Evaluation of the Performance of the U.S. Army

Discussions on WW2 in Western Europe & the Atlantic.
User avatar
Qvist
Member
Posts: 7836
Joined: 11 Mar 2002 16:59
Location: Europe

Post by Qvist » 08 Dec 2005 13:08

But that is the problem - it does not show a lot, without understanding the 'why'.
True, hence the "also". Of course, it is neccessary to realise the limitations to what the results of such models say.
I think it is now a generally accepted statement that on the tactical level, the Germans fairly consistently managed to achieve a positive casualty ratio across the length of the war, while on the operational level this changed quite drastically during the war, and the same on the strategic level. But without asking the 'why' first, you can not answer the question of 'what'. In my view.
Well, no-one has ever to my knowledge measured CEVs on the operational or strategic levels. Nor could it fully be done, among other things because it would be impossible to integrate many of the factors that go into such models - terrain and posture, for example. Essentially however, they reflect, broadly speaking, distribution of losses relative to distribution of strength. On the basis of that, it is in my view far from given that the casualty ratio relative to the force ratio changed dramatically on the operational or strategic levels during the war - the nominal casualty exchange ratio declined from the German POW, but parallell to this, so did the force relations. As stressed previously, it is not really feasible to transpose an argument or analysis of the same type directly on higher level figures for longer periods, though it is in my view occasionally possible to draw some conclusions in cases where the data are very one-sided, to put it like that.

cheers
Last edited by Qvist on 09 Dec 2005 08:30, edited 1 time in total.

Michate
Member
Posts: 1432
Joined: 02 Feb 2004 10:50
Location: Germany

Post by Michate » 08 Dec 2005 13:35

I think this point bears repeating. Unfortunately a lot of times these models are trotted out as evidence that the Germans were simply better than their adversaries at conducting combat. The counter-argument is usually that this is not so, as is shown by the fact that they lost the war. Bingo, a lovely conflation of units of analysis, and everyone can have at it, depending on their personal preference. Not very productive, but it keeps threads going forever (which is not always a bad thing, as this one shows).
and
So, if someone comes up and says that these models 'prove' that the Germans were the more effective force, I'd point them to Michate's post, and then continue to wonder what the explanatory factors are, since most of the time the person who is using it to prove something won't be able and/or interested in giving me an answer that is either relevant or interesting.
I would say that these models "prove" (actually models can "prove" things only in their own realm, which is relevant for reality only to the extent that the factors relevant in reality are included. Nevertheless I consider models to be generally highly useful in any kind of science.) the Germans for the periods covered in these models employed tactics that made somewhat more EFFICIENT use of their army's given resources. Given that armies generally have limited resources this is of course worth to be studied on its own. However that did of course not malke them more EFFECTIVE. As effect in war is close to outcome the opposite would be more true.

BTW, it would be very telling to employ these models to study some engagements in the Korean war that the Americans lost despite apparent numerical superiority in men, not even to talk of the material situation (I am currently reading Uhle-Wettler's "Light infantry in the Atomic Age" who quotes some telling examples from the American official history.). Perhaps the real killer examples of tactical efficiency in the last century would be some of the Asian armies?

Qvist:
On the basis of that, it is in my view far from given that the casualty ratio changed dramatically on the operational or strategic levels during the war - the nominal casualty exchange ratio declined from the German POW, but parallell to this, so did the force relations.
This touches one specific definition issue of these models that I am actually not so convinced of - that, given a certain loss ratio, the weaker one side is, the better its combat effectiveness value - or in roughest terms (and ignoring such factors as terrain or posture) if one party's force is outnumbered by 1:5, looses the engagement and suffers losses 3 times higher than the adversary force, it still comes out with a higher combat effectivity than the stronger winning force.

