Combat mechanics seriously flawed

[Dalwin]

The casualty rates for naval battles are just as bad. We see too many battles with doom stacks facing off, but regardless of how large the battle is far too high a percentage of ships will end up being sunk before the battle is over and far too high a ratio of those will be destroyers compared to other types. Some serious tweaking of the balance is needed in naval engagements.

 

[KevinG]

The 40 width "minigame" has been acknowledged by devs that it's something they are looking at getting rid of - but this and the fact that 1 large Div > 2 smaller divs (due to armor, breaktrough, piercing values) has been said to be a hard thing to fix properly since averaging all forces on the same side in a combat will lead to other problems.

Its incredibly easy to fix the 40 width spam. Make org additive between divisions and not averaged. Then make every unit divide its firepower equally to every enemy unit on the frontline. Now a 40 width has twice the org of a 20 width but it won't overkill enemy divisions as easily as before because its damage gets spread out. Make the amount of equipment and the effect of support companies scale to division size so that support art for a 20 width costs half as much as for a 40 width but gives half the stats. Now division size becomes a personal preference instead of mandatory 40 width tank spam. Oh and rebalance stats for every unit because now defensive points become more useful since they will always get used up instead of bypassed completely like in the old system.

 

[Novat]

https://steamuserimages-a.akamaihd....419/19DDB4D05A63FD0973FF1D8EADAA081AF9BE6162/

Tell me about it. The issue revolves around the fact that defensive stats, breakthrough and defense. Can completely negate 100% of the damage. The system compensates by adjusting the next roll, so that if the previous roll resulted in 0 damage, the next one has a higher % chance to result in damage. However, the system is utterly incapable of dealing with units that have very high stats, resulting in near 0 losses. I.e combat ends prior to one side taking any damage at all.

Furthermore in that picture. I only took losses during the very first battle, since the AI attacked with full org. Because the AI don't wait for full org, consequent battles last even shorter, resulting in even fewer deaths on the Italian side. It is also very easy to bait the AI into attacking you, by placing a deceptively weak force on advantageous terrain.

 

[fabius]

I don't min max, yet have way more zero 0 casualty battles than not. Even when taking them, my casualties will be absurdly low.

I checked the battle stats again, and it really does look like the buffs add up too much. Especially those for defence and breakthrough. Plus, I reckon reasonable thinking on attack stats would be that things like veteran, general add too much, especially when culminating.

I thank you guys posting in this thread, as it's reassuring that more players are noticing this major issue. (It is a major balance issue right? )

 

[Alex_brunius]

Tell me about it. The issue revolves around the fact that defensive stats, breakthrough and defense. Can completely negate 100% of the damage.

I'm pretty sure that is not true. I am fighting with defense or breakthrough values higher then the enemy attack all the time, and I take damage.

If you look into the defines combat balance values used in the game you will discover that units have 10% chance to get hit when defense/breakthrough surpass enemy attack stats, and 40% chance to get hit from all attacks that surpass defense/breakthrough.


What causes the 100:1 casualty ratios in most situations is a combination of this 4x damage, and the fact that your AI enemy is attacking at very very bad odds which lowers their own attack values to near zero ( Attacking into mountains, across rivers and into forts ). Throw in air superiority which reduce the AI divisions defense/breakthrough even more (50%+), and other modifiers like dig in / experience / generals / doctrines / better equipment for your divisions and it gets even more uneven.

It's common to see AI divisions with 10% attack efficiency charge into defenders with 250% attack efficiency ( 25:1 casualties before taking breakthrough/defense into account ).

 

[fabius]

I'm pretty sure that is not true. I am fighting with defense or breakthrough values higher then the enemy attack all the time, and I take damage.

If you look into the defines combat balance values used in the game you will discover that units have 10% chance to get hit when defense/breakthrough surpass enemy attack stats, and 40% chance to get hit from all attacks that surpass defense/breakthrough.


What causes the 100:1 casualty ratios in most situations is a combination of this 4x damage, and the fact that your AI enemy is attacking at very very bad odds which lowers their own attack values to near zero ( Attacking into mountains, across rivers and into forts ). Throw in air superiority which reduce the AI divisions defense/breakthrough even more (50%+), and other modifiers like dig in / experience / generals / doctrines / better equipment for your divisions and it gets even more uneven.

It's common to see AI divisions with 10% attack efficiency charge into defenders with 250% attack efficiency ( 25:1 casualties before taking breakthrough/defense into account ).

That's something like my initial impression. It's the culmination of many modifiers.

Tricky to balance them all. But a quick and dirty solution that I mentioned before could be to have the combat modifiers only affect a % of the divisions.

Just for example, most combat is attrition (in the normal casualties sense not the supply attrition mechanic in Hoi4) to some degree. So 1:1 strength losses for 50% combatants. The other 50% has benefit of modifiers. It makes sense if thought through.

The 50% is just an example and the simplest. There could be different options with it. Eg Urban, forest and jungle more attritional than plains for obvious reasons.


