Film of Stuka 87G in action

[JonS]

Well, a round fired in training could be considered 'at long range with low chance to hit'. Also, Jason does say "they expended hundreds of times as many AP rounds as there even were enemy tanks", so all the rounds that the SARs punted across the Rhine don't count. For this.

But anyway: Betsy fired 1200 or 1500 rounds, depending on which count you take. Besty is clearly an outlier anyway ("Betsy was the only survivor of the original six guns adopted by the anti-tank platoon (she also outlived every single vehicle originally assigned to the anti-tank platoon)"), but leaving that aside, how many of those were HE? Say half? The article says she KO'd two halftracks (one of which may have been unarmoured).

So we get 600 AP rounds for two armoured kills without the training rounds (300 per), or 750 including them (375 per). Doesn't seem to make much difference to me, given the point was that vastly more rounds were fired than kills acheived, or even targets to shoot at. 375 is clearly vaster, but 300 is still pretty vast. IMO. YMMV.

*shrug*

Jon

[pamak1970]

Continuing my thoughts relative to the issue we discuss here, we have as many reasons to dispute Rudells record of destroying 500 tanks as to dispute Wittman's record of destroying hundreds of tanks and antitank guns during his service.

[JonS]

Originally posted by pamak1970:

If you do beleive that

"The average tank kills ~1 tank.

The average FB kills ~1 tank."

then you have to admit that FB are as much effective as tanks in engaging enemy armor in the battlefield.

Since I inflated the FB kills, and deflated the tank kills, for convienient round numbers, I don't believe that. I was trying to explain the theory to you.

Ne'er mind. Pat yourself on the back - your averageness has worn me out.

Jon

[Michael Dorosh]

Originally posted by JonS:

Well, a round fired in training could be considered 'at long range with low chance to hit'. Also, Jason does say "they expended hundreds of times as many AP rounds as there even were enemy tanks", so all the rounds that the SARs punted across the Rhine don't count. For this.

But anyway: Betsy fired 1200 or 1500 rounds, depending on which count you take. Besty is clearly an outlier anyway ("Betsy was the only survivor of the original six guns adopted by the anti-tank platoon (she also outlived every single vehicle originally assigned to the anti-tank platoon)"), but leaving that aside, how many of those were HE? Say half? The article says she KO'd two halftracks (one of which may have been unarmoured).

So we get 600 AP rounds for two armoured kills without the training rounds (300 per), or 750 including them (375 per). Doesn't seem to make much difference to me, given the point was that vastly more rounds were fired than kills acheived, or even targets to shoot at. 375 is clearly vaster, but 300 is still pretty vast. IMO. YMMV.

*shrug*

Jon

Does your brain hurt, yet, Jon? It was a simple question - Jason is saying that since millions of rounds were expended by tank crews and that only, say, thousands of tanks were killed, that it is evidence that many shots were taken that missed.

The natural question to ask then, is, what were those rounds actually shooting at. I think this is what you're saying in your post also, no?

Anyway, perhaps Jason has some figures for AP rounds expended and a breakdown of how many were fired in training - I have no idea how often AP was used on the firing range. I do know that target tanks are still surfacing and that they were shooting something at them. If it was on the order of 20 percent of their ammunition, it would indicate less long-range shooting (or, put another way, inaccurate shooting - forgetting about the use of more than one round per kill, shooting of already dead targets, etc.)

Whole point being that, once again, I don't doubt Jason has come up with a conclusion that feels right, but I just don't see that the evidence necessarily fits the conclusion. *shrugs back*

Will send my turn later. I suspect it will provide some more data on long range tank shooting **shudder**

[ August 16, 2005, 05:02 PM: Message edited by: Michael Dorosh ]

[Michael Dorosh]

Originally posted by JonS:

So we get 600 AP rounds for two armoured kills without the training rounds (300 per), or 750 including them (375 per). Doesn't seem to make much difference to me, given the point was that vastly more rounds were fired than kills acheived, or even targets to shoot at. 375 is clearly vaster, but 300 is still pretty vast. IMO. YMMV.

But training rounds are only one instance of rounds expended that aren't leading to kills, as mentioned in my last post.

a) AP rounds used to break walls in buildings (this was done at Ortona at the very least - AP first, then HE through the hole in the wall) (rarely done?)