Best regards,
Michate

User avatar
Qvist
Member
Posts: 7836
Joined: 11 Mar 2002 16:59
Location: Europe

Post by Qvist » 08 Dec 2005 13:46

Hello Michate

Code: Select all

This touches one specific definition issue of these models that I am actually not so convinced of - that, given a certain loss ratio, the weaker one side is, the better its combat effectiveness value - or in roughest terms (and ignoring such factors as terrain or posture) if one party's force is outnumbered by 1:5, looses the engagement and suffers losses 3 times higher than the adversary force, it still comes out with a higher combat effectivity than the stronger winning force. 
Mm, if one applies that logic directly on raw figures it does yield some very strange results. But if I have understood correctly, the relation between strength and losses is not expressed in such a straightforward manner by the CEV.

cheers

Andreas
Member
Posts: 6938
Joined: 10 Nov 2002 14:12
Location: Europe

Post by Andreas » 08 Dec 2005 13:48

I would include the mass surrenders at Stalingrad and Tunis into the operational equation, the same with the encirclement and destruction of 4. Armee at Minsk, or most of 6. Armee in Romania, and also Falaise. Pretty much any German surrender before 8/9 May 45 (and allied surrenders depending on their relevant dates) I would attribute to an operation. Given that, I would be very surprised if even after adjusting for changed force ratios, this period would not look an awful lot worse for the Germans then Fall Gelb, the operations of AG Centre and AG South during Barbarossa, or Kharkov 1942. But maybe I am wrong in my assumption.

But in any case, what also appears to be important to me is that there is a difference between efficiency and effectiveness that is often overlooked when throwing around casualty exchange ratios (present company excluded). The Germans in early war were more efficient than their opponents and effective (in the sense of losses as proportion of force when achieving the desired outcome). By late war they may still have been more efficient, but they certainly were not particularly effective anymore, since (I assume) they lost a higher proportion of their forces, while not achieving their goals anymore. Also something to keep in mind.

All the best

Andreas

Michate
Member
Posts: 1432
Joined: 02 Feb 2004 10:50
Location: Germany

Post by Michate » 08 Dec 2005 13:57

It is an interesting question, but I am fundamentally skeptical towards the possibility of such models. The main reason is that it would inevitably mix together widely disparate issues, whose combination into a single model would tend to obscure rather than illuminate. Also, it is not easy to see even theoretically how they could possibly be related to each other in a menaningful way within a quantified model. Several aspects are also governed by fundamentally different logics, some of whom resist any quantification. I would much rather have a series of different types of analysis dealing with different aspects. In the main, I think it is an advantage rather than a drawback with quantified tactical models that what they say something about is really quite limited.
It would be certainly limited to some very rough relation of "strategic input - strategic output" with some high-level "theater parameters" added, while treating all other factors as a "black box". And these other factors are of course of such a broad and diverse nature that combinig them within the single resulting strategic efficiency value will not tell you anything meaningful.

Thus I am completely happy applying models to more limited questions and allow them to be more specific in return.

Best regards,
Michate

Andreas
Member
Posts: 6938
Joined: 10 Nov 2002 14:12
Location: Europe

Post by Andreas » 08 Dec 2005 13:58

Michate wrote:I would say that these models "prove" (actually models can "prove" things only in their own realm, which is relevant for reality only to the extent that the factors relevant in reality are included. Nevertheless I consider models to be generally highly useful in any kind of science.) the Germans for the periods covered in these models employed tactics that made somewhat more EFFICIENT use of their army's given resources. Given that armies generally have limited resources this is of course worth to be studied on its own. However that did of course not malke them more EFFECTIVE. As effect in war is close to outcome the opposite would be more true.
Hi, we cross-posted, proving that great minds think alike. :)

In any case, I would not immediately agree that these models prove what you say. They could also prove that the Allies were just a bunch of dimwits who never got their act together, so whether or not the Wehrmacht made efficient use of its resources was neither here nor there.

I do fully agree that models serve a function, however I think one has to be very careful about the assumptions that are underlying them - these may or may not be closely related to reality, depends on the model really. In this case, I would go so far as saying that given their specific assumptions, these models show that the Germans achieved positive casualty exchange ratios. Period. From this we can hypothese, as you or I have done (or we could just rubbish the models, also a favourite past-time in science). Some hypotheses. It could be that this reflects more efficient use of resources. It could be that this reflects continued underachievement of their opponents. It could be that it reflects their generally more defensive stance when capabilities were even, and their capability advantage when they were offensive (ooohhhh! dynamics in the assumptions!). It could be that all Germans were Übersoldiersupermen (oh wait, that was the Finns). Lots of things one can research from the result of these models really. But they 'prove' none of them outright, in my view.

