Who would win, unarmed meat wave or 1945 Panzer Division with extra King Tigers? The answer may surprise you...

[Beethoven1]

Have you ever wondered what would happen if 30,000 unarmed Soviets were to launch a human wave assault against a full TOE 1945 Panzer division? Also what if all the 81 Panzer IVs in the Panzer division were swapped out for King Tigers, would that make a difference?

There have been a couple people testing these sorts of battles.

Various things have been tested such as T-26s against King Tigers, or Panzer 35(t) against IS-3. The general conclusion we are led to from these battles is that equipment really does not matter very much at all in the combat, at least not in comparison to other factors such as morale and experience. It is not that much of an exaggeration to say that morale and experience, along with the quantity of men, pretty much singlehandedly determines battle outcomes.

Below we will examine a battle in which 30,000 unarmed (and I mean literally unarmed, no weapons at all, whatsoever, not even a single rifle or pistols) Soviet militiamen attack a full TOE 1945 Panzer division (with King Tigers).

It probably goes without saying that for such an attack we will need a custom general, namely Yevgeny Prigozhin:



Our leader will have a 0 political rating, 1 morale rating, 1 infantry rating, 0 air rating and 0 naval rating. HOWEVER he also has to have a 99 mech rating, because he can drive to Moscow faster than Guderian. And he needs a 9 initiative rating for daring to be so bold as to drive to Moscow. And he gets a semi-decent admin rating because his logistics had to have been ok to manage to drive on Moscow from Rostov in a day.



Prigozhin's meat wave in this attack were armed with precisely... nothing... nothing at all. All the Soviet rifle squads looked like this. Just 11 guys with 0 "devices" (weapons):



All they had is their bare hands. But with their bare hands alone, they somehow managed to damage 2 King Tigers and 1 Panzer IV. And in addition, other German AFVs were lost in "retreat losses." Technically it was not just their bare hands though. Private Ivanov apparently approached a King Tiger and began gnawing on its tracks with his teeth, which grievously damaged the German AFV. For his bravery in this action, Private Ivanov was awarded the Order of Lenin:



The Germans fired a bit, but not that much, against their unarmed foe:



And here you can see the Soviets fired NOT AT ALL:



0 shots fired, 0 hits of anything against any of the German elements whatsoever.

Notice in the screenshot, however, that it still lists the Soviet weapons (7.62mm Mosin-Nagant rifle) etc. That is because there seems to be some weird bug in the editor where even if you make all the elements have 0 devices, it will still use the original devices in combat (and yes, we tried restarting in case restarting the game would update data and the scenario data was "locked"). We found that the only way to actually make Soviets not fire at all was to not only edit the squads to have 0 weapons, but also to edit the weapons themselves to have 0 of anything. So for all those weapons, I also edited them to have 0 of all stats, to ensure that Soviet militia would not fire at all. For example, here you can see I changed the stats for PPSh-41 to be all 0s. It is the same thing for all the other Soviet weapons.



Here is what the Soviet unit looks like (after the attack):



Notice it has ONLY rifle squads and support squads, nothing else (and the rifle squads are all unarmed, as are the support squads).

Finally, if you look closely at that screenshot, you can find the explanation you have probably been waiting for. The Soviet militia has 99 morale and 99 experience.

Whereas the 1945 full TOE Panzer division was 50 morale and 50 experience (also both sides had 130% ammo, supply, fuel, and support at the start of the battle).




Now, of course, this battle is the extreme case, since the Soviets here are literally unarmed, but are extremely high morale (but it is the extreme case that shows how the mechanics at their core work). However, in other cases that were tested, there were other very strange results.

Here for example is a test by Albert N/Wiedrock in which a 1941 Panzer division with obsolete Czech light tanks does an attack against a Soviet 1945 Mechanized Corps. The German Panzer division was the same as it normally is on turn 1 of a Grand Campaign, whereas the Soviet mechanized Corps was 50 morale/50 experience:



With better Soviet leader:







In either case, the German Czech-tank 1941 Panzer division beats up on the late war Soviet mechanized corps and destroys its T-34/85 and SU-100s etc.

The main conclusion here is that equipment doesn't matter. Yes, it is an overstatement to say it doesn't matter at all. But basically experience and morale and number of men in a battle determine the outcome. You could change the equipment values to more or less anything and get similarish battle results as long as you don't change the morale and experience.