AP versus softskins, gunshields or other non-tank targets (intentionally or not)

c) APs fired off when a tank came across an HE target and the gunner wanted to clear the barrel fast

d) APs fired in training

e) AP fired at already killed targets

f) multiple AP rounds fired at the same target, dead or not

If you want to account for the 600 AP rounds in your example, these would all be examples. Tiny ones, perhaps, but the training rounds would be - if Betsy is an example, which I'll freely stipulate she may well not be - a consideration.

Probably much ado about nothing but thought I would throw it into the mix.

[pamak1970]

Originally posted by JonS:

</font><blockquote>quote:</font><hr />Originally posted by pamak1970:

If you do beleive that

"The average tank kills ~1 tank.

The average FB kills ~1 tank."

then you have to admit that FB are as much effective as tanks in engaging enemy armor in the battlefield.

Since I inflated the FB kills, and deflated the tank kills, for convienient round numbers, I don't believe that. I was trying to explain the theory to you.

Ne'er mind. Pat yourself on the back - your averageness has worn me out.

Jon [/QB]</font>

[JonS]

Yeah, it's a fair question, I just don't think it tips the balance much.

Originally posted by Michael Dorosh:

... training rounds are only one instance of rounds expended that aren't leading to kills, as mentioned in my last post.

a) AP rounds used to break walls in buildings (this was done at Ortona at the very least - AP first, then HE through the hole in the wall) (rarely done?)

Often enough that the technique was given the sobriquet 'pick and shovel' in Burma. Not a great deal of Japanese armour there though (But there was some!)

AP versus softskins, gunshields or other non-tank targets (intentionally or not)

Probably, but 300 per is a lot of ground to make up.

c) APs fired off when a tank came across an HE target and the gunner wanted to clear the barrel fast

I'm thinking maybe 5? Particularly since HE was the more common default load.

d) APs fired in training

Which takes us from, for example, 375 to 300. Still a lot of vast-ness to make up.

e) AP fired at already killed targets

f) multiple AP rounds fired at the same target, dead or not

Yeah, and I think I have an OR on that somewhere. It's still multiple rounds fired at only one kill/target though.

Probably much ado about nothing but thought I would throw it into the mix.

Yeah *shrug* At the macro level I can't see aything wrong with it. Same as the macro assertion 'on average, systems fail to take out their own value'.

[pamak1970]

Does your brain hurt, yet, Jon? [big Grin] It was a simple question - Jason is saying that since millions of rounds were expended by tank crews and that only, say, thousands of tanks were killed, that it is evidence that many shots were taken that missed.

I beleive you need more details regarding the way the study was made relating rounds expended and kills.

For example, how did they calulate the number of rounds expended?

Did they use data from industrial production of AP shells?

If it is this case, then there are many cases of AP rounds expended ,besides firing against targets?

Rounds sunk when they were carried in transport ships .

Rounds lost when ammunition comvoys were attacked.

Rounds captured by enemy.

Rounds lost when a tank exploded and so on.

[JasonC]

Mike - the difference between rounds and kills is several orders of magnitude. Modest percentages won't make a dent in that. If you strung all the factors you care to name together, they won't amount to a single order of magnitude.

Most kills may have occurred under conditions in which only a few shots were needed for a result. But most shots were clearly taken in conditions in which the chance of any result was miniscule. Which has been true essentially throughout the history of firearms, incidentally, and is completely rational. Ammo is not scarce compared to full weapon systems. And firing opportunities aren't scarce, even compared to ammo.

As for the point of my discussion of low specific lethality for all weapons, I was correcting a common error or illusion, created by wargames, wartime propaganda, romantic ideas about the importance of individual prowess in warfare, etc, that all impart the idea that the average soldier is constantly making effective use of his weapons and disabling multiple enemies in succession, triumphantly.

A poster thought no one would field a weapon that didn't KO a major enemy weapon system every time out, so to speak. Practically none actually do so. I was removing his objection, by showing it proceeded from a false premise - that armies only field weapons that successfully KO major enemy weapon systems every time they are used.

The reason we know FBs did not kill appreciable numbers of tanks is quite independent of this global fact. The only reason to suppose they might are the outsized claims of the pilots themselves. Every cross check on those claims demonstrates they are false.

