The Dupuy formula (so as not to get inundated by the IS-3 debate)

[JasonC]

Maas says "the western allies weren't conducting the war in a way that produced many POWs", meaning apparently, broad front vs. pincer movements. But the Allies in the west took more prisoners than the Russians did. The Germans had 1 million missing in Russia, a portion of whom were KIA rather than POW, but 3 million missing overall, before the final collapse. In other categories, 2/3rds of the losses occurred in the east, but in prisoners it was 2/3rds the other way.

He also says there were few pockets in the west. But 250,000 men were taken in Tunisia, 100,000 at Falaise, groups of 10,000 to 25,000 at a time in the fortress ports starting with Cherbourg, between 50,000 and 100,000 in the Colmar pocket, against the Alps, and on the order of half a million in the final Ruhr pocket. We can perhaps exclude the last as simply the end of the war.

Pockets account for between 1/3 and 1/2 of the prisoners taken in the west. Broad front captures make up in time operating whatever they lack in scale on each occasion. What you see in the figures is 5 figures every month the Germans are retreating (quite low before that), jumping to low 6 figures for occasional large pockets. Of course, veterans also have said it was much easier to surrender in the west.

[JasonC]

Jeff's Russian loss figures are perfectly believeable, but they do not tally with the ratio claims some have made. 28 million total casualties including wounds and non-battle are believeable. That many irrecoverable losses are not, and Jeff gives the figure of 11.3 million for that, which is about what I expected. His sources also give 3.5 million for German irrecoverables, which is 1:3 not 1:10. And I know the total battle German battle casualties, including wounded but not counting frostbite and disease, came to 6.5 million in the east, from other sources. With some non-battle included, that would probably amount to 1:3 in total casualties as well.

Moreover, his figures show half of the Russian killed and missing suffered through the 3Q of 1942 (5.5 million), thus the phase of the German offensive. German losses in that period were lighter than in the second half, of course. The feldgrau site has losses by month but overall rather than Russia (they also have Russia alone, but not by month). They give about 650K killed and missing for the Germans from Barbarossa to 3Q 1942. Other fronts were light in that period, and other non-recoverables aren't included in it. Those figures leave 5.8 million Russian IR losses for the Russian offensive part of the war, vs. 2.85 million Germans, which is 1:2 from Stalingrad on.

Overall, then, the ratios would be 1:10 in 1941 only, 1:5 in the German offensive portion of the 1942 campaign, then 1:2 average for the rest of the war. In other words, it was not their own "reckless offensives" that cost the Russians so much, it was the German offensives in the first 15 months of the campaign. Once the Russians were attacking, the loss ratio was down to the military age population ratio and the tank output ratio, and the Germans were being "bleed white" every bit as fast as the Russians were.

[ 06-14-2001: Message edited by: JasonC ]

[PzKpfw 1]

Every Russian historian who is dealing in Soviet losses, has or is, going out of their way to prove their numbers. What we do have our the major works trying to address this issue, as with Sokolov's 'Tsena pobedy' in 1991, and then the Krivosheev's group 'Grif sekretnosti sniat' in 1993 along with various other Russian historian's works Ie, Mertsalov, Volkogonov, Alekseev, Isupov, Kunetov, & Maslow etc.

The truth is, the actual cost of the war in Russian human loss terms will probably never be known, as records are incomplete and other stumbling blocks exist, as in the fact that during the Winter war with Finland due to Soviet measures taken against espopionage, Soviet troops had their pay books & id medallions taken from them, which only adds to the problem of assessing data on casualties during that time period as well as the fact their was not even an 'official' Soviet loss & burial accounting organization until March 1941.

By 1942 many Soviet troops still had not received ID medallions, or even pay books & the Commissar for Defense actually canceled the issuing medallions' altogether in an order in late 1942 & even discouraged paybook issue, as they feared the true scope of losses would become known by the general troops & result in more desertions etc.

Works cant even agree on MIA/POWs Ie, Krivosheev lists the number of Soviet forces MIA/POW from 1943 - 1945 as 604,000 while according to German records they took 746,000 during that time period.

While they also can't agree on Front Force dispositions, Ie, at Kursk, Krivosheev lists the Central Forces as of 05.07.43 as 746,000, while Sokolov lists Central Front as 704,000 on 05.07.43. Krivosheev lists the Soviet irreversible losses for 1942 at 3,258,216 while Volkogonov puts the 1942 IR losses at 5,888,236, Volkogonov's monthly IR loss breakdown for 1942 is 1.8 times higher then the Krivosheev groups figures, etc.

As to the Soviet manpower mobilization in WW2 official figures including the peace time forces was 34,476,700 men of which 3,614,000 were used in non combat positions Ie, economic positions, & non Com positions within the military as of 01.07.45 11,390,600 men were in the Soviet armed forces with another 1,046,000 in medical facilities and another 425,000 Soviet POWs returned from captivity.

Sokolov states that the official call up data may be in error of 2 to 4 million effectives if or not it includes the over 4,000,000 volunteers of which over 2,000,000 were conscripted & that the official mobilization numbers do not reflect men who were drafted directly into front line combat units, which was believed to be in the millions, Sokolov puts the final mobilization at 42.9 million with the addition of an Soviet miscalculations in manpower Ie, volunteers conscripted party formation transition to the armed forces, & direct conscription into frontline units at 12,000,000 effectives.

