Casualty Madness

[raizer]

delib attack, think assault-think taking ground
hasty attack, think meeting engagement, movement to contact

[Davekhps]

ORIGINAL: ComradeP
Also keep in mind that the forces in the AGS area retreat rather than surrender after many battles, which will mean you will have to take hundreds of losses attacking the same isolated unit a number of times. Somehow, their slightly higher morale/experience has a huge impact.

FYI, I've noticed that the propensity of isolated units to retreat vs. surrender is affected *greatly* by having available an enemy-controlled hex to retreat to. Meaning, if you really want to kill the maximum number of pocketed units with the least amount of your own combat power, try this:

1. First turn, form your isolated pocket, don't attack within the pocket.
2. Second turn, now that you have your proper pocket, you can attack within the pocket... but BEFORE you do this, use any free units (including 1 or 2 mobile units if you have them to spare) to eat away at all the enemy-controlled hexes within the pocket.

E.g., an isolated Soviet pocket may have five infantry divisions present, but the total number of enemy-controlled hexes in the pocket are ten. If you attack those five infantry divisions first, unless their morale and strength are exceptionally low, all you are likely to do is bounce those enemy divisions into retreats all around the pocket until they either finally surrender, or run out of hexes to retreat to.

HOWEVER, in our ten-hex pocket example, if you already move into those five clear hexes to take control of them, you've reduced the areas that the enemy can retreat to. Note that we're not talking about forming a completely new sub-pocket with an unbroken line of your units-- all you have to do is switch those empty enemy-controlled hexes to your control by moving your units next to them or through them.

Once this is done, enemy retreats will result in great numbers of surrenders-- enemy units can't overstack via retreats (e.g. two hexes, 4 units, 2 in each hex-- you force a retreat, only one enemy unit can retreat, the other surrenders).

One warning: I *did* notice a number of times, very few but they did happen, where a retreating enemy unit did retreat out of the enemy-controlled hexes in the new sub-pocket, into a friendly-controlled hex. Usually, this happened right next to one of my HQs, darn it [:@].

But the bottom line was that this was VERY rare, and a lot less frustrating to deal with than forcing enemy retreats all across dozens of enemy-controlled hexes in isolated pockets, particularly in the late game (i.e., the Soviets in 1941 will shatter and surrender far more easily than in the later game, making it more important to force surrenders on as few attacks within pockets as possible rather than wasting time and strength bump-chasing retreating enemy units across big open pockets).

[bwheatley]

ORIGINAL: Singleton Mosby

ORIGINAL: abulbulian

bwhealty also posted that he feels the loses to a hasty attack should be on average more than deliberate attacks. I agree and I think logically in most cases it makes perfect sense. Deliberate attacks have invested the time to scout best attack routes, plan for contingencies if things go wrong, etc. Thus lessening the causality factor.

I disagree with this suggestion. A hasty attack doesn't pack the punch a deliberate attack does because only a small portion of the unit is used for the attack. Hence the casualties are relative light. A deliberate attack is a prepared attack by an entire unit, mostly used against a stronger defense.

I respectfully must disagree.

Code: Select all

15.2.1.	 HASTY ATTACK
 Defined as “...an attack in which preparation time is traded for speed in order to
 exploit an opportunity,” hasty attacks will generally result in higher attacker and
 lower defender losses than a deliberate attack. A hasty attack will require the
 expenditure of three MP’s for a motorized combat unit and two MP’s for a non-
 motorized combat unit. Only a single stack of combat units can participate in a hasty attack
 and their Combat Value (CV) will be reduced by one half for all steps in which CV is calculated.
 Support units can only be committed from eligible headquarters units that have not expended
 any movement points during the current turn. Note that support units attached directly to
 combat units will always be committed to battles to which the combat unit is a participant.

ie coup de main. It says nothing about only a portion of the unit being involved. And while traditionally yes some of the coup de main style attacks by german units were just small portions of a unit that is no modelled in the game.

Code: Select all

15.2.2.	 DELIBERATE ATTACK
 Defined as “A type of offensive action characterized by pre-planned coordinated
 employment of firepower and manoeuvre to close with and destroy or capture the
 enemy,” deliberate attacks require the expenditure of sixteen MP’s by motorized
 units and six MP’s by non-motorized units. Multiple stacks of combat units
 can participate in a deliberate attack against an adjacent defending stack. Unlike a hasty
 attack, support units can be committed from eligible headquarters units that have moved
 during the current turn. In addition, Artillery combat units that have sufficient movement points
 remaining may participate in a deliberate attack from two hexes away from the defending unit.
 The artillery combat unit must be selected just as any unit would be selected to add into a
 deliberate attack (5.3.1). If all units launching an attack are artillery combat units that are two
 hexes from the target hex, then only artillery units from both sides can fire and no support,
 reserve or air group units will be added into the battle for either side.
 