There aren't enough dead tanks for it, overall. The specific causes of loss, where known, cannot sustain so high a portion awarded to air attacks. Specific instances of large kill claims, investigated in minute detail, prove pilot claims completely unreliable, regularly off by a factor of 10 or 20 and uniformly too high.

Someone suggested perhaps a FB took out a tank over its operational life, but we know that isn't so. The Russians fielded 35000 Sturmoviks, comparable to the number of German AFVs deployed against Russia. It is not remotely true that all German losses were to air attack.

Again, we know most were lost to AT ground fire from tanks and towed guns, others to mines and mechanical issues etc. Only a modest portion might have been taken out from the air, meaning the numerator of tanks KO is much smaller than the denominator of ground attack planes doing the KOing. Ergo, the average IL-2 did not take out a single tank, not just on a typical mission, but ever, over its entire operational life (estimated by the Russians at 25-45 sorties, rising as the date progresses).

As for why they were fielded in such numbers anyway, obviously because they were useful against other targets besides fully armored AFVs, as I have repeatedly explained.

As for the idea that they might have had high specific lethality against tanks per engagement, but still have rarely KOed a single tank, it does not withstand scrutiny. If there were only a handful of IL-2s, looking for very rare targets, it might appear plausible. But there were tens of thousands of them (thousands at a time) operating over a huge front opposite tens of thousands of AFVs (thousands at a time), actively looking for them, and frequently engaging their units, as numerous combat reports tell us.

If they regularly KOed full AFVs successfully on a per engagement basis, they would have killed many more tanks than they can actually have killed. Because sorties are 35 times the size of planes, which are the size of tanks. If they had high specific lethality against a tank - say something like one chance in three, even - and encountered tanks even once in ten missions, then the average one would have bagged a tank sometime.

Since we know only something like 1/10 at the most, and more likely 1/20 German losses were actually to air attack, this can't remotely be right. One might bump the chance of finding tanks down to 1/20, and the specific lethality down to 1 in 10. Then you'd expect tanks kills from the air to be 1/200th of sorties, 35/200ths of plane numbers.

But that has Russian planes alone killing 17.5% of all German tanks, still far too high, probably 2-3 times too high. The encounter chance is already lower than is plausible. The chance of getting a tank when one is targeted therefore has to fall well below one in ten. You see, there are factors of 10-20 on the kill side and factors of 35 on the sortie side. There is no plausible way to span them and leave kill chance per engagement high.

What is missing on the other side in this discussion is a basic underlying awareness of how distorted the conventional picture of warfare people have been fed actually is. Officers, armies, propaganda, doctrine, weapons producers and advocates, games, writers, all systematically romanticize war as a tale of individual prowess, not as a largely anonymous, voracious industrial accident on a continental scale, whereby men are ground up by machines.

This isn't just another theory of strategy or doctrine about war, it is the truth about it. A truth all the aforementioned parties are anxious not to look in the face. Both themselves, and for others. They do not want the average person thinking of individual soldiers as mere targets, at least as likely to be shot before he achieves a darned thing, as to accomplish anything. They don't want you focusing on the dry accounting identity that says, the average man's chances are piss-poor.

They want you to imagine the heroic warrior rising to do battle every day, confident in the safety afforded by his prowess, his daunting enemies collapsing before his triumphant weapons and skill repeatedly, day after day, until scores are piled up at the feet of each such happy warrior. You are never to ask how happy the warriors down in the piles are. They must just be stupid or something. Probably, they think in attrition terms.

If instead you believe in maneuverism and prowess and providence and skill and your luck, then maybe just maybe when the machinegun fires in your direction you will throw caution away and fire back - likely getting chopped into fresh ground hamburger for your pains, but the attempt is wanted. The cogs in the collosal man eating machine should not think of themselves as cogs, nor are targets, nor as accounting entries, but as heros and gods.

The merest tincture of math is enough to explode this soap bubble illusion, and to demonstrate it is a lie for practically all combatants. But the illusion having been energetically constructed, often fervently believed for deep psychological reasons by those doing the constructing, the conclusion is predictably resisted. Men willingly die for that psychological illusion, or perhaps it is more accurate to say, that psychological illusion makes bearable the threat of death they labor under regardless.