As to the IR totals Sokolov brings up that the Moscow museum's WW2 database records contain information by name on 17,000,000 KIA/MIA Soviet troops, & as of 1996 another 1,000,000 names was to be added by the 50th anniversary of WW2. This data does not contain the names of the 6.3 million Soviet POWs, in German captivity, of which only 1,836,000 returned to Russia.

Krivosheev's group list's Soviet battle casualties from 1941 - till the end of the war as 14,685,593, and the number of ill as 7,641,312. This data was calculated by military medical facility records according to Krivosheev, medical losses according to Soviet troop reports was 15,296,473 battle casualties & 3,047,675 ill.

The discrepancy in Krivosheev's work is explained in that the former figures refer to all casualties, while the the 2nd refer only to active duty duty personnel which would explain the difference in the ill numbers, but not the difference of over 600,000 in active duty battle casualties.

Sokolov believes the data represents evacuated personnel and does not reflect troop numbers who were wounded & returned to their units or died in the field aid stations Ie, as of 1945 Soviet records show that over 1,191,000 Soviet troops had been wounded at least twice during their service the St. Petersburg Military medical museum holds records on over 32,000,000 Soviet personnel who were admitted to medical facilities during the war, their are no records to date on Soviet personnel who died, or were patched up by frontline Soviet medical Bn's.

These are but a few of the examples I can give that show just how much confusion their is concerning, Soviet loss studies.

Regards, John Waters

[ 06-14-2001: Message edited by: PzKpfw 1 ]

[Jeff Duquette]

I know there was some recent "controversy" regarding Krivosheev's brake down of specific Red Army operation casualty statistics. It had something to do with Krivosheev's choice to leave out the figures for Zhukov's disastrous Operation Mars. The gist was that a second addition of Soviet Combat Casualties is in the works that will apparently touch on new information, which has recently surfaced. Perhaps new archival information dug up by Sokolov.

John:

You edited out an interesting observation regarding the horrendous cost paid by the Red Army and how these casualty stats may be a reflection of Soviet Operational Art. To bad. It was an observation\opinion I share.

[PzKpfw 1]

<BLOCKQUOTE>quote:</font><HR>Originally posted by Jeff Duquette:

John:

You edited out an interesting observation regarding the horrendous cost paid by the Red Army and how these casualty stats may be a reflection of Soviet Operational Art. To bad. It was an observation\opinion I share.[/QB]<HR></BLOCKQUOTE>

Yes Jeff I was pressed for time,Also i would add that the Mars data was not available at the time anyway. The post listed just a few of the examples but was more to show just how muddled the whole feild is.

The new spin on the Russian Front has become the Soviet's evolved & became masters of operational art, the casualty data paint's a much difrent picture as loss rate climbed not declined during this period

of 'operational art'.

Sokolov & others are makeing no bones pointing out the cost during the suposed increase in Soviet performance.

Regards, John Waters

[ 06-14-2001: Message edited by: PzKpfw 1 ]

[Maastrictian]

<BLOCKQUOTE>quote:</font><HR>Originally posted by JasonC:

Maas says "the western allies weren't conducting the war in a way that produced many POWs", meaning apparently, broad front vs. pincer movements. But the Allies in the west took more prisoners than the Russians did. The Germans had 1 million missing in Russia, a portion of whom were KIA rather than POW, but 3 million missing overall, before the final collapse. In other categories, 2/3rds of the losses occurred in the east, but in prisoners it was 2/3rds the other way.<HR></BLOCKQUOTE>

Interesting. As I said, I don't have any sources, this is just the impression I have from the books I have read. Some questions follow, the first of which is:

-What are your sources for your data?

<BLOCKQUOTE>quote:</font><HR>

He also says there were few pockets in the west. But 250,000 men were taken in Tunisia, 100,000 at Falaise, groups of 10,000 to 25,000 at a time in the fortress ports starting with Cherbourg, between 50,000 and 100,000 in the Colmar pocket, against the Alps, and on the order of half a million in the final Ruhr pocket. We can perhaps exclude the last as simply the end of the war.

<HR></BLOCKQUOTE>

My intent was to say that the really large pockets were in the east. Tunisia is a special case, I was only speaking of western europe really. And with their backs to the sea I think we should expect a massive surender.

<BLOCKQUOTE>quote:</font><HR>

Pockets account for between 1/3 and 1/2 of the prisoners taken in the west. Broad front captures make up in time operating whatever they lack in scale on each occasion. What you see in the figures is 5 figures every month the Germans are retreating (quite low before that), jumping to low 6 figures for occasional large pockets. Of course, veterans also have said it was much easier to surrender in the west.<HR></BLOCKQUOTE>

That doesn't really make sence. "5 figures every month the Germans are retreating," I assume you mean Sept '44 to May '45. That's 8 months of fighting at 50,000 a month? That's only 400,000 for the entire retreat due to broad front. If your figures are correct that implies that 1.5 million Germans were captured in pocket (strategic level) actions.

-Can you please put up or provide a link to your figures? I would love to take my own crack at analysing the stats.

-Also, how many of your 2 million figure POWs were taken Jan '45 to the end of the war? I bet most of the POW discrepensy can be acounted for by the preference Germans had to surendering to the western powers vs. Soviet Union.

Thanks

--Chris

[Jeff Duquette]

This is just my opinion, and it's not likely to change (I'm a lost cause in this regard), but there is this "School of Glantz" arising which at times subtly and other times overtly encourages disregarding German History of the Russo-German War. This is somewhat of an odd approach in my mind to studying history of any event. In addition, it has been my impression that one of Glantz's guiding philosophies has been one of opening the door to the Soviet side of the story. This is a good thing. However I disagree with the philosophy of opening one door and at the same time subtly implying that folks should shut the German door.