The only way equipment really matters is if the morale/experience differences are a lot smaller between the two sides of a battle (changing that might make the combat engine significantly more realistic), and if the attacker doesn't stack an excessive number of men into the attack (if they do, they basically get a guaranteed win).




Back to the unarmed meat wave attacks, however, for one last final thought. If you repeat the same attack with only 80 morale and 80 experience for the Soviet meat wave rather than 99, you get this:



On the other hand, what if you keep the Soviets at 80 morale and 80 experience, but just increase the number of men they attack with 10x? If so, you get this:



The final test (an in some ways the funniest but most informative) is from Caedus, who tried this same attack (with the 10x more men), but where the Soviets have only 15 morale and 15 experience:



In that case, the Soviets win the battle, but take.... 82k casualties...

There is obviously a massive difference here between how well unarmed Soviets with 15 morale and 15 experience do and how well they do with 99 morale and 99 experience. With the 15 morale, 82k losses. Whereas with my original test, the 99 morale Soviets only lost a few hundred men in their unarmed meat attack.

Now, think about that for a minute or 2... Is it logical that the morale and the experience of UNARMED attackers (or even just poorly armed attackers, perhaps more realistically armed only with rifles and some grenades) should so drastically alter how many casualties they take when (successfully) attacking a 1945 Panzer division with King Tigers?

Sure, there would be some difference because the more experienced troops would be better able to use cover. But I do not think they would realistically take anything like 200x less losses. Instead, the main determinant of how many casualties the Germans manage to inflict ought to be the morale and experience of the Germans. They are the ones doing the firing, and the only ones with weapons in the battle, after all.

[AlbertN]

Without fishing the screenshots - tried also the swap with Experience 90 '41 Soviet Tank Division, stripped of the KV and T34s, and upped to 300 T26 - beating down a '44 TOE GD Division kept at Morale / Exp 50. (The inversion of the Czech tanks shanking the Su100s and T34/85) -- Just to ensure it works both ways.

[EwaldvonKleist]

Testing things systematically with the editor is a very good thing, thanks for doing this. Inserting extreme variables in formulas is a good way to test the model.

WitE1/2 calculates hit chances, firing rates and CV by consecutive unit mrl, unit exp, leader mrl and leader inf/mech rolls.

If you consider that usually XP=Morale since XP converges to the Mrl. upper bound over time, then Morale enters the CV and loss calculations to the second and perhaps even third and fourth power. Now, 0.99^2=1 (almost), 0.5^2=0.25

In addition, there are jump points in the function since certain extra modifiers only fire above/below a certain exp or mrl level, further increasing the huge difference between high and low mrl units.

The system overall is highly non-linear.

[Sammy5IsAlive]

I'm not sure how productive it is thinking too hard about the combat engine. The information the player is presented with is on two very different levels, and the two do not necessarily complement each other. On one level the engine is resolving and presenting the outcome of a week's operations, normally at at least a division level over a c. 200 square km area. At another level the engine is presenting the outcome of engagements at an individual unit level. I think it is important not to conflate the two levels too much. So the element vs. element combat only forms a small part of what goes into determining the overall outcome (i.e. elements lost or damaged during this combat will no longer contribute to the final CV calculation).

Further to this, I would argue that because of this "two level" issue, trying to test the combat engine by entering extreme values may not give meaningful results (meaningful in terms of giving us an insight into the combat mechanics overall).

With this in mind I am not sure that

Have you ever wondered what would happen if 30,000 unarmed Soviets were to launch a human wave assault against a full TOE 1945 Panzer division? Also what if all the 81 Panzer IVs in the Panzer division were swapped out for King Tigers, would that make a difference?

is the right question. For me the question would be more like

"What would happen if 30,000 elite Soviet troops spent a week sharing the hex area with a barely trained Pz Division. Would they be able to engineer a situation whereby the Germans were forced to withdraw. In this scenario how much difference would the German division possessing King Tigers make to the outcome. Similarly what difference would the issued personal armament of the Soviet troops make?"

This post is not intended to argue for a particular answer to that latter question.

I do appreciate the time and effort that goes into tests like this. The player base has managed to uncover various bugs/unintended consequences by looking forensically at the various information that the game provides re combat and other aspects of the game. I just think that sometimes it is possible to get overly caught up in drawing conclusions from the minutae of what is meant to be a large-scale game, particularly when you are basing these conclusions on extreme situations that are far from anything the engine deals with in "normal" gameplay.