There are men however who prefer their reality straight, whatever it says.

[pamak1970]

The reason we know FBs did not kill appreciable numbers of tanks is quite independent of this global fact. The only reason to suppose they might are the outsized claims of the pilots themselves. Every cross check on those claims demonstrates they are false.

There aren't enough dead tanks for it, overall. The specific causes of loss, where known, cannot sustain so high a portion awarded to air attacks. Specific instances of large kill claims, investigated in minute detail, prove pilot claims completely unreliable, regularly off by a factor of 10 or 20 and uniformly too high.

Now ,this is the appropiate argument for the issue we discuss here and i will agree.

Someone suggested perhaps a FB took out a tank over its operational life, but we know that isn't so. The Russians fielded 35000 Sturmoviks, comparable to the number of German AFVs deployed against Russia. It is not remotely true that all German losses were to air attack.

Again, we know most were lost to AT ground fire from tanks and towed guns, others to mines and mechanical issues etc. Only a modest portion might have been taken out from the air, meaning the numerator of tanks KO is much smaller than the denominator of ground attack planes doing the KOing. Ergo, the average IL-2 did not take out a single tank, not just on a typical mission, but ever, over its entire operational life (estimated by the Russians at 25-45 sorties, rising as the date progresses).

The above logic implies that all missions of IL-2 were directed against tanks and that all 35000 assault aircrafts were destroyed .

Therefore the conclusion that the denominator of 35000 aircrafts is much larger than the numerator and the assult aircraft did not take out a single AFV during its operational life.

That is something that needs a deeper investigation in order to understand the magnitude of denominator.

Think for example that in the same way we can talk about the size of tanks employed by Russians and compare it to the actual loss of German tanks cause of Soviet armor.

Again the denominator is much larger than the numerator.

Same with antitank mines and so on.

We can also talk about the German number of tank busters compared to the actual number of Soviet armor destroyed by aircraft.

What are the denominators and numerators in that case?

This can not lead to conclusions about the effectiveness of a platform in engaging another one.

If they regularly KOed full AFVs successfully on a per engagement basis, they would have killed many more tanks than they can actually have killed. Because sorties are 35 times the size of planes, which are the size of tanks. If they had high specific lethality against a tank - say something like one chance in three, even - and encountered tanks even once in ten missions, then the average one would have bagged a tank sometime.

Since we know only something like 1/10 at the most, and more likely 1/20 German losses were actually to air attack, this can't remotely be right. One might bump the chance of finding tanks down to 1/20, and the specific lethality down to 1 in 10. Then you'd expect tanks kills from the air to be 1/200th of sorties, 35/200ths of plane numbers.

I think now we are really getting inside the core.

of the issue we discuss here.

First i do not agree again with the implication that all mission are directed against tanks.

"sorties are 35 times the size of the plane" -plane fleet i assume.

That needs more data.

The other thing i have reservations is the observation of 1/10 more likely 1/20 of German losses cause of air attacks.

The 3-4% of losses cause of aircraft attacks according to some Soviet sources talks about the number of tanks beyond repair .

The a ctual losses of all tanks put out of action is more than that.

I recall a figure of 180 tanks put out of action which at the end became 70 tanks (110 were repaired).

This number of 70 tanks was equivalent of 3-4% of all losses but if we use the number of 180 tanks instead we automatically see a percentage of 10% put out of action.

So, there is an issue of definition here regarding what is a loss and what is not.

Antitank mines fore example result in mobility kills.

Do they inflict losses or not?

We have also other issues unresolved.

The chances to spot a tank are not so big in my opinion inspite the thousands of planes and tanks. If we count weather effects ,clouds ,smoke and thousands of interceptors which are also present in the same area, it is not so easy to spot a unit.

I would be interested to see the percentage for example of missions having to abort cause of enemy aircraft,not counting the aircrafts shot down

(we have already taken this effect in consideration when we talk about 20-40 sorties per each aircraft during its life) .

We can also see the above from a different perspective.

Since the denominator of Soviet armor is much larger than the nominator of actual german armor lost cause of Soviet tanks, and since the actual number of engagements that a tank was able to participate during its life period is sginificant .can we conclude that an armor unit during an average armor class was not so lethal against enemy units?