I think there is also this impression that the German side of the war is old, and dare I say "not sexy" anymore. Old history = Unreliable history or some such thing. To a certain degree the further one moves away from the time period of the historical event being considered, the more one looses as far clarifying the human aspects of the event. Memory fades, individuals involved in the event die off.

I thought it quite interesting regarding Kips earlier comment on the Depuy Institute, and the implication that only post 1995 data is of value. Berlin wall and all…no need to delve into opening of Soviet Archives please. I just thought it an odd comment. Disregard all history pre-1995. Even post 1995 Russian Historians can't seem to agree on actual casualty statistics for the war. They'll loose themselves for 10 years in Russian Military Archives and then come to the conclusion that German statistics on the war were not so far off after all.

The reason why German histories of the Eastern Front have "traditionally" been excepted as accurate accounts of the struggle is directly related to the study of German war histories, memoirs, archival evidence etc etc of campaigns and battles fought against the Anglo-Americans. The campaigns\battles in North Africa, Sicily, Italy, Normandy, Lorraine, etc. Moreover, German histories of these aspects of the war in Europe jive fairly well with Anglo-American history of the same subject matter. The two sets of histories detailing the same event are mutually corroborating. Glimpses at military history from both sides of the hill so to speak. It is therefore not unreasonable to accept German accounts of Eastern Front combat as truthful, if not as detailed in the workings of Soviet tactical and operational practices as one might like.

This tendency toward less emphasis in German accounts, such as "Lost Victories" or "Panzer Battles", on what the Russians were doing operationally during "such and such" a battle is however not an exclusive trait of German Veterans turned writers. Histories written by participants of any event tend to write about there personal experiences of said event. Von Mellenthin was not typically present in Soviet divisional or Corps level staff meetings, so his tendency not to delve into the details of what Gen. Vatutin or Gen. Badanov may have been thinking when the Soviet 24th Tank Corps was destroyed by the 11th Panzer at Tatsinkaya should be seen in that light.

From the opposite extreme one only needs to pick up Dmitriy Loza's "Commanding the Red Army's Sherman Tanks" or "Fighting for the Soviet Motherland" to realize that Soviet Soldier historians focus on Soviets aspects of the war, and in the case of Loza are no less prone to memory lapse that Von Mellenthin. Case in point Loza (like Glantz) slips on the Panther armed with an 88mm. Perhaps Loza was Glantz have the same technical advisor.

[ 06-14-2001: Message edited by: Jeff Duquette ]

[Vanir Ausf B]

<BLOCKQUOTE>quote:</font><HR>Originally posted by Jeff Duquette:

I thought it quite interesting regarding Kips earlier comment on the Depuy Institute, and the implication that only post 1995 data is of value. Berlin wall and all…no need to delve into opening of Soviet Archives please. I just thought it an odd comment. Disregard all history pre-1995. Even post 1995 Russian Historians can't seem to agree on actual casualty statistics for the war. <HR></BLOCKQUOTE>

I very seriously doubt that Kip meant to imply that anything written before 1995 or written by Germans is no longer valid. I think he simply meant that before then we were mostly relying on German estimates of Soviet strengths and losses and now we have the numbers from the sourse. Unless you think the Germans knew how many men the Soviets had and lost better than the Soviets.

<BLOCKQUOTE>quote:</font><HR>They'll loose themselves for 10 years in Russian Military Archives and then come to the conclusion that German statistics on the war were not so far off after all.<HR></BLOCKQUOTE>

I'm not sure who these historians are, but it would appear from what I have seen that the Soviet numbers really do not match up with the German. I don't see any justification for dismissing the new numbers. If they make the Soviets look better than we had believed them to be for the past 50 years, so what?

[Jeff Duquette]

<BLOCKQUOTE>quote:</font><HR>Unless you think the Germans knew how many men the Soviets had and lost better than the Soviets.<HR></BLOCKQUOTE>

I think it’s quite evident that the Soviets don’t know how many casualties they suffered during the war. Even the “definitive work” on the subject by Krivosheev is being subjected to scrutiny and criticism. Krivosheev himself has indicated he is working on the 2nd Edition of “Soviet Combat Casualties” which will shed additional light on the subject and presumably rectify oversights in the first edition.

<BLOCKQUOTE>quote:</font><HR>I'm not sure who these historians are, but it would appear from what I have seen that the Soviet numbers really do not match up with the German.<HR></BLOCKQUOTE>

Well we’re all sharing numbers here, what German numbers are you referring to that so grossly exaggerate what has already come to light on this thread. Are these the German figures that indicate the Red Army suffered 32,000,000 casualties during the war? Because this number is clearly a huge exaggeration of the 30,000,000 that contemporary Russian Military Historians are coming up with.

<BLOCKQUOTE>quote:</font><HR>If they make the Soviets look better than we had believed them to be for the past 50 years, so what?<HR></BLOCKQUOTE>

Again that’s exactly the point, the numbers don’t make the Red Army’s performance look any better or any worse than it looked 50 years ago. This is the heart of what this thread is about. Russian Combat Effectiveness during WWII.

Lest we diverge from the intent of my last post, most German memoirs and histories of the War – at least those I have read -- don’t portrait the Red Army as a band of blithering idiots. Far from it.

[Michael Dorosh]

<BLOCKQUOTE>quote:</font><HR>Originally posted by JasonC:

Pockets account for between 1/3 and 1/2 of the prisoners taken in the west. .<HR></BLOCKQUOTE>

Source?