[Joel Billings]

When attacking a unit, you always want it in a position where it can only retreat to a hex adjacent to an enemy unit. This ZOC to ZOC retreat will cause extra losses and greatly increase the chance of surrender if the unit is isolated. So move and attack in such a way to put the units in a pocket in this position.

No one has commented on the fact that a very large part of the 700 losses in the example above might actually be damaged elements that repair, leading the actual long term (KIA/disabled) losses to be much less than 700. This is especially true for high experience German units that tend to take lots of damaged results and less destroyed results, and given their high experience tend to repair more quickly. I could be wrong, but you need to study the permanent losses that result from these battles, not the "casualties".

[dwesolick]

Well, I've been cleaning up the "Lenigrad pocket" (finally got that "right hook" strategy to work, but man it was a grind!) in late fall 42 of my GC and I can only DREAM of suffering 500-700 casualties per attack.
I am routinely seeing casualties of 2000-5000 per attack. Now granted, most of these Soviet stacks are in level 5 forts, but they have been isolated for weeks and I am attacking each stack with overwhelming force (and total air control, over the Leningrad area anyway), and usually from 3 or 4 (sometimes 5) sides.

In a couple cases I even had results where I suffered 7000+ casualties while the (isolated....for weeks) defenders suffered around 1000 and held! I had to restart the turn...cheesy as hell I know [8|] but I just couldn't stomach such results. I know, I know, these things can happen in war....Thermopylae, the Alamo, Bastogne, but still.....

[gingerbread]

I have absolutely no source to quote, but didn't a large part of the Kiev pocket (after Guderian's escapade) surrender en masse?

[Flaviusx]

Joel, the recovery of damaged elements is an under the hood game mechanic with no easily quantifiable metric. It's therefore hard to analyze it in the context of disabled losses generally, even if they are representing the light and easily recovered wounds.

If the loss tables tracked this, maybe folks would be more comfortable and get a better handle on the larger picture. But for now it's a kind of black box. And even taking that into consideration, I still have to wonder about the recovery rate of the "hard" disabled losses.

[moses]

ORIGINAL: Joel Billings

No one has commented on the fact that a very large part of the 700 losses in the example above might actually be damaged elements that repair, leading the actual long term (KIA/disabled) losses to be much less than 700. This is especially true for high experience German units that tend to take lots of damaged results and less destroyed results, and given their high experience tend to repair more quickly. I could be wrong, but you need to study the permanent losses that result from these battles, not the "casualties".

I was planning on looking into this tonight as I have no turns to work on[:(].

This seems just like WITP where the same issue arose time after time. As I currently understand it, men don't actually die, instead elements get damaged or destroyed and this is converted into a casualty figure. So for example 200 damaged infantry squads might give the same casualty figure as 50 destroyed infantry squads. But the destroyed squads must be replaced. The damaged ones will repair in time without using replacements.

I'm wondering if the figures in the casualty screen (as opposed to combat reports) include damaged elements or only destroyed. If what you are suggesting in true then I'm thinking that after a combat report in which the german losses "700 casualties" you should be able to go to the casualty screen and see that fewer soldiers actually died. I'm hoping that the casualty screen matches with the production screen so that you can accurately assess if your losses are exceeding production.

[karonagames]

I'm thinking that after a combat report in which the german losses "700 casualties" you should be able to go to the casualty screen and see that fewer soldiers actually died.

You will and you do, if you run a combat then look at the casualty summary you will see that the recent casualty column shows the number that matches the the combat results report, but the permanent loss column is less, and this represents the damaged squads that were included in the combat results report.

edit: you will see occasional discrepancies, but I think I put it down to pilots crash landing when they got back to base, and this not being shown in the combat report.

[abulbulian]

ORIGINAL: Joel Billings

When attacking a unit, you always want it in a position where it can only retreat to a hex adjacent to an enemy unit. This ZOC to ZOC retreat will cause extra losses and greatly increase the chance of surrender if the unit is isolated. So move and attack in such a way to put the units in a pocket in this position.

No one has commented on the fact that a very large part of the 700 losses in the example above might actually be damaged elements that repair, leading the actual long term (KIA/disabled) losses to be much less than 700. This is especially true for high experience German units that tend to take lots of damaged results and less destroyed results, and given their high experience tend to repair more quickly. I could be wrong, but you need to study the permanent losses that result from these battles, not the "casualties".