Experience shows that actually armor classes were short and brutal with many tanks K.O among the sites which participated.

On the other hand statistics regarding ammunition expense related with losses implies that probability of kill is very low

Is there a contradiction here?

This is relative with what i posted earlier regarding the method they used to link number of rounds expended and actual shots against a target and if it was contacted the way i suspect-either studying industrial production,or even various unit ammo requests, then the results are not credible cause in this way you count shells that were lost without having them directed against enemy targets.

For example an ammunition dump being bombed resulted in the loss of thousands of AP shells

In my opinion if the typical number of ammunition carried inside a tank was not sufficient to give a good chance to K.O an enemy tank in a typical fight, then i become suspicious of the theory.

Decent chance to K.O a tank with a stock of 40 let say shells means an actual average chance of few percentages for each shell to score a hit in actual battlefield conditions.

[ August 16, 2005, 11:04 PM: Message edited by: pamak1970 ]

[dieseltaylor]

When I see people talking about averages I get very nervous. I can assure everyone that the climate in a year in the Sahara is a pleasant 75F 23C? on average.

Of course at midday it is over 100F and at night it drops to freezing but the average is true. Now of course you may say that most activity takes place in daylight and therefore my mathematical average is bogus in relation to human activity.

If we average anything over a selected time we can reach a figure more or less favourable to a proposition. To the 10-20 German tanks nailed by air-power at Mortain the time period of a day gives an impressive average.

Now the other interesting point is proximity. If you can only exert force at a particular point and your enemy has the ability to project power by artillery, fighter bombers speedily to where you attack these together with the ground forces defeat your local ground superiority.

So for all the talk of life-time averages the ability to apply power to a point is actually rather important and seems to be excluded when talking of the value of a weapons system.

As someone wisely posted the value of a platoon on the beachhead outweighed a division in Calais and that surely is the point. As Pamak1970 has pointed out the load-out on a tank was that number for a reason. So to re-phrase it to average speak the tank carried enough ammo for to be effective in an encounter.

Getting sufficient advantage in a sector to make a costly attack is a function of mobility, resources, and time. The rest of the front can be ticking over with desultory artillery fire, recon raids etc. When the big attack goes in those resources doing little are going to provide a huge averaging effect to the big battle where lots of serious ground taking/killing is taking place.

So there you are folks averages are fun but you choose your time frame and it makes everything look different. So on average in 1939-1940 Stuka's rule : )

BTW the other big point surely is taking systems in isolation to discuss their effectiveness vis-a-vis the enemy system is positively perverse as it is the mix of systems you deploy that provides an army's effectiveness. And that effectiveness is a function of terrain, climate, and what time span you are looking at.

I despite the talk of the individual being subordinate to the macro picture I suppose rather romantically think that Hitler and Churchill had some importance in the war,the commander at Anzio, the code breakers, the putting of the Merlin in the Mustang. I am sure you grogs can think of many instances where big decisions, and little acts of thought or heroism, had quite a large impact on the wars progress.

[Baneman]

A fascinating thread.

If I may chip in with a few thoughts ( no maths and stats from me, my brain canna do that stuff ).

Originally posted by JasonC:

... The cogs in the collosal man eating machine should not think of themselves as cogs, nor are targets, nor as accounting entries, but as heros and gods.

The merest tincture of math is enough to explode this soap bubble illusion, and to demonstrate it is a lie for practically all combatants. But the illusion having been energetically constructed, often fervently believed for deep psychological reasons by those doing the constructing, the conclusion is predictably resisted. ...

This is obviously the reason behind the popularisation of "Aces" - take an individual who has beaten the odds ( luck, skill and right-place-right-time ) and say "you too can be like this." It aint even remotely true, but it keeps the cogs happy.

However, this is not to say that their exploits are untrue. We all know that the German air-to-air claims were subject to fairly stringent checks and many aces had several victories disallowed because no one was able to corroborate their AAR.