Definition of "pocket"?

You cite the example of the channel ports, but many German troops in other channel ports did not surrender until May 1945.

You need to define what you mean by a pocket before throwing out a simplification like this. How broad are you casting your net? Was the Bulge a "pocket"?

[Vanir Ausf B]

<BLOCKQUOTE>quote:</font><HR>Originally posted by Jeff Duquette:

Well we’re all sharing numbers here, what German numbers are you referring to that so grossly exaggerate what has already come to light on this thread<HR></BLOCKQUOTE>

My comment was in reference to a post Kip made in a different thread. I'll cut and paste here:

I see you mention the Dupuy formulas.

I am a huge fan of his analysis...

...However, when it comes to the Eastern Front, although his formulas still hold in every detail, it must be remembered that the “data” he entered into his formulas was almost completely wrong, with regard to the Soviets.

By this I mean that in order to use his formulas one has to “know” the force ratio, the size of the Soviet forces relative to the Germans, and, the causalities suffered by the Soviets. In the 1970s and 1980s the only people who knew the true size of Soviet forces, and even more so, their true causalities were the Russians, and they were not telling. Now we know both.

As it turns out Soviet forces were far smaller than was believed in the 1980s; also their causalities were far less. The Germans were simply not killing as many Soviets, post Kursk, as had been believed.

The average force ratio during the period July 43 to March 45, Soviet to German, was only 2.7:1. Not the 4-5:1 as had been believed. Also the causalities ratio over that period, Soviet to German, was only 1.64:1, not 2-3:1 as had been believed. (Remember these figures exclude the round-up of April-May 45.)

To cut a very long story short if you run the Dupuy formulas, using the new data, you get a combat effectiveness figure for the Germans against the Soviets, during the period given above, of 1.15, not the 1.8-2 or 3 given in Dupuy’s books.

This is no criticism of Dupuy; the figures were just not available in his day.

The conclusion is that the Soviets were very nearly as good as a Germans, battalion combat team V battalion combat team, post Kursk. Note that is “post Kursk” not at Kursk. At Kursk the Soviets really fought in the “old” style for the last time.

All the best,

Kip.

I wish I could give you the exact numbers, but Kip did not state where he got the numbers he used for his above figures. If he comes back to this thread hopefully he can shed some more light.

<BLOCKQUOTE>quote:</font><HR>Again that’s exactly the point, the numbers don’t make the Red Army’s performance look any better or any worse than it looked 50 years ago.<HR></BLOCKQUOTE>

Obviously, there does not seem to be universal agreement on this. I'm not in a position to judge who is right or wrong, as I don't know what raw numbers Kip used.

[Tero]

Here is a post on the subject from the OnWar forum by a guy called Wolf. Makes interesting reading:

>Many of the topics at this board ends up

>with a lot ”playing” with numbers from

>different sources, especially about the

>eastern front. What conclusions can be draw

>by looking at casualties?

>

>If I look at the losses in Normandy Campaign

>(Zetterling&Tamelander 1995)

>

>German 23.019 KIA, 67.240 WIA, 198.616 MIA.

>U.S 20.838 KIA, 94.881 WIA, 10.128 MIA.

>British 16.138 KIA, 58.954 WIA, 9.013 MIA.

>

>I don’t know if the Poles, French or

>Canadian losses are included. The 16.417

>lost allied airmen are not included,.

>

>likewise lost German airmen are not included

>what I know.

>

>Conclusion 1: The allies lost 209.592

>against 288.875 Germans a 1:1,38 ratio.

>

>So Tommy Atkins and G.I Joe saved Western

>Europe and inflict higher casualties despite

>attacking over the Channel against a

>defending foe. But the cost was high.

>

>Conclusion 2: The number of KIA and WIA in

>the German forces was 90.259 against 190.451

>a 1:2,11 ratio. So despite fighting against

>a foe with total air superiority and an

>unlimited(?) amount off artillery

>ammunition(very good logistics) and good

>information about the enemy(resistance,

>Ultra and aerial recon) superior tactics by

>the German forces gave Tommy and Joe a hell.

>

>Conclusion 3: Using the definition

>irrevocable losses (I suppose that it’s

>means KIA and MIA, should invalid soldiers

>be included to?) the numbers are 221.635

>Germans against 56.117 Allies a 3.95:1

>ratio. The Allied army soundly defeated the

>German army with relatively low casualties.

>

>Conclusion 4: The number of deployed troops

>of the Germans was around 600.000 and the

>Allied numbered 2.052.299. That means that

>every German soldier inflicted 0,34

>casualties(209.592/600.000) and every Allied

>soldier inflicted 0,14 casualties

>(288.875/2052.299). The old Landser (and the

>very young) with it’s better training (but

>shorter) fought still better than Tommy and

>Joe in 1944.

>

>Depending what I want to prove about the

>Normandy Campaign I can use any off the

>above conclusions. And that's only from one

>source.

>

>I think all four are pretty good

>conclusions...

Comments ?

[Vanir Ausf B]

<BLOCKQUOTE>quote:</font><HR>Conclusion 4: The number of deployed troops of the Germans was around 600.000 and the Allied numbered 2.052.299. <HR></BLOCKQUOTE>

What a bunch of nonsense. The Germans deployed a grand total of 600,000 troops in the ETO?! IIRC it was several times that many.

[Tero]

>What a bunch of nonsense. The Germans

>deployed a grand total of 600,000 troops in

>the ETO?! IIRC it was several times that

>many.