[Wiedrock]

WitE1/2 calculates hit chances, firing rates and CV by consecutive unit mrl, unit exp, leader mrl and leader inf/mech rolls.

If you consider that usually XP=Morale since XP converges to the Mrl. upper bound over time, then Morale enters the CV and loss calculations to the second and perhaps even third and fourth power. Now, 0.99^2=1 (almost), 0.5^2=0.25

That's exactly the "issue". Things multiply into one another.

1. CV (that one is observable and clear)
   55MOR/EXP Unit with 100CV becomming GUARDs makes it 100CV/0.55/0.55x0.65x0.65=139.7CV
   So just looking at CV this means it's not an increase by 10%, but by 40%.
   (completely ignoring that it also means getting a better TOE additionally)
2. Increasing Combat Performance
   Then the numbers also/supposedly change how many elements can shoot at all, how often they shoot and how often they hit. I guess noone knows the impact in this calculation with certainty, but assuming for a second that it is all three of the factors, which again multiply into one another, this is another layer of "3-timing" the numbers/effects.
   
   So now we have an MOR increase by 10MOR, followed by 10EXP as a result, increasing the CV by 40% and the Combat performance by X%.
   But I feel like there is one more layer not mentioned/missing.
3. The "camouflage effect" (MY GUESS)

   In that case, the Soviets win the battle, but take.... 82k casualties...
   
   There is obviously a massive difference here between how well unarmed Soviets with 15 morale and 15 experience do and how well they do with 99 morale and 99 experience. With the 15 morale, 82k losses. Whereas with my original test, the 99 morale Soviets only lost a few hundred men in their unarmed meat attack.
   
   Now, think about that for a minute or 2... Is it logical that the morale and the experience of UNARMED attackers (or even just poorly armed attackers, perhaps more realistically armed only with rifles and some grenades) should so drastically alter how many casualties they take when (successfully) attacking a 1945 Panzer division with King Tigers?
   
   Sure, there would be some difference because the more experienced troops would be better able to use cover. But I do not think they would realistically take anything like 200x less losses. Instead, the main determinant of how many casualties the Germans manage to inflict ought to be the morale and experience of the Germans. They are the ones doing the firing, and the only ones with weapons in the battle, after all.

   For me it seems like added to the two previously stated there is another layer which decides whether a element is actually hit or not. As we can see, the high MOR/EXP meat wave suffers almost no casualties while the 15 MOR/EXP unit suffers many. Altough both not being able to shoot themselves (so not DIS/DAM/DES the defender).
   The reasoning to me is not fully clear

   In addition, there are jump points in the function since certain extra modifiers only fire above/below a certain exp or mrl level, further increasing the huge difference between high and low mrl units.

   The question now is if this is related to
   1. actual "jump points/tresholds"
      only these,...or these and additionally one of the following points
   2. or if there is a linear "camouflaging"
      (like 90EXP means all hits you would suffer multiplied by 0.1)
   3. or some "comparing values"-thing
      (like 80EXP vs 65EXP means 65/80=0.81 -> the enemy will only hit 81% of his rolled (Point Number 2) shots.
      whereas the other side would hit 80/65=1.23 -> 23% more than against a unit with same EXP.)
   As stated, points B and C are a guess (and formulas are just so you know what I genrally mean by this), maybe they are mentioned somewhere in the manual, but then I must have forgotten about that.

Just taking points 1 and 2 makes it already skyrocket, factoring in the combat internal "hit more, get less hit" system inceases it further. If my observation/assumtion pointed out in point 3 holds true that one would further snowball into that.
Even only taking points 1 and 2 as being given, this is already reason enuff why combat for many is completely unintuitive, since units turn by 180° once they become "elite" or a side suddenly overwhelms the other becuase of a +5NM to one and -5NM to the other side set at certain dates.
This big of an impact on performance undermines the equipment/TOE changes, especially when considering that these were already part of/reasoning why a unit was "elite" or why a NM (representing training and so on) would increase/decrease.
The changes in increments of 5 seem logic and easy to remember/understand, like "yea paratroopers were better trained, so 5% more EXP seems alright" and additionally they were more of the "volunteer type unit, so 5% more MOR is reasonable too". But both multiplies into one another and multiplies again, and again and again.