Well with a 1% repair rate, I think damage is almost as good as lost. Still the numbers make no sense given the odds and the probably difference between the attacker's moral/exp vs the defender's. Keep in mind also that the terrain was light woods and clear weather. [:(]

[abulbulian]

ORIGINAL: dwesolick

Well, I've been cleaning up the "Lenigrad pocket" (finally got that "right hook" strategy to work, but man it was a grind!) in late fall 42 of my GC and I can only DREAM of suffering 500-700 casualties per attack.
I am routinely seeing casualties of 2000-5000 per attack. Now granted, most of these Soviet stacks are in level 5 forts, but they have been isolated for weeks and I am attacking each stack with overwhelming force (and total air control, over the Leningrad area anyway), and usually from 3 or 4 (sometimes 5) sides.

In a couple cases I even had results where I suffered 7000+ casualties while the (isolated....for weeks) defenders suffered around 1000 and held! I had to restart the turn...cheesy as hell I know [8|] but I just couldn't stomach such results. I know, I know, these things can happen in war....Thermopylae, the Alamo, Bastogne, but still.....

Yeah, I too experience these types of inflated loses in my AI game when trying to clear out ISOLATE (2-3 turns) units the Leningrad area. I had all the ports along Lake L. My attacks suffered 2k-4k and the units in most cases did surrender in the end. So I'm not saying it's way out of whack, but something is a bit fishy here.

[karonagames]

the 1% figure has nothing to do with damaged units being repaired. The 1% is referring to the Disabled pool which is created separately from damaged units.

This section explains how the damaged units return; the disabled pool return is a separate sub-segment of the logistics phase:

18.2.1. THE REPLACEMENT SEGMENT
The replacement segment consists of four subsegments;
return of damaged ground elements,
return of excess support squads, refit, and normal
replacement (4.2). Units must be in supply to
participate in any part of the replacement segment.
First, half of all damaged ground elements from
units are returned to the production and manpower
pools and made available to return as replacements,
however, only eighty percent of the manpower from
the damaged ground elements goes to the pool; the
other twenty percent being placed on the disabled
list. All other things being equal, returning ground
elements have a better chance of going back to
their original units. Next, excess support squads will
be returned to the pool and their manpower made
available to build up other ground elements.

[Joel Billings]

The casualty screen left most column should match the combat report. In these cases damaged elements are counted as 1/2 of a kill. So 3 damaged 10 men squads will show as 15 casualties. The next column, the permanent turn losses are counting only those elements that are destroyed. The first attack I made in a turn showed casualties of 650 men, while permanent losses were 376 men. Of this 364 were KIA and 12 were disabled. So what really happened was:

364 KIA
12 Disabled
548 in damaged elements (reports this as 274 casualties - 364+12+274=650)

When the the damaged elements are returned to the pool, 20% of the men are considered disabled instead. Also, 2 damged elements may combine into 1 ready element, in which case the lost element is considered KIA. Also, the damaged element may be repaired in the unit in which case there are no permanent losses.

So in this battle with 650 causualties, you could end up with as few as:

364 KIA
12 Disabled

But more likely for the Germans, perhaps half of the elements will be sent back to the pool and half will be repaired, and in this case you'd end up with:

364 KIA
66 Disabled (12+(274*.2))
220 Wounded but returned almost immediately to duty
274 Effectively out of action for a short period due to the elements damage (these could also be considered lightly wounded as well)

Notice that the number of disabled is very low relative to the total number of men wounded or put temporarily out of action.

This is just one comat result, and what exactly happens with the damaged squads is impossible to say, but it needs to be considered when discussing "disabled" totals and return rates.

[Zovs]

Still the numbers make no sense given the odds

In the first example I don't see how 33,788 attacking German soldiers losing 631 makes the number and odds out of whack. It seem quite reasonable to me. Also, the Soviets had 16,729 men and lost 7,322 again I fail to see the issue here.

I too experience these types of inflated loses in my AI game when trying to clear out ISOLATE

Again, I fail to see these types of losses as "inflated".

I found this source on the web and the attached image as well:


[font="courier new"]German casualties in Barbarossa 1941

According to Das Deutsche Reich u. d. 2. Weltkrieg, Bd. 5/1, p. 885 :
[Verluste d. Ostheeres ohne AOK Norwegen] :

|--------------------------------------------------------|
| 1941                                                   |
|--------------------------------------------------------|
| Month | Dead   | Wound.  | Miss. | Illness + Frostbite |
|-------|--------|---------|-------|---------------------|
| June  |  8,883 |  29,474 | 2,701 | 54,000              |
| July  | 36,144 | 120,409 | 8,435 | 17,000              |
| Aug.  | 39,334 | 142,807 | 7,672 | 34,000 (estimated)  |
| Sept. | 27,492 |  99,635 | 4,560 | 56,800              |
| Oct.  | 23,804 |  86,396 | 3,562 | 66,000 (estimated)  |
| Nov.  | 16,945 |  64,116 | 2,990 | 73,092              |
| Dec.  | 14,752 |  57,747 | 4,594 | 90,907              |
|--------------------------------------------------------|
| 1942                                                   |
|--------------------------------------------------------|
| Jan.  | -->  87,182 (combined)   <--         | 127,718 |
| Feb.  | -->  88,014 (combined)   <--         |  85,086 |
| Mar.  | --> 105,042 (combined)   <--         |  62,858 |
|========================================================|
| Total:  --> 1,082,690 (combined) <--         | 567,461 |
|--------------------------------------------------------|[/font]

