Jump to content

Film of Stuka 87G in action


Recommended Posts

  • 2 years later...
  • Replies 327
  • Created
  • Last Reply

Top Posters In This Topic

This thread comes from the past as an additonal evidence for the issues i have with Jason in another thread.

I point page 8 of this thread and whoever has some basic knowledge of OR is welcomed to enjoy.

I have to correct Professor Jason, an expert in operations research who messes with a simple formula. Notice that in none of his subsequent post made any effort to counterargue anything.

Note also that this is not some issue of brainfart that everybody can have or some just wrong addition.

The guy fails to identify the right equation which is one of the most basics.

Jason originally posted

But let's apply the missing principle of charity in argument the other side mislaid, and assume as naively as possible that low kills per Russian tank were simply a byproduct of their overwhelming numerical superiority in Lancesterian fashion. Its just math, let's do it.

20,000 tanks to start plus 8,000 lend lease plus 102,000 produced equals 130,000 Russian tanks. Germans around 50,000. Let's take 10000 tanks are our unit and call it 13 to 5. The squares are to remain equal under Lancestrian attrition. Russian power 169, German power 25, net outcome Russians left with 144, Germans with zero. Square root again to get back to numbers, we have 12. In our units, that is 120,000 tanks, that Russia was supposed to end the war with. The actual figure is 30,000 or less, their losses were 100,000 not 10,000. So our fine Lancestrian estimate gets Russian losses wrong by a factor of ten.

My response

The crucial point is the following

The above was the main issue I had to address in detail first.

Let’s return to your initial equation.

Now it will be much clear the problem of your study.

As I said earlier you used a specific form of the general equation.

If you compare the equation you used with the general one i provided, you will realize that you inserted an assumsion which destroys completely the value of the study .

Without realizing it , you insert the assumsion that A/B=1

That means that indirectly without knowing it, you imply that the "lethality coefficient" of germans and Soviets is the same.

So you treat all factors that influence "lethality coefficient" like training, leadership, equipment quality and so on , as equal for both sides. So you say for example later

their losses (Soviets) were 100,000 not 10,000. So our fine Lancestrian estimate gets Russian losses wrong by a factor of ten.

Although as i said from the beginning noone claims that lanchester estimations of numbers are accurate, your wrong approach makes it easy for someone to contradict you.

Of course there is going to be a totally different estimation from historical results , since

you imply that Soviets had the same quality of germans ( A=B) ,you are going to find totally different results.

In fact someone can claim that if Soviets had the same historical quantity superiority combined with the same quality as Germans ,then it would not be surprising to see their losses much more lower than what they had historically.

A 10.000 Soviet losses in this scenario is certainly a possibility.

Here is the whole explanation cause obviously i am speaking with an amateur and i have to explain many things.

As a side point, it is interesting that although the professor is so confused with the concept of average and the lethality of tank battles ,claiming that it is actually low, I recently saw by chance in another thread that today seems to have changed his opinion. I beleive our conversation helped him eventually ,although he would never admit this in public

First point.

I thought it was clear since it was mentioned earlier that Lanchester equations are used to value principles and examine simple scenarios.

They do not predict numbers in complex situations,although there are some projects that use advanced and extensive modifications of Lanchester equations trying to predict results (without big success).

Therefore the whole post is totally irrelevant.

The issue is that we know and accept some rules like that the ratio of forces - affect outcomes-kills favoring the big site (with all else equal) ,which is something that numerators and denominators do not consider at all,(better to say they distort it ,but i will talk about that later)

The problem is that although we are aware of the limitations of Lanchester theories, you do not seem to be aware of the limitations of your theory and you are very bold in using it to describe battlefield effects.

So, even if you could use the proper method to evaluate Lanchester applicability in a historical example and prove that results do not match data, you would not point anything new .

So now we come to point two.

That is the way you try to prove that Lanchester equations do not give accurate numerical results.

It is totally wrong from start to the end.

First mistake is that you use the overall production numbers.

Lanchester equations apply when neither side receives additional reinforcements or replacements during the resolution of the fight.

Since neither germany or Soviet Union had the total number of tanks you use from start ,but they actually "build it" during the fight , there is no logic in using these numbers in Lanchester equations.

Second,

let's ignore the above mistake and assume that it is not so important.

You continue by using the wrong type of equation!!

Search your papers and see the GENERAL equation used to find the number of units that will survive after the end of the fight using the theory of square law.