All the best

Andreas

Michate
Member
Posts: 1432
Joined: 02 Feb 2004 10:50
Location: Germany

Post by Michate » 08 Dec 2005 14:06

Mm, if one applies that logic directly on raw figures it does yield some very strange results. But if I have understood correctly, the relation between strength and losses is not expressed in such a straightforward manner by the CEV.
Unfortunately my knowledge on these models is much too limited to dig into such questions, this was just an impression I had when superficially looking at some of the cases, and I may be wrong here.

User avatar
Qvist
Member
Posts: 7836
Joined: 11 Mar 2002 16:59
Location: Europe

Post by Qvist » 08 Dec 2005 14:22

Hello Andreas
I would include the mass surrenders at Stalingrad and Tunis into the operational equation, the same with the encirclement and destruction of 4. Armee at Minsk, or most of 6. Armee in Romania, and also Falaise. Pretty much any German surrender before 8/9 May 45 (and allied surrenders depending on their relevant dates) I would attribute to an operation. Given that, I would be very surprised if even after adjusting for changed force ratios, this period would not look an awful lot worse for the Germans then Fall Gelb, the operations of AG Centre and AG South during Barbarossa, or Kharkov 1942. But maybe I am wrong in my assumption.
I am not at all sure, given that the force relation changes more strongly than the casualty ratio. But here we are all groping in the dark, as there is AFAIK no methodology in existence that could guide such an assessment. If one just makes a "raw" comparison, it seems to raise far more questions than it answers: For what it's worth (which almost certainly ain't much), this a purely nominal overview of the combat losses in the East, where the quarterly force ratio have simply been multiplied with the quarterly casualty ratio:


Force.........Caslt.........comb........period
1:1.2 1:4.8 5,76 3q41
1:1.1 1:5.4 5,94 4q41
1:1.7 1:5.9 10,3 1q42
1:2.0 1:6.2 12,4 2q42
1:2.0 1:6.0 12 3q42
1:2.2 1:7.0 15,4 4q42
1:2.1 1:3.8 7,98 1q43
1:2.3 1:3.9 8,97 2q43
1:2.4 1:4.9 11,76 3q43
1:2.5 1:5.0 12,5 4q43
1:2.6 1:4.4 11,44 1q44
1:2.6 1:3.1 8,06 2q44
1:3.1 1:2.0 6,2 3q44
1:3.6 1:3.6 12,96 4q44

Primarily a pretty good illustration of your previous point concerning the need to understand why these figures are as they are, and quite possibly an altogether pointless exercise.

One thing that is very clear from the above (from the casualty ratios alone) is that major annihilation battles often has a major effect on the macro casualty exchange ratios, which is one of the reasons why it is difficult to apply the same type of analysis to higher levels. Interestingly, this seems to be the case primarily for the Soviets: there are marked improvements in the period where the Stalingrad losses were written off, and even more during the summer of 1943. There is no corresponding effect in periods of major German annihilation victories. Quite possibly this in itself could tell us something interesting about the two armies. I wonder just what though. :).
But in any case, what also appears to be important to me is that there is a difference between efficiency and effectiveness that is often overlooked when throwing around casualty exchange ratios (present company excluded). The Germans in early war were more efficient than their opponents and effective (in the sense of losses as proportion of force when achieving the desired outcome). By late war they may still have been more efficient, but they certainly were not particularly effective anymore, since (I assume) they lost a higher proportion of their forces, while not achieving their goals anymore. Also something to keep in mind.
Of course - "effectiveness" and "efficiency" can be understood as two different things. I think that usually the distinction isn't observed, and most people simply mean the latter.

cheers
Last edited by Qvist on 09 Dec 2005 08:36, edited 1 time in total.