I'd argue using a more conservative approach to NM changes and bonus NM by being Elite or by being some special type of Unit (Mt., Cav., Airb....). But that's just me. People want their super heroes.

---

With better Soviet leader:

Rokkosovsky rly did change the odds quite a bit.

In either case, the German Czech-tank 1941 Panzer division beats up on the late war Soviet mechanized corps and destroys its T-34/85 and SU-100s etc.

The main conclusion here is that equipment doesn't matter. Yes, it is an overstatement to say it doesn't matter at all. But basically experience and morale and number of men in a battle determine the outcome. You could change the equipment values to more or less anything and get similarish battle results as long as you don't change the morale and experience.

Somewhat that's what happened IRL, outmaneuvering badly trained, badly coordinated tanks essentially.
So imo there is a point in making MOR/EXP have an effect/impact, tho the question is if it should be that huge. Eastern front had extreme results, I guess this high multiplying nature was the best way to achive such extreme results in game.

[Beethoven1]

Further to this, I would argue that because of this "two level" issue, trying to test the combat engine by entering extreme values may not give meaningful results (meaningful in terms of giving us an insight into the combat mechanics overall).

To me the issue is what it implies about combat in 2 different types of battles:

1) Battles with a very high morale unit against a weak/normal one (for example, SS divisions against a Soviet infantry division or a Soviet guards corps against Romanians or something).

Basically what we can see from the test is that a very high morale wins just by showing up, even totally regardless of any combat at all or any equipment at all.

I certainly DO agree that morale/experience should be a major factor, and you can also make a strong case it should be more important than equipment. But it should not be the ONLY factor, and these tests are basically showing that it pretty much IS the only factor.



2) Attacks with a large amount of numerical superiority. If you just attack with a sufficient number of men, then you automatically win. Even if you do not fire at all at the enemy. You win solely for existing.

This is also an indication that the ability to stack too many men into one attack probably breaks the combat engine. And so perhaps the ability of Soviets to stack as many men (combing rifle corps with attached rifle brigades etc) in one hex/battle is just too much.

But from 1943-45 or so, what is an attack with, say, 150-200k Soviets against 15k Germans...? That can be basically EVERY SINGLE ATTACK that the Soviets do. All they have to do is just attack with more men against a single German division, and they can just automatically win, literally 100% of the time, even totally regardless of anything that happens in combat whatsoever. They can get completely massacred in the actual combat, but they actually win nevertheless and the Germans retreat. Even if substantively the Germans do not need to retreat (i.e. even if the Germans have taken 0 losses and still have plenty of ammo left).

If you can automatically win every single battle just by existing, then that means that each turn you can automatically advance at least 2 hexes or so.

That means that if you are 50 hexes away from Berlin on turn 100, you will be in Berlin on turn 125 just by virtue of the fact that you exist.

Which is probably also relevant to how Stamb managed to take Berlin in 1943 in his Stalingrad to Berlin game.

[AlbertN]

I do not have a big problem with a high morale / exp unit beating a low morale / exp unit so long both are armed; it helped me understanding though how to fine tune stuff for the mod (or assume I am quite right, for now, in the direction taken).

The problem was when the unarmed ninjas take over the low training / quality unit. That's the excess where the system breaks, I think. Since the system cannot envision some sneak operation where the ninjas seize weapons from the enemy and use them ...

It also help me underline the removal of brigades from multi role is much good and needed - at least Soviet end. (and it may be case Axis end too over time)

[Teo41_ITA]

Thank you for pointing out something that I could confirm in my independent tests about unit morale and experience (setting the experience Grossdetuschland Mot. IR to 250 makes it worth having as a CU).

Based on my unreported results, I have to add that:
- unit morale automatically self-adjusts during the first movement step (I had units with morale of 600 just to see it rolling back to 99-95 after being moved for one hex)
- any change into TOE will result in the new elements lowering the overall exp (and CV) of the unit;
- sending units to any TB automatically resets their exp to be closer to the national morale one.

[EwaldvonKleist]

..........................

Indeed, if you have mrl=exp as usually is the case and treat them as one thing, then during the battle phase, morale influences losses taken and inflicted. Then morale influences CV of remaining elements to the second power and thus strongly influences if unit retreats and takes retreat losses, and then will influence how big the retreat losses are.

Regarding the discontinuities, WitE had some combat events that e.g. lowered retreat losses, which only fired above a certain mrl or exp level. Not sure if WitE2 has some too, but I would be surprised if not.