ETO or Normandy ?

[Vanir Ausf B]

Ok, I didn't see he was talking just Normandy.

But, really, why just Normandy? If you're trying to gauge overall combat effectiveness why only talk about a small part of the war in the ETO? The terrain in Normandy was a defender's wet dream.

If I took stats from just the Ardenes Offensive I bet the Germans would come out looking rather poor. That doesn't tell you much about the bigger picture, which is what I thought this thread was about, not any one battle or campaign.

In short, his conclusions are way too simplistic to be taken seriously. Not that what he says is totally wrong, just that he doesn't seem to take all factors into account, and it is of questionable relevance to this thread due to the limited scope, and the fact that he obviously isn't using the Dupuy formula.

[ 06-15-2001: Message edited by: Vanir Ausf B ]

[Tero]

>But, really, why just Normandy? If you're

>trying to gauge overall combat effectiveness

>why only talk about a small part of the war

>in the ETO?

The conclusion he did draw are admitedly a bit far reaching. But I think that when broken down into portions like this the events in ETO, indeed any front, can be handeled so that we get trends that can be said to be representative.

>The terrain in Normandy was a defender's wet

>dream.

And it showed.

>If I took stats from just the Ardenes

>Offensive I bet the Germans would come out

>looking rather poor.

Not really. The POW figures bring down the overall value. If you lose them the loss figures adhere to the general trend, ie. the Germans were able to get 1:1 or better loss ratios in combat.

This is why I advocate the POW figures should be left out, unless their inclusion is warranted by other related factors.

>That doesn't tell you much about the bigger

>picture, which is what I thought this thread

>was about, not any one battle or campaign.

>

>In short, his conclusions are way too

>simplistic to be taken seriously. Not that

>what he says is totally wrong, just that he

>doesn't seem to take all factors into

>account, and it is of questionable relevance

>to this thread due to the limited scope, and

>the fact that he obviously isn't using the

>Dupuy formula.

I agree, more windows are needed to complete the picture if more far reaching conclusions are to be drawn to form these "national biases".

But you do agree in principle that some formula, not necessarily the Dupuy formula, could be used to derive some factual evidence of "national biases" IF enough related variables are processed ?

The raw figures from Normandy do seem however to indicate that the Germans were able to extract better ratios, if not else then in the goulish way of cost efficiency in RM expended per enemy soldier killed or maimed. The Allies extracted better overall loss ratios but their cost efficiency in $'s per KILLED/MAIMED enemy soldier was very poor indeed compared to the German figure.

I am not saying it would be easy but with a little effort, a lot of hard facts and an established, well based guide line the comparison to establish the "national biases" could be done.

[Vanir Ausf B]

<BLOCKQUOTE>quote:</font><HR>Originally posted by tero:

But you do agree in principle that some formula, not necessarily the Dupuy formula, could be used to derive some factual evidence of "national biases" IF enough related variables are processed ?<HR></BLOCKQUOTE>

I see your steering this into the same "nationality modifier" debate going on in the other thread. I don't want to get caught up in that as it has been debated to death in the past and nothing we say here is going to change BTS's mind on it anyway.

Yes, it is possible to isolate and identify real trends particular to one side or the other, but these are only generalities and therefore should not be applied to a small scale game like CM.

Example: let's say that we accept as fact the Dupuy calculation that on average a German soldier in the ETO was equal to 1.2 American soldiers in combat effectiveness. Does this mean that this holds true for every unit in every battle? Hardly. There is a huge variation from unit to unit and at different times in different battles. The 1.2 figure is just an average, as all trends are. But if you make all German units in CMBO equal to 1.2 American units, then it becomes true in every battle with every unit. Not realistic. If you are making a scenario you can simulate this my making the Americans "green", so the game does allow this to an extent without forcing it upon you.

If CM were a strategic level game you may be justified in using nationality modifiers, but at the tactical level they make little sense IMO, unless the difference was overwhelming (10:1 effectivenes, for example).

[Jeff Duquette]

<BLOCKQUOTE>quote:</font><HR>Vanir Said: I wish I could give you the exact numbers, but Kip did not state where he got the numbers he used for his above figures. If he comes back to this thread hopefully he can shed some more light.<HR></BLOCKQUOTE>

Kip is obviously a bright fellow. His posts are always interesting to read and require no defense. It is evident from the information he has presented (on this thread) that he has researched the topic and has approached the question at hand from an analytical rather than an emotional approach. In this regard I respect his opinion, I simply don't agree with it.

I look at the overall numbers associated with Operation Bagration (numbers presented by Russian Sources), and I see a Red Army operational numeric superiority of 5:1 in men. A 12:1 numeric advantage in Tanks and Assault guns. And a 5:1 advantage in Guns and Mortars.

Combine this huge force ratio advantage held by the Soviets with Hitler imposed operational constraints on German Army Maneuver. Add in Red Army strategic level intelligence on AGC regarding dispositions of German troops, reserves formations, plans, etc by Red Orchestra. Now consider that the average German Infantry division on the MLR - most of which were severally depleted in strength -- was defending 1.5 to 2 times their doctrinally prescribed frontage.

And yet by Krivosheev's numbers AGC still managed to inflict an overall operational casualty ratio of over 2:1 in men, and also succeeded in destroying close to 3,000 Soviet tanks.