User avatar
Qvist
Member
Posts: 7836
Joined: 11 Mar 2002 16:59
Location: Europe

Post by Qvist » 08 Dec 2005 14:27

Hello Andreas
In any case, I would not immediately agree that these models prove what you say. They could also prove that the Allies were just a bunch of dimwits who never got their act together, so whether or not the Wehrmacht made efficient use of its resources was neither here nor there.
and
It could be that this reflects continued underachievement of their opponents.
There is no distinction between these in this context. Efficiency as understood in such models is relative - the efficiency of one side is per definition the inverse of the other. If one side has a CEV of 1.20, then that of the other is .80. It's a zero sum game, so the stupidity of one side is the brilliance of the other. :)
It could be that it reflects their generally more defensive stance when capabilities were even, and their capability advantage when they were offensive (ooohhhh! dynamics in the assumptions!).
If so, that should emerge from the cases themselves, as some of them have the allies defending while others have the allies attacking. Also, the effects of posture are a part of the model.
In this case, I would go so far as saying that given their specific assumptions, these models show that the Germans achieved positive casualty exchange ratios.
Actually, they don't neccessarily show that even if they show a higher German CEV. What they strictly speaking show (as far as the Ardennes and Italian databases are concerned) is that on average, the Germans tended to achieve a larger number of casualties inflicted relative to the military force (as modified by relevant factors) they deployed in an engagement. In fact, the Germans suffered more casualties in the Ardennes than the allies did.

cheers
Last edited by Qvist on 09 Dec 2005 08:39, edited 1 time in total.

Andreas
Member
Posts: 6938
Joined: 10 Nov 2002 14:12
Location: Europe

Post by Andreas » 08 Dec 2005 14:41

Qvist wrote:Hello Andreas
In any case, I would not immediately agree that these models prove what you say. They could also prove that the Allies were just a bunch of dimwits who never got their act together, so whether or not the Wehrmacht made efficient use of its resources was neither here nor there.
and
It could be that this reflects continued underachievement of their opponents.
There is no distinction between these in this context. Efficiency as understood in such models are relative - the efficiency of one side is per definition the inverse of the other. If one side has a CEV of 1.20, then that of the other is .80. It's a zero sum game, so the stupidity of one side is the brilliance of the other. :)
Thanks for that, I was not aware that this is how they worked. Nevertheless, I understood Michate's point to be relating to something outside the specific model. But in any case, here is a hypotheses that should work - it could be that this reflects the superiority of the HMG42. Better? ;)
Qvist wrote:
In this case, I would go so far as saying that given their specific assumptions, these models show that the Germans achieved positive casualty exchange ratios.
Actually, they don't neccessarily show that even if they show a higher German CEV. What they strictly speaking show (as far as the Ardennes and Italian databases are concerned) is that on average, the Germans tended to achieve a larger number of casualties inflicted relative to the military force (as modified by relevant factors) they deployed in an engagement. In fact, the Germans suffered more casualties in the Ardennes than the allies did.

cheers
In this case, I would go so far as saying that given their specific assumptions, these models show that the Germans achieved positive casualty exchange ratios.
;)

Thanks for the table of losses etc. I agree with you that it raises more questions than answers.

All the best

Andreas

User avatar
Qvist
Member
Posts: 7836
Joined: 11 Mar 2002 16:59
Location: Europe

Post by Qvist » 08 Dec 2005 14:53

Thanks for that, I was not aware that this is how they worked. Nevertheless, I understood Michate's point to be relating to something outside the specific model. But in any case, here is a hypotheses that should work - it could be that this reflects the superiority of the HMG42. Better?
Better yes, but perhaps still not quite there. :) Here we come up against the limits of my familiarity with the detailed workings of these models, but - one of the factors modelled is usually firepower, and it is quite possible to take account of the effects of different weapons systems rather than just count them.
In this case, I would go so far as saying that given their specific assumptions, these models show that the Germans achieved positive casualty exchange ratios.
This is really, really nitpicking, but a CEV is fundamentally not the same thing as a casualty ratio, regardless of specific assumptions. Sorry.

cheers

Andreas
Member
Posts: 6938
Joined: 10 Nov 2002 14:12
Location: Europe

Post by Andreas » 08 Dec 2005 15:04

Qvist wrote:
In this case, I would go so far as saying that given their specific assumptions, these models show that the Germans achieved positive casualty exchange ratios.
This is really, really nitpicking, but a CEV is fundamentally not the same thing as a casualty ratio, regardless of specific assumptions. Sorry.

cheers
It takes me a while, but I'll get there in the end (maybe):
In this case, I would go so far as saying that given their specific assumptions, these models show that the Germans achieved a higher CEV than their opponents.