[ErickRepie]

This confirm my uneasiness during playing ... armour div lost to infantry division in non fortified - clear - hex.

[Joel Billings]

While it's always interesting to see these kinds of tests, I'd point out two things about the combat engine:

1) Since CV values are based on the type of squad, all rifle squads have the same base CV value, not matter the weapons. It's assumed that a rifle squad will have some kind of weapon. You won't find rifle squads with no weapons in the game. The weapons themselves have an impact on the losses and disruption that the squads can deal to the enemy during the combat. So having better weapons does impact things. Obviously the elements with no weapons will cause no combat losses (the only losses to the defender would be to weapons breaking down in combat and retreat losses).

2) There are major penalties for having more units in a battle. Having one unit that contains 280k men again goes around this part of the code. IIRC that part will make it easier to cause losses to the larger force and harder for each of the larger forces elements to engage in combat. So again, not as valid a test as having created 10 rifle corps that contained the men, which would probably have some serious density issues.

That said, yes, morale (doctrine) and experience are critically important. You can't really explain many of the WWII campaigns without this (so many to mention, but how about the British defeat of the much larger Italian force in late 1940 North Africa as one example).

[CaedusZ]

2) There are major penalties for having more units in a battle. Having one unit that contains 280k men again goes around this part of the code.

@Joel
Could you please elaborate on what stacking means and how it influences battle outcomes? As far as I understood your answer, stacking comes from adding more divisions, brigades, regiments and batallions into a battle independend of the number of men, guns and tanks belonging to them? Does this specific stacking penalty only influence battle outcomes (in terms of CV) or also performance in battle?

[Wiedrock]

While it's always interesting to see these kinds of tests, I'd point out two things about the combat engine:

1) Since CV values are based on the type of squad, all rifle squads have the same base CV value, not matter the weapons. It's assumed that a rifle squad will have some kind of weapon. You won't find rifle squads with no weapons in the game. The weapons themselves have an impact on the losses and disruption that the squads can deal to the enemy during the combat. So having better weapons does impact things. Obviously the elements with no weapons will cause no combat losses (the only losses to the defender would be to weapons breaking down in combat and retreat losses).

2) There are major penalties for having more units in a battle. Having one unit that contains 280k men again goes around this part of the code. IIRC that part will make it easier to cause losses to the larger force and harder for each of the larger forces elements to engage in combat. So again, not as valid a test as having created 10 rifle corps that contained the men, which would probably have some serious density issues.

That said, yes, morale (doctrine) and experience are critically important. You can't really explain many of the WWII campaigns without this (so many to mention, but how about the British defeat of the much larger Italian force in late 1940 North Africa as one example).

You are right, the test just show the impact of MOR/EXP. But as with the "normal Division vs Corps test" there was no stacking forgotten.
About MOR
For me I do not like Guards getting +10MOR as I don't like SS getting +15MOR later on. For the SS it should be reflected by Equipment and potentially massed additional MOR/EXP up until that point. For the Guards it is that thousands of units were called Guards even freshly built units. Imagine Germans making 20% of their Divisions "elite" by giving them a new name and a cuff title. And as we can see the impact of MOR is huge so those extreme changes by being "elite" are concerning to me.

About the Stacking (diminishing returns):
Obviously some of the "crazy" tests are flawed by not having multiple CUs/stacking penalties, but if you shoot 0, a penalty can't reduce your chance to hit.
But stacking rules generally needs to be looked at imo. I think for equal sizes of CU/SU and using about the same attachment rules (Western Front/early eastern) everything works out as you say it, things make sense. Stacking penalties cause less hitting and more being hit, thats working as intended, but in later Eastern Front it's often not changing the outcomes. If you can amass 300k in an attack, you do not care about being hit more/shoot less, since if you win, you won't suffer RTR losses. That's how the 1943 in Berlin StB meat wave was possible (why would you go around if you can go through?).

For earlier periods of the game, stacking should be alright (somewhat like Western Front), but once Corps CUs arrive numbers change drastically. Since in WITE2 you can assign 3x5k Men Rifle Brigades to 30k Rifle Corpses (so +50%manpower). Those 3 Brigades (since being a SU) do not contribute to Stacking penalties, while they would add 5 Stack Penaltiy each as an CU.

