does each and every percentage of combat strength actually mean what it says then (Having +10% Strength meaning you deal 10% more damage, rather than the current +2 or even a +10 often meaning no change whatsoever), but you can no longer buff anachronistic units to contemporary era levels, simply because their base is smaller. At the same time, while you can certainly improve your units to a huge degree, even building the above ~40 soviet weapons would only give +80% combat strength (If each one gave +2%, which would give +1 on up-to-date era troops with 50+ CS and less on obsolete ones), which at the end of the day would actually mean dealing 180% damage (relative to unbuffed, less when other bonuses are taken into account), and taking ~55.5% (100/180).

Which while powerful, does not break the game per se, it does not even reach the boundary of oneshots against formerly equal units (36-54 damage at a 180% ratio in CS), and it can't reach an artificial damage taken floor because there isn't one, either (11-17 damage at a 1/1.8 ratio). Furthermore, this approach allows for fractions to be properly taken into account in the calculations (even if they aren't displayed in the UI) which can swing the damage very slightly.

On the other hand, under the current formula, that same amount of combat strength (+40) will vs units of even two eras above your own (which don't exist while you are a contemporary soviet) result in you oneshotting them while taking the minimum damage. Potentially even if you use ancient archers which cost two orders of magnitude less in industry and upkeep than proper units.

shakee wrote:

May I add an additional option? Currently every possible damage roll has the same statistical chance to occur. Regardless if the Unit is fighting in the same era or vs. later eras. 

So:

Each era uses different dice size (+4) and are rolled for attacker and defender. (I excluded neolithic here)

Ancient: 2d10; Classical: 2d14, and so on

Damage Formula: <Attacker> / (<Attacker>+<Defender>] * <Base Damage>

• the Damage will align to the middle of the range, highs and lows are unlikely to occur
• combat strength of the unit may adjust the min. damage
• flat or percentage damage modifier for anti-tank vs armored; small arms vs armored are possible instead of fiddling around with the CS
• fighting against later era units will have a natural falloff the higher the difference is
• maybe add advantage or disadvantage to different situations (keep highest 2 or lowest 2 of 3 rolls)

....

Not sure if instead of changing dice size, adding more dice is better for later eras (+2 for each era), but is an alternative.

Hm, making the result tend towards the middle of the range would be a stabillizing factor in combat, however depending on how narrow the range is, it might not be very impactful, besides making small to moderate differences in combat strength matter even less (whether you deal 60 or 80 damage per hit, its still a two shot, its why in the current formula a +8 and a +15 act functionally the same - same damage taken and dealt). If you have damage tend towards the middle, then a say 70-110 range of damage would hardly ever one shot, and a 30-55 would hardly ever two shot enemies, whereas without such a stabillizing factor that can happen quite often.

Furthermore dice rolls ultimately fall into the pitfall of being quite a controlled range environment, albeit less so than a formula based on addition&subtraction, and they are not as easy a formula to understand for an average player as a combat strength ratio.

I think the advantage/disadvantage mechanic is intriguing though. Maybe certain unit types could have a natural advantage or disadvantage against others (like polearm troops have advantage vs cav, whereas cav has disadvantage vs them), and it would take the higher of two rolls (advantage) or the lower (disadvantage), Kind of like D&D, except for damage taken and dealt.

That would make for a sort of rock-paper-scissors system where certain unit types perform much better vs X, yet are countered in turn by Y which X tramples on. All while being easy to understand. I think hard and soft damage is a bit of a stretch in the earlier eras - however each unit could have an armor class value maybe, which halves damage taken from all enemies that do not have combat strength higher than that value? And being inside walls or such would provide a bonus to that. It would allow for diversifying tankiness between units, without outright differentiating damage and toughness as separate parts of combat strength, or artificially inflating HP.

I disagree, the current version with patch is fine.  once you are at 16+ you one shot them, and that probably means era differences that reduce their minimum damage.  Also the current version is Very good on understandability. (some units need to be tuned)

Krikkitone wrote:

I disagree, the current version with patch is fine.  once you are at 16+ you one shot them, and that probably means era differences that reduce their minimum damage.  Also the current version is Very good on understandability. (some units need to be tuned)

Very TL;DR in case what was written failed to convey the meaning.

The current formula does this:

- Ancient Era Archers deal and take exactly the same damage to and from: medieval era knights, industrial era construction teams (settlers with shovels) and contemporary era tanks. This makes them the superior choice to use across all ages past classical, due to unit industry costs scaling exponentially, as you can make ~186 Archers per Tank, but 3 Archers will already suffice to beat a Tank in combat. Said Tank will also cost 3 population much like 3 archers will, and so it is not even cost-effective in terms of pop.

- Units between +8 and +15 combat strength relative to the opponent are effectively the exact same in efficiency, because they will need two hits to kill the enemy, and will receive 5-25 damage per hit all the same. Therefore even if the values are within the very narrow range of 20 points where more or less CS does anything at all, even as many as 7 points of it can effectively change nothing.

You are free to think this state of affairs is fine, of course. I'm just making a very TL;DR version to make sure the point got across, whether it is agreed with or not. Also updated the main post with a pretty table to better illustrate what I'm talking about.

Thank you Crunbum for your work on this post.

I think you have right. The weak units do too much damages to the powerfull units (one or two more eras). I don't know what would be a good solution.