In any case, the fundamental point is that we have these models, and this should be close to what they give us. Ways for dealing with this are plentiful:

a) Question the applicability of the model (If I understand him correctly, this is what Jon is doing)
b) Question the quality of the model
c) Question the quality of the data fed into the model (I think Igorn is often doing that)
d) Accept the model and results, and wonder what is going on to get to these results (that's what I like to do)
e) Use the results of the model without hesitation or questioning to proclaim that the Wehrmacht rocked (Seen that a few times as well)

The list is not exhaustive, and a-e can be freely mixed. A matter of preferences.

All the best

Andreas

Michate
Member
Posts: 1432
Joined: 02 Feb 2004 10:50
Location: Germany

Post by Michate » 08 Dec 2005 15:07

Better yes, but perhaps still not quite there. Smile Here we come up against the limits of my familiarity with the detailed workings of these models, but - one of the factors modelled is usually firepower, and it is quite possible to take account of the effects of different weapons systems rather than just count them.
It depends on whether the models lump together e.g. any type of light MG or whether technical properties such as rate of fire, effective range or weight are paramterised within the model. Then there is additionally the question of intangible factors (or at least factors where it is difficult to find enough hard data, though in principle they can be measured) such as reliability, ergonomity and so on.

Another question to these models, falling under point b) - do they explicitly factor in ammunition consumption and availability?

User avatar
Qvist
Member
Posts: 7836
Joined: 11 Mar 2002 16:59
Location: Europe

Post by Qvist » 08 Dec 2005 15:23

At this point, we might perhaps profitably have a look at the link Jon previously provided. It deals with capture rates, but it builds on several databases and contains also considerable information of more general interest. Interestingly, it also does in fact contain operational level modelling - news to me. Refer pages 29-36, and 52-58.

http://www.dupuyinstitute.org/pdf/e-4epw1and2final.pdf

cheers

User avatar
Qvist
Member
Posts: 7836
Joined: 11 Mar 2002 16:59
Location: Europe

Post by Qvist » 08 Dec 2005 15:28

This post by Christopher Lawrence quotd from the TDI Forum puts the above linked-to study in context, and also comments on scoring of weapons systems (my emphasis):
quote:
--------------------------------------------------------------------------------
Originally posted by yadernye:
It is this: regarding the German and Allied casualty rates for the Italian Campaign data listed by Col. Dupuy in NPW, the Germans suffered slightly higher total casualties and a higher rate of casualties than the Allies. Yet, Col. Dupuy asserts that the Germans had a higher score effectiveness than the Allies, regardless of posture and whether or not they lost the engagement. This does not seem consistent with the casualty data for the database.
--------------------------------------------------------------------------------

If one is consistently outnumbered and outgunned, then one would expect to take more casualties and loose more battles. The model scores all the weapons (a very rough approximation of combat power) and these scores are then compared. Furthermore, the scores are modified by the conditions of the combat (terrain, fortifications, weather, etc). What generated the conclusion that the Germans had generally higher combat effectiveness is that when all these factors are taken into account, they tended to performed better than the predicted outcome of the model, whether on attack or defense (meaning the model had no implicit bias in one mode or the other). As the model is complex, and the CEV is based upon a comparison to the predicted result of the model, then some people just called it "voodoo" and dismissed it outright.

There are other ways to get there than the QJM. For example, in our EPW Report Phase I & II, we do the same test using simple aggregate averages and statistical comparisons. In particular, we compare the performance of different nationalities in low odds attacks on both the attack and defense. There is no scoring and no QJM/TNDM used. This still produces very similar answers to what the QJM produces, using a completely different methdology.

Furthermore, the British OR community has also looked into this subject, primarily David Rowland, and also has come up with similar results.

The fact that there are three different methodological tests that have produced similar results seems to be simply ignored.
cheers

Return to “WW2 in Western Europe & the Atlantic”