This general equation in our example is the following

If we name

Y1 the number of soviet tanks surviving the engagement

yo the initial number of Soviet tanks at the start of the engagement (accepting in spite the previous point your number - 130000)

x0 the initial number of German tanks (accepting again your number of 50.000)

then

y1^2 = y0^2 -(A/B)x0^2

where y1^2 means y1 squared (same with y0^2 , x0^2)

A and B are factors i will explain later.

Now your equation is a specific form of the above.

In your equation A=B.

Now what is this A and B?

According to the square law theory A and B are the LETHALITY COEFFICIENT of Germans and Soviets.

Now i have to explain to you the meaning of this coefficient cause by reading your post i noticed that the main problem is that you confuse different concepts.

I will give my best shot to explain it clearly cause this is the main sourse of your misunderstanding.

The equation we use in this case is a bi- product of the following equations.

Dx/dt= -By and

Dy/dt=-A.x

This is where we find for the first time A and B.

Let's focus on the first equation.

The first equation points that

the rate of attrittion of germans at any time during the engagement (Dx/dt) is proposional to the number of participating Soviets at that time (y) and the factor B.

Notice that since we have attrition (x becomes lower as the time passes) and since Dx is defined as x at a subsequent time - x at a previous time therefore Dx= negative.

That is why you see the (minus) at the other portion of the equation.

returning to my effort to explain the meaning of the term "lethality coefficient",i will say the following.

The first equation points that at ANY TIME during the engagement , the Russian "lethality coefficient" B is always the same.

This factor is not affected by the quantity x or y or ratios of forces .It is independant of all that.

So what does this mean in real world experience?

We can define it in different ways.

One might use the phrase that it is the kill probability of the weapons participating in an engagement if ALL ELSE factors that affect perfomance (like morale,leadership ,terrain and so on) are considered the same for bothe sides.

Another one might put it differently.

Try to imagine yourself inside a single tank ,trying to engage a single enemy tank with your gun.

Before you pull the trigger to send the shell away ,you realize that you have a certain percentage of probability to kill the target.

Your equipment, skill,training, relative position and so on and so on combined all together in this specific situation,give you a certain percentage to kill the target.

This is what "lethality coefficient" stands for.

It is the percentage of killing generated by all the above factors.It expresses your ability to kill and it is a "quality" indicator.

Notice nomatter how we describe it , it is true that this percentage has no relation with x or y or force ratios.

So consider that you are part of a company of 16 tanks ready to fire simultaneously a volley of 16 shells towards the same single target.

Again all other factors of training,skill,equipment and so on is the same as in the previous case.

In this situation and inspite the fact that the ratio of forces has changed, you should be able to realize just before you fire the shell as part of the team , that you have the SAME percentage of probability to kill the target.

The presence of your fellow tankers do not affect your training,relative position, quality of gun and so on that shape your "lethal coefficient"-meaning ability to destroy the target.

Now here is where there is the main confusion.

Using statistics and numerators denominators.

You claim that when you are alone and ready to engage a single enemy tanks, you realize that your "average effectiveness" after you destroy it, will be one tank. (one shooter -one loss)

On the other hand, when you are part of a company ready to engage the same single tank during the same conditions you are aware that your "average effectiveness" after the german tank is destroyed will be 1/16 of the tank ,(16 shooters causing one loss).

This is what you do not realize from the beginning.

This type of "average effectiveness" you consider ,which is a bi product of your numerator and denominator, has nothing to do with the "lethality coefficient" which is the real factor indicating your ability to engage and kill a target.

This "lethality coefficient" is independent from force ratios as i said.

The “average effectiveness” is certainly affected by force ratios and might have a very different value .

So someone might claim in the previous second scenario that the “lethal coefficient “ of a tank at that specific engagement is 20% and you argue that this can not be true cause according to your “average effectiveness” you expect to see a percentage of 1/16 or about 6% .

The above was the main issue I had to address in detail first.

Let’s return to your initial equation.

Now it will be much clear the problem of your study.

As I said earlier you used a specific form of the general equation.

If you compare the equation you used with the general one i provided, you will realize that you inserted an assumsion which destroys completely the value of the study .

Without realizing it , you insert the assumsion that A/B=1

That means that indirectly without knowing it, you imply that the "lethality coefficient" of germans and Soviets is the same.

So you treat all factors that influence "lethality coefficient" like training, leadership, equipment quality and so on , as equal for both sides.

So you say for example later

their losses (Soviets) were 100,000 not 10,000. So our fine Lancestrian estimate gets Russian losses wrong by a factor of ten.

Although as i said from the beginning noone claims that lanchester estimations of numbers are accurate, your wrong approach makes it easy for someone to contradict you.

Of course there is going to be a totally different estimation from historical results , since

you imply that Soviets had the same quality of germans ( A=B) ,you are going to find totally different results.