Some interesting note to history I recently came across.
There were plenty of different Reduced Strenght TOEs the Soviets used, starting from mid 1943 and going till the end of the war (Soviets preferred to built more and more Gun/Mortar/AFV units, so manpower for Rifle Units was reduced vastly). So there were TOEs for 4k-6k Rifle Divisions. Squad sizes went often down to 6-8 Men. Meanwhile in game there's 32k Corps with 3x5k Brigades (nothing reduced) assigned with 11Men Squads.
Side note to the increase to 11 Men squads in late 1943 (in game) - which historically can only be found in a single "Reduced strength TOE from Aug 1943", caused someone to decide to use 11Men Squads starting from 09/1943 for Soviets in WITE2. But this increase of Squad size was bound to less Squads so a Reduction of manpower density per Division/Corps/Regiment.

Combined you get...

• Large non Reduced TOE CUs (massing of Manpower in single CU)
• Large 11Men Squad sizes never seen (more CV, more Rifles per manpower invested)... about ~10% too much CV for a "non Reduced TOE" Rifle Corps
• up to +50% of manpower without this adding Stacking penalties (3 Brigade SU)
• one side getting more and more of Guards (+10NM)

= Berlin 1945 impossible.

Yes Soviets have limited manpower, but the game cares about Force in the individual battle. There is no incentive for a Player to go through Rumania and Hungary if he can force his way to Berlin in a straight line with a completely unhistoric composition/density of Forces, using 300k+ Manpower attacks. Similtaneously there is no incentive for Soviets to build smaller SUs which I feel is actually sad.
The supply system could reduce the advancement, but that it doesn't has been shown.

[Joel Billings]

Stacking adds up the value of all the units in the attack based on the size of the unit. IIRC, once you get over about 3 division equivalents modifiers start to kick in (I think it uses the 12=corps, 9=division, 5=brigade, etc. value system). IIRC, they make it less likely that each larger force unit will fire, and I think it also makes it easier to hit their elements when fired at. At one point, this modifier was even larger, and you had very large concentrations getting hit very hard by the modifier. That was reduced some, but I don't remember if that reduction came before or after the release of the game (I think after release during one of our combat system adjustments, but at least 2 years ago). If my memory is correct, this means that if the attacking force was in a realistic number of units, it would suffer because of it. The modifier does not impact the CV values, just the firing, IIRC.

[Beethoven1]

Stacking adds up the value of all the units in the attack based on the size of the unit. IIRC, once you get over about 3 division equivalents modifiers start to kick in (I think it uses the 12=corps, 9=division, 5=brigade, etc. value system). IIRC, they make it less likely that each larger force unit will fire, and I think it also makes it easier to hit their elements when fired at. At one point, this modifier was even larger, and you had very large concentrations getting hit very hard by the modifier. That was reduced some, but I don't remember if that reduction came before or after the release of the game (I think after release during one of our combat system adjustments, but at least 2 years ago). If my memory is correct, this means that if the attacking force was in a realistic number of units, it would suffer because of it. The modifier does not impact the CV values, just the firing, IIRC.

...

The modifier does not impact the CV values, just the firing, IIRC.

I think this bit here which I picked out to put in a separate quote is the key. I also searched through the manual yesterday and found 0 reference to any stacking penalty effects other than on firing.

But remember that in the tests in this thread that the attacking meat wave did literally zero firing. So any penalty that reduced how much they fire would reduce their firing from zero... to zero...

This means that (unless there is some other unknown or undocumented effect of stacking), that there would be zero difference in the results of these test battles if we were to attack with 10 corps of unarmed men and the test where I had it for simplicity in 1 single unit. In both cases there would be zero firing by the meat wave.



This probably also interplays with why Soviet logistics seems to be too good, especially in PVP games (less so in AI games). Human Soviet players will do massive attacks. As a result of the stacking penalty, this means they fire less. However, they still win the battle despite not firing very much due to the fact that the additional meat always adds CV (even if it does nothing at all in combat). But since the Soviets are firing less, they use less ammo, which means their supply is better than it would be if they fired more.

Whereas AI Soviets do not do such massive attacks as much and are not as impacted by the overstacking penalty on firing, and consequently fire more, which means they use more ammo and have worse logistics as a result.

So the way the stacking penalty currently works is a buff to Soviet logistics.