One of the key quote is here:


Ancient Era Archers deal and take exactly the same damage to and from: medieval era knights, industrial era construction teams (settlers with shovels) and contemporary era tanks. This makes them the superior choice to use across all ages past classical, due to unit industry costs scaling exponentially, as you can make ~186 Archers per Tank, but 3 Archers will already suffice to beat a Tank in combat. Said Tank will also cost 3 population much like 3 archers will, and so it is not even cost-effective in terms of pop.

Realistiquely, archers should do 1 damages to one tank. Not more. The cost comparaison between 186 archers and one Tank, is funny ^^

Promethian wrote:

Didn't they just tweak this in a patch so severely out of date units can't hit high rolls against modern units?

They did, but that is not actually the core issue, and its a bandaid fix for a deep flaw in the formula. If you look at the numbers, any more (or less) combat strength outside of a very narrow range does not affect combat at all, and even in said range there is huge swatches of effectively "dead" points, where you can gain +7 combat strength and the matchup will go the exact same way in every circumstance.

One problem is the uneven impact of combat strength per point (for ex. going from +0 to +13, you go from 20 to 63 damage (+43 or 3.3 per point), and then suddenly to 100 at +16 (+37 or +12.3 per point). This breeds balance issues (and is not intuitive, and neither are the results the current formula gives intuitively predictable for a human mind I feel), and then the far larger problem is that its a small, closed bracket of CS that matters, only - and you often hit its floor between units in the exact same era (like archers dealing floor damage of 5-25 against chariots) and always do it with one era ahead (archers vs swordsmen). After that, upgrading your units against an "inferior" opponent is pointless, because their efficiency stays the same, yet their costs skyrocket, making them far less efficient.

The damage floor is -4, and it is achieved by almost everything, no need for using fortifications, taking the high ground or anything - because you will still be taking capped damage anyway, which is 75% of the amount that you would be taking in a +0 str matchup. In other words, there is no way to actually reduce damage to any real degree, and that makes melee units quite pointless to use.

Archers will beat the same industry cost in Chariots on neutral terrain plains without any defensible locations (that they can always use), despite Chariots literally hitting them at 1 CS shy of the damage cap (+15 CS), and getting hit at the damage floor (-5, and floor is at -4 CS).

And then there is the fact that modifiers to combat strength are flat, instead of percentages, which unbalances units between eras even further - one of the more severe consequences of this is that you can buff long obsolete units like archers to literally oneshot tanks without even exploiting the game, by just building some soviet districts and playing as intended.

You can fix each and every side-effect of a flawed formula with a band-aid, or you can fix the formula itself and the problems will go away. I know its kind of preachy, and yet it is what it is.

I agree that there is an issue here. And that the issue is rooted in the damage floor (minimum damage a unit will do) being quite high.

Kaerbear wrote:

So, maybe I'm missing something, but don't you need 186 population to make 186 archers? Who are going to hit for 6 basically. Why would you use these? Do you actually win with this strategy? I'm curious how it works in practice.

Yes and no.
If you spam archers (or an appropriate cheap unit), the game will end well before tanks is one thing, they are merely a good example to point out the issue.

But higher era units tend to cost more than 1 pop. For example, tanks cost 3. And 3 archers will already beat a Tank because they do not hit for 6 (or at least did not when I first made this thread, I am not aware of the exact extent/formula of the penalty which anachronistic units now face - only that it restricts them to a lower bound within the damage range, so I cannot say for sure how many it would take, but certainly not more than 5 even if it were that severe - and the flaws in the core combat formula remain). Therefore you are paying zero industry and upkeep and the same amount of population to beat the opposition.

You can't make 186 Archers to beat a Tank because population. But you don't need to make 186. You just need to make 3, which is the same pop as the tank. And, whereas the opponent will be hard pressed to mass-produce units, you will just produce 50 in a single turn in one backwater city with hardly any industry even, converting all its pop to an army that will be able to fight 62 turns of the enemy's production in a similar city on even grounds. If completely anachronistic units turn out not to work now because of some patchwork bandaid? Well, you make slightly less anachronistic ones to much the same effect.

You don't need a standing army (within reason). You just make units instantly when you need them. You pay no upkeep, you don't waste industry (or pops jobs) on units that you will need 10 or 20 turns later. You just spam districts and if you suddenly find yourself needing 100 archers for a war - you make 100 archers in one turn.

And even if you make a respectable standing army, well... you need 3 Archers to beat one tank, so you now have the leftover 183 Archers worth of industry and gold to convert into districts and whatnot. And that's what happens every era past ancient. You get way more industry than other civs, because you are spamming obsolete units that cost almost none of said industry. Therefore you get more districts, more industry, more gold, more science, more food, more pops...

It snowballs.

Archers are possible to counter once artillery takes the field, via bombardment. But at this point you are so far ahead it doesn't even matter, and they will still clear other contemporary units with ease if the enemy doesn't bombard you. Furthermore, by that point you can take Soviets, build a bunch of their weapon districts, and have your Archers one shot their tanks while taking the damage floor themselves, because all the CS modifiers are flat instead of % so its not even hard, especially with how much you've snowballed by then.

Okay, hyperbole. But you can feasibly make your Archers (or a slightly better equivalent) be not one-shot by the enemy's contemporary units, and at that point they can bombard all they want.