In fact someone can claim that if Soviets had the same historical quantity superiority combined with the same quality as Germans ,then it would not be surprising to see their losses much more lower than what they had historically.

A 10.000 Soviet losses in this scenario is certainly a possibility.

Later

Let's still pursue it anyway, to see what it predicts about average achievements, relative performance of the winning and losing sides, etc. Total kills are 10,000 + 50,000 = 60,000, total shooters are 130,000 + 50,000 = 180,000. Average kills per shooter are 0.33. Necessarily less than unity by the already mentioned identity. Here, vastly less, because the Lancester equations predict run-away wins ("snowballing") for the more numerous side.

Russian kills are 50,000, spread over 130,000 shooters, average kills 0.385, above the 0.33 average. Despite their larger numbers, Lancestrian odds effects raise their average effectiveness, compared to the overall average of both sides. The Germans on the other hand have 10,000 kills spread over 50,000 shooters, average kills 0.2, far below the average for both sides and barely better than half the Russian figure. The relatively performance and the initial odds multiply out to the predicted 5 to 1 loss ratio, favor the more numerous side.

Now we know the war was not such a lopsided win, and the kill ratio was 2 to 1 favor the Germans, not 5 to 1 favor the Russians.

A clear example of how much you are confused by inserting "average effectiveness" in the Lanchester theory.

.

You have already inserted the equation that the German "lethality" coefficient" is the same as the Soviet one.

We do not know the actual value but we DO know that whatever is this value ,is going to be the same for the whole duration of the scenario regardless of numbers or force ratios and is equal for both sides .

In other words, if you are in a single tank ready to exchange your first volley against an enemy tank,regardless if the target is alone or part of a bloody division , you will still have decent chances to take out of action ,even if hundreds of tanks of a bloody division in the second case are going to "evapotare" you during this exchange of volley.

When you try to use the "average effectiveness" as you define it through numerators and denominators, you miss something important.

If 1000 tanks engage a single tank in a single volley, they will inflict MULTIPLE KILLS according to lanchester theory.

The single tank might receive 20 30 40 different rounds.

In Lanchester theory these rounds are considered as KILLS cause the attrition as described in the theory is calculated through the use of "lethality coeficient" probability.

In Lanchester law we assume a SIMULTANEOUS EXCHANGE OF VOLLEYS until one side is completely eliminated.

The answers of a question regarding the statistical results of x shooters with A "lethality effectiveness" " acording to lanchester requires a totally different procedure

than a simple division of numerators and denominators..

We do not care so much about the shooters, we basically care about the shells and if we do indeed take force ratio measurements ,is because our final objective is to determine number of volleys and effective shells.

By using your "average effectiveness" you fail to address the shells.

The "average effectiveness" sees just a single casualty and divides it to all members which participated in the engagement.

How was this casualty produced is an issue it does not address at all.

Lanchest does point the fact that big ratios affect casualties in favor of the big guy, but this is not a result of less "lethality coefficient".It is a result of shorter duration of the battle until the small guy is destroyed and this is what affects the actual outcome of losses.

That is not so strange if you think of terms of "lethality coeficcient" and get over with "average effectiveness".

The more enemy shooter trying to kill you ,each one having a certain Lethality probability, the less chances for you to survive long enough to try your lack using your "lethality' ability in subsequent volleys.

That is the whole story.

That applies even in the extreme case of 1000 tanks firing at you.

At first you might think that no matter if they are 1000 or 1000000, they are going to get you anyway during the first volley.

So you would not expect to see different large numbers giving different chances to you to score a hit during an engagement since the duration will be basically the same in any case-meaning one exchange of volleys before you are destroyed.

But since the theory "sees" multiple shells-kills against you and since the more the numbers ,the more these multiple shells destroying you, it should not be surprised that even in such extreme cases, the "losses" you inflict in theory are getting lower.

Maybe the best example i can use is that the theory "in the absense" of other targets besides you engaging 100 or 1000 tanks, "treat" you as a single vehicle with 100 "life points".

The same with anyone of your targets.

So 1000 tanks destroy your "points" while you might destroy " a single point" from an enemy vehicle.

That is because the theory "needs" to focus on the effectiveness of SHOTS ,not shooters and the definition of attrition in this theory (Lethality probability multiplied by number of shooters) ,is such as that it does not "accept" that a round landing together with many others in your vehicle is "worthless".

In order to be able to "reward" these rounds that "passed" the "lethality coefficient probability check" as effective in killing you, it needs to see your single target as a sum of "life points".

For the "average efectiveness" all these things about shells are irrelevant.