[bwheatley]

ORIGINAL: Joel Billings

When attacking a unit, you always want it in a position where it can only retreat to a hex adjacent to an enemy unit. This ZOC to ZOC retreat will cause extra losses and greatly increase the chance of surrender if the unit is isolated. So move and attack in such a way to put the units in a pocket in this position.

No one has commented on the fact that a very large part of the 700 losses in the example above might actually be damaged elements that repair, leading the actual long term (KIA/disabled) losses to be much less than 700. This is especially true for high experience German units that tend to take lots of damaged results and less destroyed results, and given their high experience tend to repair more quickly. I could be wrong, but you need to study the permanent losses that result from these battles, not the "casualties".

Maybe we need another line on the report to show damaged units. So it's less misleading?
That's probably my one big irk with the game. Not enough visibility into long term review of numbers. To say oh lets look at long term losses per turn. Etc.

Also when you play PBEM the manual says hit 'l' first thing to see your losses but i think that is boned. T7 i see 0 german losses and only 35k for me.

I'm sure you guys will get it figured out. But with wargame nerds like use the more reports and numbers the better.

[jomni]

ORIGINAL: Cannonfodder
Taking losses while clearing out a pocketed but fairly determined enemy should be costly...shouldnt it?

Yup just like every island in the Pacific.

[raizer]

ORIGINAL: jomni

ORIGINAL: Cannonfodder
Taking losses while clearing out a pocketed but fairly determined enemy should be costly...shouldnt it?

Yup just like every island in the Pacific.

disagree with this strongly...you talking 1941, where russian pockets were a disheveled mob-rabble...not unlike the french 55 infantry division who ran from the posts outside of sedan in 40. Comparing the russians to the Japanese soldiers who were fighting and not surrendering is folly. Japanese didnt surrender...russians surrendered in the millions. The two did not have equal morale and fighting tenacity, particularly in the early periods of the war, prior to the defense of mother russia being introduced as a theme to the soldier, during the periods when the soviet soldiers were not sold on protecting the motherland. During the time when the field grey soldiers in their hob nailed boots were undefeated and looked upon with a kind of awe by the peasant army (this would change temporarily outside moscow in 41 and change for good at Stalingrad) and We talking summer 41, a time when the soviets did surrender in the millions, did have a mistaken belief that maybe that captivity wouldn't be so bad-not Okinawa or Peleliu where the Japs were fighting, as bushido warriors for their divine emperor.

The soviets woke up in time, around winter 41 for a short time and then winter 42-but not in summer 41 where they did, in many locations, literally just lay down when encircled

[Oleg Mastruko]

Lotsa axis fanboy myths in this thread....
&nbsp;
I apologize in advance for using the term "fanboy" again - it appears that some people are especially sensible to this word.
&nbsp;
How else do I call stuff like "Russians laying down when encircled" or the very name of this thread "casualty madness". Words as strong as "madness" being used for what to me, and many others, seem like pretty normal casualty figures.....
&nbsp;

[raizer]

actually Im a sov fanboy at heart but you cannot tell me that russians didnt surrender in the millions during the first couple months of the war-so Im not "sensible" to the word-you are fine with the soviet fighting spirit in the summer of 41 being equated with the troops from Japan holding out in coral caves-im not, not a biggie.  I got the game like you, and will enjoy playing russians-

[Oleg Mastruko]

Russian morale in 41 was actually very variable and "impulsive", for lack of better word (I am not a native english speaker). Many authors were commenting Germans never knew whether a unit in front of them will surrender easily, or will fight like berserkers or Japanese fanatics&nbsp;till the last drop of blood and the last man alive&nbsp;- lots of examples for both ends of the "morale spectrum" so to speak.
&nbsp;
Putting that aside, this thread, like others before it, failed to convince me that there is anything wrong with the basic combat and casualty model. Perhaps the casualty reports are not clear or user friendly enough, but the model appears to be fine. Calling it "madness" is certainly overreacting.
&nbsp;
Also, reading AARs, seems to me problem lies with people who thought playing Axis will simply be easier, or who constantly complain about problems from only one side (you guessed it - Axis).
&nbsp;
&nbsp;