Do I think the Red Army of 1944 and 1945 was tactically and operational far more proficient than it was in 1940 through 1943? Yes I do. (Steve's comments on the superior quality of The Red Army of 1940 not withstanding …clearly the 1940 Red Army's performance in Finland is a model of combat effectiveness which should be emulated and revered).

[Andrew H.]

It's very silly (as Vanir pointed out) to base troop effectiveness just on Normandy and ignore, say, the Breakout or the Falaise pocket. Particularly as most of the fighting in Normandy was designed to attrit the Germans to as to make a breakout possible. That would be sort of like trying to figure out German/Soviet combat effectiveness at Stalingrad, but cutting off consideration of the encirclement part of the battle to focus only on the city fighting. (Which, from the sov. side was so fierce so as to permit the encirclement).

I do think that the Dupuy numbers are interesting, and I also believe that German defenders at Normandy were more "effective" in some vague sense, than the allies.

But to just use Dupuy's numbers, even assuming that they are accurate, takes all the fun out of wargame design. The really interesting question is *why* certain troops were more effective. To the extent that Germans were more effective than the allies, or Finns than the soviets, it is certainly not because they were born German or Finn. It has to do with things like morale, training, firepower, equipment, tactics, etc.

CM, or any game, should model these things in such a way that they are consistent with historical experience; that is the ultimate touchstone of realism in this genre, after all. But CM shouldn't take ahistorical shortcuts to achieve historical results; that's both cheating and not as much fun.

For example, if we assume that German troops in Normandy were, say, 1.5 times as effective as US troops in June, CM could make German and US squads completely identical, but give the German troops 1.5 times the firepower of a US squad. This might lead to results that completely square with the historical record. But it's a completely worthless exercise because, for all of the reasons that German troops of this period might have had a 50% effectiveness bonus over the US (and once again, I am just using this number for illustrative purposes), it was *not* because they had 150% of the firepower.

Instead, increased effectiveness was probably due to a combination of a lot of things, including more experienced troops, flexible leaders, better use of combined arms, more use of automatic weapons, etc. It's these things that CM needs to model.

[gunnergoz]

I have to wonder about these loss ratios that grandiosly proclaim that one German was equal to x.x numbers or Allied troops.

I've read Dupuy and like other staff analysts he is very fond of cranking numbers into formulae to come up with predictions of the future and justifications for the past.

Because real war (as opposed to remembered, re-written or revised history of war) is so messy and imprecise, I take all of these number-fests with an enormous grain of salt.

An almost bewildering number of variables come to mind that couldn't possibly be distilled into clean equations and predictors.

For example, the armies were very dissimilar in terms of tooth-to-tail ratios. One soldier at the front thus had a very different balance of force multipliers behind him, depending upon his side and time or war. How does one depict this precisely? How does one capture a snapshot of conditions that really equate to reality at that past moment in time?

Other factors, like morale, training, leadership and doctrine all are difficult if not impossible to quantify. We can devise our own pet looking glass and peer through it, but it remains an artifice; it is not a time machine as we would like to pretend it to be at times.

That's why I like CMBO. It only quantifies what is pretty accessible and which lends itself to numbers.

Above that level, say at the operational and strategic level, one can quantify and devise rules and formulae all one likes, but it is so abstract as to be almost meaningless. All one does is devise a gaming system, not a reality. Statistical analysis and simulation has its limits.

In the end, I guess I'm an inveterate "gut feeling" sort of grog at heart. Throw anything less concrete than armor thickness and velocity numbers at me and I begin to think in terms of my instincts about the subject. And also that's when my hard-coded BS detector comes on line!

[Tero]

>I see your steering this into the

>same "nationality modifier" debate going on

>in the other thread. I don't want to get

>caught up in that as it has been debated to

>death in the past and nothing we say here

>is going to change BTS's mind on it anyway.

Sorry. I got caught in the mood. But then again the two topics are not totally unrelated.

>Yes, it is possible to isolate and identify

>real trends particular to one side or the

>other, but these are only generalities and

>therefore should not be applied to a small

>scale game like CM.

I think Dupuys intention was to get some... I will NOT say national bias ... quantifiable, nation specific trends that could be compared even if the two forces had never actually met on the battle field.

>Example: let's say that we accept as fact

>the Dupuy calculation that on average a

>German soldier in the ETO was equal to 1.2

>American soldiers in combat effectiveness.

>Does this mean that this holds true for

>every unit in every battle? Hardly. There

>is a huge variation from unit to unit and

>at different times in different battles.

Agreed. That is why the battles of time periods have to be viewed as a whole. There may be variables at play that affect the outcome to a surprisingly large degree. Such as weather, ammo shortages, the force commander getting killed in a car crash etc. Also positive things like new equipment being introduced are factors.

>The 1.2 figure is just an average, as all

>trends are. But if you make all German

>units in CMBO equal to 1.2 American units,

>then it becomes true in every battle with

>every unit. Not realistic. If you are

>making a scenario you can simulate this my

>making the Americans "green", so the game

>does allow this to an extent without

>forcing it upon you.

Perhaps we should substitute the names of countries with "blue" and "purple" so there would be less emotinal things that cloud the debate. But if we did that we would have to throw away also many force specific details which are already present but which we take for granted.

>If CM were a strategic level game you may

>be justified in using nationality

>modifiers, but at the tactical level they

>make little sense IMO, unless the

>difference was overwhelming (10:1

>effectivenes, for example).

I disagree to some extent. These national specific differences manifest themselves in the tactical level, not in the strategic level. It is the fundamental differences in the small units tactics and doctrine that make these differences. Sure, these is something to be said about the level of experience being a major factor. But different armier learn different things, even if they go through similar circumstances. You only have to read through the basic training manuals to spot the differences, which points are virtually the same and where the differences are.