Special troops could improve the % getting the buff stats.

 

[Alex_brunius]

That's something like my initial impression. It's the culmination of many modifiers.

Tricky to balance them all. But a quick and dirty solution that I mentioned before could be to have the combat modifiers only affect a % of the divisions.

Just for example, most combat is attrition to some degree. So 1:1 strength losses for 50% combatants. The other 50% has benefit of modifiers. It makes sense if thought through.

The 50% is just an example and the simplest. There could be different options with it. Eg Urban, forest and jungle more attritional than plains for obvious reasons.


Special troops could improve the % getting the buff stats.

Another part of the issue when it comes to using "attrition" is that the HoI4 implementation of attrition = equipment loss only, so not a single manpower can be lost from it meaning it's not a driver of casualties at all ( In today's HoI4 ).

 

[Zwirbaum]

It is more of a compounded (is that the correct word?) issue.

First let's delve into the issue number 1 - connected to the imbalanced results - AI.

AI as was often mentioned - performs attacks with almost zero chance of success constantly with all factors being against it. (Low planning bonus, bad terrain modifiers, fortifications, lack of the MIL high command boosts, bad equipment/manpower count in divisions, lack of air superiority, being backwards with research, rigid and not the greatest templates etc.)

Now let's see how constant attacks tie even more to the 'negative' factors. Constant attacking means you do not get planning bonus build up, equipment/manpower goes down etc.

If AI divisions will perform attacks over and over, with having some distance to the front it may even result in them missing/not getting replacement equipment and manpower.

Equipment and manpower also have double dipping effect on division stats. Division with 90% equipment (depending on what is missing etc.) will get less stats (not necessary 90%). Division with 90% manpower also gets 90% of combat stats that it would have from equipment.

So division with 50% combat stats from equipment (let's say division of 10 INF with only 500 Inf EQ) and 50% manpower will get only 25% of the combat stats -> this makes it even worse for constant AI attacks.

AI will often not get that much of a breakthrough (mostly infantry attacks) and as such it will deal even more serious blow to their manpower-equipment problem because of the relatively high ORG of the infantry divisions.

Attrition also plays a role in this (player will minimize attrition suffered to the divisions, resulting in less equipment lost and lower % stats missing) while AI will often disregard the issue.

And there are probably more and more factors that lead to the imbalanced result. (combat stat inflation, multiplication of all the modifiers etc.)

 

[Alex_brunius]

multiplication of all the modifiers etc.)

Disagree a bit on that one because while multiplication benefits positive modifiers (players) it also benefits negative modifiers (AI), perhaps even more so.

If the AI attacks with -50% modifiers from two sources, and the modifiers were additive instead the AI would attack with -100% = 0% efficiency, but currently it attacks with "only" -75% = 25% efficiency. That is infinity more, and IMHO worth it having the positive modifiers multiplicative.

 

[Zwirbaum]

Disagree a bit on that one because while multiplication benefits positive modifiers (players) it also benefits negative modifiers (AI), perhaps even more so.

If the AI attacks with -50% modifiers from two sources, and the modifiers were additive instead the AI would attack with -100% = 0% efficiency, but currently it attacks with "only" -75% = 25% efficiency. That is infinity more, and IMHO worth it having the positive modifiers multiplicative.

It is a double edged sword to be honest but stacking positive modifiers is much easier and leads to even more imbalanced situation. There could be the case of having either additive and multiplicative modifiers depending on the situation, or bringing the minimum % of combat stats no matter of the modifiers, which could be modified by certain extra variants (forts lowering the minimum %, air superiority etc.) This way positive stacking would be less of an issue (+10% and +10% is +20% instead of +21% and so on) and you'll no longer be able to penalize enemy too much.

**** EDIT ****

It also comes from the current combat stat inflation and values together with the modifiers. It could also use value rebalancing for both positive and negative modifiers.

 

[fabius]

Another part of the issue when it comes to using "attrition" is that the HoI4 implementation of attrition = equipment loss only, so not a single manpower can be lost from it meaning it's not a driver of casualties at all ( In today's HoI4 ).

I've edited my post. I meant attrition in the normal casualties sense not the supply attrition mechanic in Hoi4.

I still think all the modifier stack way too much.

Doesn't matter what training level, general skill, better equipment- units should still take significant casualties when attacking down an urban street covered by AT and Machine Guns, or taking an arty or Katusha barrage etc etc etc

 

[xXNightEagleXx]

There are so many things messed up here that is not easy to analyze them all. However i believe that one of the most strange one are the whole bonus and penalty system. It leads to ludicrous values and sometimes it doesn't even make sense.

For me the easiest to criticize are skills (both division and commander). Skills should affect human factor, right? The capacity to hold a good performance for longer at the same amount of supplies, against a lower experienced, and the capacity to better exploit the battle field and equipment thus improving movement, defense and attack. Unless i'm wrong 'soft attack' basically represents the firepower...guns...so why the hell a skill/experience would improve 'soft attack'? It should improve things like organization and supply or maybe even a new variable but anyway things that has directly relationship with the human side and not equipment side. Currently you can have division xp that gives you +75% + a 5 level commander that provides a +25%, which will end up by having ludicrous soft attack amount. This alone can be the source of various issues.