It just sees a single loss and 1000 shooters and since you confuse this "average effectiveness" with the "lethality probability" when you shoot trying to kill a target, you come to totally different conclusions.

you fail to distinguish the two following cases

The first one is the case when the "lethality coeficient" is very small ,say 1% and therefore as a result 100 tanks will have poor perfomance inflcting maybe one kill if they all fire together.

And the second is the case where 100 tanks with a much better "lethality coeficient" say 20%

will land successfully 20 rounds on a single target, but since you focus only on denominators and numeratos, you treat this as 1 kill per 100 tanks again like in the previous case.

And when someonr points you that it IS possible to have good "lethality coeficient" inspite what numeratots and denominators point, you disagree

My post is too long so i will stop here.

I think all your points are relevant with what i described until now.

The funny thing is that i do agree with you that Lanchester theory has limitations and can not predict accurate results.

Still i chose to address the way you think cause

i noticed at some point that the whole thing leads you to very strange results.

So although i could accept your claims about gross inflations of pilots and i could agree with at least some things regarding CAS, i started to have big issues with other conclusions that do not match battlefield experience as i gather it though AAR, combat diaries,stuff studies and so on , about the great difficulty to score kills during a typical armor engagement.

P.s I am sorry if at some points i am very direct saying clearly that you are wrong.

This is nothing personal.

It is obvious that i think you are wrong and you think the same about me.

nevertheless i consider this as a good discussion even though it seems that it is just the two of us.

[ April 27, 2008, 03:29 AM: Message edited by: pamak1970 ]

Link to comment
Share on other sites

  • 8 years later...

JasonC,

Was trying to find a post I did regarding a Russian war film about an 85 mm AAG roled as an ATG in Barbarossa when I came across this great thread from back when. In part of it, you talk about Napoleonic battle practices and also supply some shocking information about troops firing from as far as a 1000 yards out. Please supply the source/s for anything over typical engagement ranges for musketry. Have never encountered any reference (am hardly an expert, though) to 1000 yard engagement ranges. Field artillery? Even 6=pounders didn't do such things that I know of on the battlefield, let alone infantry armed with the Brown Bess, Charleville and the like. Did someone do something crazy like that once or twice? Perhaps. But I would take a report of muskets being fired at long range being maybe twice normal range, if that. Strongly disagree with your contention Napoleonic infantry combat was mostly skirmishers. The evidence shows anything but that as having been the case. For line as being the standard combat mode, per the period regs, see the beginning of this great analysis of Napoleonic infantry tactics. The same article also goes into great detail on the composition, quantities and use of skirmishers on the various sides. My remarks are in Italics. Have Bolded some key points.

http://www.napolun.com/mirror/napoleonistyka.atspace.com/infantry_tactics_4.htm#infantrycombatintroductionskirmishers

 

Marshal Davout's instructions from 1811 order one company from a battalion deployed as skirmishers,

(Fair Use)

  • According to Marshal Davout's instructions issued in 1811, when a company was send forward to act as tirailleurs it first marched 200 paces away from the battalion. Here the center section halted while the left and right section of the company marched forward a further 100 paces.

  • - The centre section was under captain and acted as a principal reserve for the skirmish chain.
    It consisted of 1 captain, 1 sergeant major, 1 sergeant, 2 corporals and 40 privates (total: 45 men).
    The communications were maintained by using NCOs as runners.
  • - The left section was under lieutenant and deployed its 1st and 2nd rank (total: 26 men)
    as skirmishers, with the 3rd rank being held in reserve. With the reserve were 1 lieutenant,
    1 drummer or trumpeter, 1 sergeant, 3 corporals and 13 privates (total: 19 men).
    The reserve furnished replacements and also formed rallying point when attacked by cavalry.
  • - The right section was under lieutenant and also deployed its 1st and 2nd rank (total: 26 men)
    as skirmishers, with the 3rd rank being held in reserve. With the reserve were 1 lieutenant,
    1 drummer or trumpeter, 1 sergeant, 3 corporals and 13 privates (total: 19 men).
    The reserve furnished replacements and formed rallying point when attacked by cavalry.

Typically nobody else had skirmishers in a battle on the same scale as the French, who used skirmisher formations as big as a battalion to accomplish important tasks. As for skirmisher effectiveness , from the same source, I have provided a telling list of examples of how nasty French skirmishers were in documented instances. Examples from other nations are presented, too, at the link. The first example is particularly telling and shows unmistakably the line was the classical formation on the target end.