Here is what I think happens in the actual game, where Soviets don't literally fire 0, but also are not particularly combat effective (until later war when their morale goes up anyway) ---

Despite the overstacking penalty, Soviets are incentivized to attack with as many men as possible, despite the overstacking penalty, because attacking with more men virtually ensures a win, regardless of how the actual combat goes. This means:

1) You are virtually ensured that your units will gain +1 wins towards guards - and the more units you attack with, the more get guards progress.

2) You are virtually ensured that the Axis will suffer retreat losses and that you will not, which ensures a reasonable K/D ratio even if your units do very little in combat.

3) Every unit that gets a win in the attack has a good chance for its morale to go up. And as we have seen, morale/experience is probably the most important factor in combat, so the more you can get it up by having close to 100% wins on your attacks, the better things will get and the more you snowball.

4) The enemy likewise has its morale go down, and this is virtually guaranteed since it is virtually guaranteed that they will lose the battle.

5) Generals randomly gaining a stats point depends on their win/loss ratio. So by doing huge meat wave attacks, Soviets can improve their generals skill more (on average) and subsequently get better leader rolls in future battles.

6) The initial large meatwave attack, with the certain victory it provides via superior CV/superior meat, ensures the destruction of German forts and causes them to retreat. This means they can then be hit with a follow up attack. The follow up attack doesn't necessarily have to have as many men committed and where maybe the Soviets do actually care about the actual combat more and don't want to overstack as much) to inflict excessive losses on the beaten fortless Germans.

Like this for example in the gasteris vs etc game:





Note this is without the new reduced rout losses option, but even retreats from follow up attacks can be devastating for Germany.



I think the potential solution is obvious - make the stacking penalty directly modify CV, rather than having it only influence firing. Because firing is not what determines the outcome (win/loss) of a battle. CV is.

[Joel Billings]

What is the combat effectiveness of that 8700 man infantry division? That unit starts the 43 campaign with roughly 17k men and an unadjusted CV value of 10 (so 100 in combat). With a CV value of 9 in combat, it sounds like it's less than 10% combat effective. You have a point that the game may overweight quantity (although it also possibly overweights very high and low quality), but you just can't analyze examples like these without looking at the details.

As for a penalty other than on firing, I can't remember for sure if it is easier to hit large quantities of troops, but that was the feeling I remember from my tests when we were making these adjustments.

The upcoming new game option would definitely reduce the manpower losses on the routed units.

[Beethoven1]

What is the combat effectiveness of that 8700 man infantry division?

Gasteris sometimes attacks the same German division 3 or 4 times in a single turn with repeated follow up attacks.

So IMO the more relevant question is what was the combat effectiveness of that 8700 man infantry division at the start of the turn. It could have been considerably higher.

Personally I don't exactly know because I am not playing that game myself and can only go by whatever screenshots they post (which is not everything).

FWIW it is also very normal for the CVs/combat effectiveness of German units (and Soviet units) to plummet dramatically after they have retreated a single hex, even if they took hardly any losses in the first battle.

Spolier alert, but what is happening currently in that particular game is both of the two players are trying to aggressively grind each other down as hard as they can and inflict high losses, with the result that both OOBs are declining sharply in 1943 and both sides have a lot of weak units.

[Beethoven1]

Here is a better example I found in the AAR which will give you a fuller picture of how this happens (screenshots crudely grey out parts of the map which could help et if he saw them).





So for example in this case, the German infantry division started out the turn with 14073 men and 247 guns (and 24 AFVs from attached Panzerjaeger battalion).

At the start of the 3rd attack, it was down to 11525 men (lost 993 more in the 3rd battle), 146 guns (38 more lost in the 3rd battle) and 9 AFVs (6 more lost in the 3rd battle).

So that is how this happens.

Mind you, et (Axis player) is doing the same to the Soviets as best as he can also, so it is happening both ways.

[AlbertN]

These are the total men though, not the combat capable men.
By now the German division has most of these men accounted for into 'damaged' ground elements; and a bloated amount of that manpower as well will just be into support squads that are 66% of a German TOE.

It means that after the 1st defeat the ratio worsens as the combat troops gets hit and the support elements not so much.

So if at the beginning there are like 9-10k out of 14k in support, at the end it will be like 8-9k in support and 1k of combat elements.
That's why Axis / Germans are so brittle as well, among other things.

Russians have a 33% ratio of Support Elements to their max ToE, making the Manpower far more efficient in terms of being on map.