On the other side organization is a harder to analyze because on the paper it seems right, 2x20 offers double organization than a 1x40 which is expected since it should reprensents the ability of the division to fight properly and the more manpower/vehicles a division has the harder should be to keep the organization up thus extract the max from that division. Specially in current implementation where division moves and engage as a whole blob. However we all know that in practice it offers no difference because 40 width will have less org but so much SA and Defense that will shot 20 width target before organization starts to matter. Maybe it is necessary an organization and supply penalty that is directly related to the amount of battalions, manpower/equipment and or regiments of same type (as multiple of module of 5 for given battalion type). By doing so adding more battalions, specially when over historical values, should provide bigger penalties to simulates increased commanding and logistic complexity.

 

[xXNightEagleXx]

BTW i forgot to add that those bonus values are just too high....probably it even makes tweaks harder.

 

[Crowarior]

From my experience problem is in huge debuff for attackers (They drop to something like 10 soft attack) and huge buff for defenders (They get like 400-500 defence). You simply cant attack with those stats unless you sorround them and outnumber them heavily or use tank breakthrough divisions. Basically tone down both buff/debuff for attack/defence.

 

[SpeedKatMcNasty]

Well let's clear something up.

Offense is much stronger than defense in HOI4.

This is for a reason. If defense were stronger than offense the game would just stall, much like it often did on the Soviet border in HOI3, the classic wall defense. If the Germans found a way through it was over, if not, everyone just sat around.

 

[Dalwin]

Well let's clear something up.

Offense is much stronger than defense in HOI4.

This is for a reason. If defense were stronger than offense the game would just stall, much like it often did on the Soviet border in HOI3, the classic wall defense. If the Germans found a way through it was over, if not, everyone just sat around.

It might be more accurate to say that in HOI4 offense is stronger than defense unless the AI is attacking.

 

[jaredstanko]

The 40 width "minigame" has been acknowledged by devs that it's something they are looking at getting rid of - but this and the fact that 1 large Div > 2 smaller divs (due to armor, breaktrough, piercing values) has been said to be a hard thing to fix properly since averaging all forces on the same side in a combat will lead to other problems.

10 width is OP if you can maintain more defense than your enemy has soft attack because all divisions have the same ORG. Regardless of size... which is why 20 width is normally optimal for defense. Same stats, double the org of 40 width. Which are perfect for attack because it's the best way to have more small attack then your enemy has defense.

Combat width meta is not only the dumbest thing ever, but the meta surrounding org is too. Ideally, 8 10 width, 5 16 width, and 4 20 width would fight very close to identically. Where in this scenario, 16 width sucks because in most settings it doesn't fill combat properly. And 20 width has double the attack and defense of 10(per division, not total) 10 has double the org of 20(total, not per division.) If the 10 width has more defense than the 20 has attack, they fight the same, but have twice the org, and will last twice as long, and reinforce quicker (but not twice as quick, there's a signifigant stacking penalty on reinforce rate. But it's still a major boon for 10 width)

Its the absolute dumbest thing in the world, and I can't wait for it to be fixed.

 

[EntropyAvatar]

Its incredibly easy to fix the 40 width spam. Make org additive between divisions and not averaged. Then make every unit divide its firepower equally to every enemy unit on the frontline. Now a 40 width has twice the org of a 20 width but it won't overkill enemy divisions as easily as before because its damage gets spread out. Make the amount of equipment and the effect of support companies scale to division size so that support art for a 20 width costs half as much as for a 40 width but gives half the stats. Now division size becomes a personal preference instead of mandatory 40 width tank spam.

You'd also need to do something about how additional width is penalized (current penalties are pretty severe). I'd make the last division (the one that puts it over-width) bear the penalties, in proportion to the fraction of it's width that fits in the combat line (e.g. if combat with is 80 and 70 is already used, another 20 width division operates at half attack and defense).

I'd also divide firepower based on the width of the target unit. So if a 140-attack unit is facing a 10, a 20 and a 40 width, it's attacks divide up 20,40,80.

Oh and rebalance stats for every unit because now defensive points become more useful since they will always get used up instead of bypassed completely like in the old system.

I'm not sure how it would play out. I mean, with the current system, each division is potentially the target of every division on the other side. Every point of def counts when you are in danger of being overwhelmed. If you only get "your share" of the enemy firepower, then maybe you can tailor your def to how much firepower you think you are going to need in a typical situation.

 

[billcorr]

and has really bad templates.

The reason I hit "respectfully disagree" is that the A.I.'s templates have improved. The Paradox team has dedicated effort into making the templates better.

To characterize the A.I.'s templates as "really bad" is a subjective observation.

As an observation, "really bad" could be true...but there could be a bit of exaggeration or a better way to describe A.I. templates in "Death or Dishonor."

 

[Nuklearius]

#