Lets take an example (this is based on hearsay but I can believe it to be true ): I claim the best aircraft of the war was the Brewster Buffalo.

Bull !!! I can hear you crying out.

The matter is resolved if you use as the criteria the number of aces per airframe built. Look it up. Most of the Finnish aces flew the Buffalo at one time or another during their career and scored most of their victories and became aces flying it. The Finnish kill-loss ratio of the Buffalo is in the order of 44-1.

The funny thing is there are few strategic factores involved in this equation. It was a miracle that our ground crews could keep them up for the duration of the war because there were absolutely no spare parts avaiable.

[Jeff Duquette]

<BLOCKQUOTE>quote:</font><HR>Andrew Said: For example, if we assume that German troops in Normandy were, say, 1.5 times as effective as US troops in June, CM could make German and US squads completely identical, but give the German troops 1.5 times the firepower of a US squad. This might lead to results that completely square with the historical record. But it's a completely worthless exercise because, for all of the reasons that German troops of this period might have had a 50% effectiveness bonus over the US (and once again, I am just using this number for illustrative purposes), it was *not* because they had 150% of the firepower.<HR></BLOCKQUOTE>

I understand what you are saying and I feel your pain What we are talking about and what you seem to be eluding to would not - in my opinion - be an appropriate method of artificially enhancing German combat effectiveness, or any other nations combat effectiveness for that matter. In that since, I totally agree with you. Moreover at the tactical level, I agree that firepower ratings should not be artificially "cranked up" to reflect historical operational level Combat Effectiveness.

In the case of Normandy German Tactical effectiveness is a function of combat experience and tactical savvy on the part of German troops, combined with tactical use of terrain. In addition the Norman countryside is well suited to defense. At the level of CM (almost lowest tactical level…i.e. squad and individual vehicles) CEV should a function purely of the mouse wielding Human General. A bad player's digital troops will typically perform badly. I would disagree with the idea that firepower of a squad should be artificially increased in order to make an incompetent player capable of achieving what he perceives are more "historical" game results. Firepower at the squad level should be purely a mechanical exercise in weapon rates of fire, number of men, etc.

"Bad" commanders are typically weeded out rapidly by natural selection. Sure some slip through the cracks and many good ones "buy the farm". But in the Case of the German Army in Normandy, a fair number of their formations had seen several years of this weeding out process on the Eastern Front. Most Allied formations going into Normandy had seen no prior action. German tactical level leaders had a strong feel for terrain and how to best weave the weapons at their disposal into the terrain advantages offered to a defending army in Normandy. This can't be reflected in a tactical level game by artificially enhancing a squad's firepower.

[Jeff Duquette]

<BLOCKQUOTE>quote:</font><HR>gunnergoz said: I have to wonder about these loss ratios that grandiosly proclaim that one German was equal to x.x numbers or Allied troops.<HR></BLOCKQUOTE>

This is not really the intent of CEV, although it tends to be brandished about in this manner. Depuy is presenting a purely antiseptic quantification of the results of a large data base of battles. There is no emotional qualification of the results. It is purely an exercise in number crunching based upon parameters, stimuli and boundary conditions that are well laid out in "Numbers Predictions and War".

Read through Jason C's post of 06-13-2001 12:36 PM. He does a good job of explaining\ranting about what CEV is or was (a shocking post considering his Techno-phobia

[kipanderson]

Hi,

May I first apologise for not having responded for a few days. It looks as though some people have asked my direct questions and I should have got back to you before. Those that did ask me questions will understand when I say that real life does sometimes get in way of me being able to spend as much time on this stuff as I would like, no doubt the same for all of us. As far as I can tell, where people have been good enough to ask me questions I will try to answer them.

When it comes to my point about the Soviet/German casuality ratio at Stalingrad and at Kursk it is quite correct that I was using figures for the “schwerpunk” or attritional centre. The figures for Kursk come straight out of the Glantz/ House book on the battle and are for the “14 days of the hot spots”. I should have used one of those small smiley faces to show that I was stirring things up somewhat. However, the point I was trying to make I stick by. Had the casuality ratios of 41/42 and the first half of 43 continued the Soviets would not have won in the Eastern Front, in my view. My point about the fall in the casuality ratio, Soviet to German, being a “necessary” but not “sufficient” condition for victory I hold to.

This brings me on to the discussion about the Krivosheev figures. In discussing his figures I will include some discussion of the Dupuy formula, as all are inter-connected. When it comes to such matters we all have to take a view on which authority is most likely to be correct. I am not in a position to go to Moscow, learn Russian, and go through the raw documents myself. (I have gone through raw documents in archives on some other subjects such as T34 sights.) The fact that “both” John Erickson and David Glantz regard his book as reliable and unbiased I take as very important. I believe they are the world’s number one guys on the Eastern Front, in the English language, others will disagree. However, just because they say something does not mean that I automatically agree. When I run the Dupuy formula and find that German combat effectiveness, with regard to the Soviets, was only 1.15 during the second half of the war, there is a way to do a check. One way to check is to check across to other types of analysis, as I described earlier. A second way is to take the same sources, i.e. Krivosheev’s casualty figures, and run the Dupuy formula for the “first” half of the war in the east. When I did that I found that, certainly in terms of “combat power” if not combat effectiveness, the answer was that one German unit of a given size was as effective as a Soviet one 5-6 times its size. Importantly this tells us that the Krivosheev figures, and Dupuy formula, are consistent with German estimates of Soviet combat power during the first half of the war. If Krivosheev’s data is accurate when dealing with the first half of the war why should it suddenly become unreliable when dealing with the second half of the war? If Krivosheev’s data had resulted in a German combat effectiveness of, say, only 2 for the first half of the war, I also would have thought “ hmm.. maybe this guy simply understates Soviet losses”, but it didn’t. The problem with German estimates of Soviet combat power during the second half of the war is that they had no way of knowing how many causalities they were inflicting on the Soviets. They did not control the battlefields after the fighting, capture millions of prisoners or vast quantities of documents.