 

(Fair Use)

  • Other examples:

  • - In 1806 at Jena, French skirmishers took on enemy's line infantry deployed in open field. Maude described this action: "Now followed one of the most extraordinary and pitiful incidents in military history. This line of magnificent infantry, some 20,000 strong, stood out in the open for 2 hours whilst exposed to the merciless case and skirmishing fire of the French, who behind garden walls offered no mark at all for their return fire. In places the fronts of the companies were only marked by individual files still loading and firing, whilst their comrades lay dead and dying around them." (Maude - "The Jena Campaign, 1806" p 156)

Absent other information, such as how many and of what weight, it is hard to know what the killing potential was of the cannon firing case ammunition. But we do know that the ratio of cannon to infantry in the period was quite low. But let's move to a case with pure skirmishers, if the account is accurate.

  • - Petre described on p 137 in "Napoleon's Conquest of Prussia" another action of French tirailleurs at Jena: "Harassed by a galling fire from the swarms of skirmishers in and on either side of Vierzehnheilegen and unable to return it, the Prussian infantry was already shaken and demoralised before it was ready to begin volley firing. Even when it did begin, the fire had little effect on the French skirmishers, adepts as they were in finding cover in the gardens and potato fields or behind the walls of the village." and on p 178: "The Prussian attacks on Vierzehnheilegen were a good example of the impossibility of succeeding with the parade-ground tactics of the Seven Years War. The unhappy Prussians, attempting solemnly to form line before opening fire, were decimated by the fire of the French, ensconced behind the walls of the village and the gardens, or hidden in the furrows of the potato fields."

Again, it is clear that line is the typical formation for infantry combat.

  • - Englishman Sir F. Ponsonby was an involuntary witness to French skirmishing at Waterloo. He found himself wounded and immobilized in a sector of battlefield occupied by French skirmishers. One of these threatened to kill him and demanded his money; Ponsonby let himself be searched .... A second skirmisher with the same intentions arrived but left disappointed after an even more meticulous search of colonel's person. Stll later, another skirmisher came by and decided to use the immobile Ponsonby as a screen ! He stayed for a long time, reloading and firing over Ponsonby's body again and again, and conversing with great gaiety all the while.
  • - At Waterloo after one of the attacks on Hougoumont, many French skirmishers instead of withdrawing to their own lines, climbed the slope in the direction of the British and German positions, concealing themselves amid the high-standing grain. British battery of 6 9pdrs under Lt.-Col. Webber-Smith suffered casualties when the French tirailleurs opened fire. In the course of a few minutes, many gunners and horses were hit, and Webber-Smith had to give orders to limber up the guns and very quickly abandon the position.
  • - At Waterloo the enormous number of French skirmishers in action, together with the intensity of an artillery fire, had caused an unusual pessimism to spread among the British officers. Col. Gould confessed that he found this situation desperate. Not the least of his reasons was his belief that the only road through the forest would be bottled up in an instant. Captain Mercer: "It does indeed look very bad... Meantime gloomy reflections arose in my mind, for though I did not choose to betray myself (as we spoke before the men), yet I could not help thinking that our affairs were rather desperate, and that some catastrophe was at hand. In this case I made up my mind to spike my guns and retreat over the fields ..."
  • - In April 1813 at Halle, General Maison detached a battalion of French 153rd Line to the Giermeritz farm, on an island formed by two branches of the Saal River. The battalion's voltigeurs took up a position where their fire began picking off the Prussian gunners serving the guns in the earthwork on the far side of the river. The 2 6pdr cannons and 1 howitzer were hurriedly withdrawn due the voltigeurs' fire.
  • - In 1813 at Dennewitz, French tirailleurs came out of the village and drove back Prussian Horse Battery Nr. 6. They also attacked Prussian Foot Battery Nr. 16 and took 4 guns.
  • - In 1805 at Austerlitz "The Hussars, excellent ones of Hessen-Homburg, had many men and horses killed by the French skirmishers ..." (Amon von Treuenfest - "Geschichte des k.k. 11 Huszaren-Regimentes" p 233)
  • - In 1813 at Kulm the Prussian 9th 'Silesian' Landwehr found itself under heavy skirmish fire. The Landwehr "turned heel and broke.

 Returning now to body text, if you take the time to read the article in detail, it becomes quite clear your characterization of the fundamental of Napoleonic ground combat tactics is incorrect. Given this, your related conclusions are suspect. Sorry for spare bullets. Couldn't get rid of them. To end this on a bit of fun, may I present this clip on rate of fire?
 

  •  
  •  
  • Regards,

John Kettler

 

 

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Unfortunately, your content contains terms that we do not allow. Please edit your content to remove the highlighted words below.
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...