This brings me to another point I believe Jeff raised. That is the weight that should or should not be given to the German accounts of the war in the East from the nineteen fifties and sixties. I do give a lot of weight to them and nothing that Glantz comes up with changes that. However, it has always been my view that on some points, such as Soviet causalities, the figures given in German accounts should not be considered as reliable. As I have said above, and many others on this thread have also said, the Germans had no way of knowing, particularly for the second half of the war. When you want to know what Commonwealth causalities were in NW Europe you look to Commonwealth sources not German estimates of our losses. Non of this reduces my interest or respect for books such as Panzer Battles by Von Mellenthin. But even in the early eighties when I first read the book I always assumed German estimates of Soviet causalities were likely to be wrong. They clearly were just “estimates”.

Quick point on Soviet casuality data. The Soviets did not keep personal records in the sense of Ids and names and addresses. But that is no reason to for assuming that combat commanders strength returns, casuality reports, as part of their SOPs were not reliable.

The source I used for German casualties was a table at the back of one of those nineteen fifties studies done by the Germans and US on the war in the east. Defence against Russian Breakthroughs. The version I have is one of the originals, not a Greenhill version, and I do not know if the Greenhill versions have all the tables of data. It is made clear that the causality figures are the German chain of command’s own figures. In both cases, Soviet and German, I use KIA and missing, i.e. POWs. The Russian count slightly differently, they call such figures “irrecoverable”. The Soviet figures also include wounded that were not expected to recover.

Lastly I will just take another quick look at the figures for Bagration.

If you remember Glantz describes German losses thus,

“ In the twelve days between 22June and 4 July, Army Group Centre had lost 25 divisions and well over 300,000 men. In the ensuing weeks it lost more than 100,000 additional men.”

Then consider that Krivosheev gives total Soviet losses, this time including sick and wounded, at 765,815 for a 68-day operation. If one looks at the detail of the Glantz statement on German causalities, as given above, it is reasonable to assume a figure of around 500,000 for the full 68 days Krivosheev is discussing. This would give a casuality ratio, Soviet to German, of 1.6:1. Exactly the same as for most of the second half of the war. If you focus down on killed and missing, irrecoverable, you find the Soviet figure was 178,507. I have seen a figure of 250,000 for German killed and missing but cannot remember where of the top of my head. Anyway, whichever way one looks at it the Soviet casuality ratio was clearly no more than the Dupuy formula would tell us to expect from forces of similar quality to the Germans themselves.

The fact that we will all never agree adds to the fun.

Greatly looking forward to CM2,

All the best,

Kip.

PS. this post does not take account of some of the "lastest" posts above, things are changing so fast I cannot keep up!

Jeff, thanks for the compliments,even though we do not agree. Quick take on Bagration force ratios. Glantz puts total German forces involved at 850,000, does not say over what period. Total Soviet at 1.6 million. Even if you take the higher Krivosheev figure of 2.3 million Soviets involved it still gives a force ratio of around 2.7:1 as I would have expected if the "big" pictures was taken. Given the "big" picture loss ratio of 1.6:1 as I give above, it is all unnervingly close to the global picture for the entire second half of the war.

[ 06-15-2001: Message edited by: kipanderson ]

[gunnergoz]

To Jeff Duquette -

I know what a CEV is, like I said, I've read Dupuy...

My point is that all these figures we love to bandy about are ultimately meaningless because they are based upon contemporary data that is utterly unknowable. Some factors were not quantifiable data at all but were instead the result if imponderables acting in concert with random chance and fortune.

As much as I was enamored in my youth of the neat little combat-factored units in my AH Stalingrad wargame, I knew that they were just components of a self-contained game system. I could not credibly argue that the war could have turned out one way or the other just because some particular strategy worked on my hex-based map.

To my way of thinking, leadership, vision, morale, training and timing decide more conventional battles than any other factors, asssuming the combatants make maximum effect of their given technology.

Combat effectiveness is as much a spiritual matter as it is mathmatical, I would argue.

The exception is, of course, those times when you simply bombard your enemy into the stone age. No amount of courage, training, leadership or morale will deter a nuke or a B-52 alpha strike from making a mess of your day. That is a quantity of violence applied to a situation that simply overwhelms all other factors, regardless of their quality. And like Stalin said, "quantity has a quality all it's own."

At some point in time, I would argue, we must all take a step back and concede that we're doing nothing less than arguing about "how many fairies can dance upon the head of a pin" as the Victorians would say.

As to CM, I'll happily enjoy the results of our little mathmatical machinations because I know that the armor and firepower values are based upon something vaguely realistic (for my tastes) and beyond that, I'll accept the subjective decisions applied by Steve and the boys when they gin up the troop effectiveness formulae. I'll never pretend, however, that I'm seeing anything but a game, our made-up vision, an approximation of what happened 60 years ago to some brave and very real people.