So despite all the evidence given above as to the relative ineffectiveness of fighter-bombers versus tanks, we have to consider that perhaps someone like Rudel DID in fact achieve what he is credited with. Ok, given the difficulty in verifying ground kills, perhaps he only actually managed half of his credited total - say 260 tanks. That's still a huge number, but he is ahead of the curve ( again to relate to fighter pilots, we know that approx. 10% of all fighter pilots accounted for approx. 80% of all kills - to probably misquote a stat ). So yes, there could be many Stuka pilots out there who just plain couldn't hit a tank, but that doesn't mean there weren't some who could, to pull the average up. ( I don't buy that the German Ju87G trials could have been so flawed that they thought it could kill enemy armour when it couldn't - at worst, they must have been testing against static targets, so that could throw results off. )

But, there are also other factors that could be working in Rudel's favour. A fighter ace like HJ Marseille is reputed to have shot down enemy planes with an average of 7 rounds. Now these planes were manoeuvering at high speeds, so it is not inconceivable that a skilled pilot could hit a vastly slower ground target from a somewhat slowish plane.

Ok, let's consider that our hypothetical Rudel has hit the target tank.

He is firing a so-called "dinky" 37mm. The ammunition is tailored to tank killing, so that helps offset its dinkyness. He is firing from above/behind/to the side which will also help. But most importantly, not all tanks are created equal - the Soviets fielded "obsolete" tanks ( BT7's, T26's, T28's ) right up until the end of the war ( in recce and other roles ) - so did the Germans, fielding PzII's in the recce role. Not to mention the Lend-Lease Stuarts and Shermans. All of these would be easier to kill than a T34/85 or IS tank. Plus, our target "tank" could be an open-topped AFV of some kind. To our pilot, it's just Thing-With-Tracks-and-Pointy-Bit = Tank.

So it is within the realms of possibility that an above-average pilot like Rudel could pull off multiple kills on tanks without skewing the overall average ineffectiveness of the tank-buster aircraft.

Bear in mind also that as described for the West Front, the fact that the aerial attack was ineffective may not preclude the chance that the crew of the tank would panic/bail out/run away/get killed by humble infantryman's bullet. And even if the Stuka only immobilised a T34, say - in a place like Kursk, the crew is unlikely to stick around. This would be a valid "kill" for the pilot having removed a weapons system from the battle ( not really caring whether or not it can be repaired )

It's all very interesting.

[dieseltaylor]

John S said

"Right, but given all that, CM planes never seem to then miss. Which is, of course, what these two threads have been all about. "

As it was because of this remark I did the trials that I posted elsewhere I will copy the results here for those who may not read the "Does flak work" thread. I thought it interesting enough in its own right and thought many readers might give up before reaching it if posted solely here.

"Italy, September 1944 midday, fine, cool, moderate trees, gentle slopes, medium map

For the Germans: platoons of Stugs, Tigers and Hummels one gun, no flak.

For the US two Mustangs with bombs 868 points

Result: 1 gun damaged Tiger, three bailed Hummels – I neglected to make the distinction between abandoned and knocked until a later stage.

I then added two 20mm flak ,which is a little light in my book for the points involved.

Result: Three bailed Hummels, one killed flak, one killed gun [ – too near the flak!] Interesting as the flak was taken out first cutting the AA fire. Also the only Hummel under a hide order in the open was the last one attacked – three times and lost a crew man – they did not bail.

Ran the test again. No near misses. 1 knocked out Hummel – ablaze. And one bailed Hummel. It was amusing to see the plane again attack the bailed Hummel presumably going for a kill.

Third Test: Tiger bomb bracketed by both fighters so crew was shaken twice. Two Hummels crew bailed with 3 casualties each time. The one parked in the open was attacked last again three times and was abandoned by the crew on the second pass.

New map, same parameters but now a quad 20mm and a 37mm flak. Quite gratifying as all the bombs missed and the strafing attacks were abortive. I was pleased to note that both planes had been shot down in the results. However I got no man casualties BF please fix or something!!!!

From this I deduce that the game results approach real life provided you actually provide reasonable flak cover. All purchase for the Germans were about 2000 so a flak spend of 40pts is not overly generous whereas a hundred points was lethal. : )

Note that there might be mileage in RL for the crew to remount their Hummels and ride out of the battle area. Only one was destroyed on the battlefield so if one were to make an analysis of the battlefields it would reveal for sure one burnt out Hummel for 10 sorties, two dead aircraft.

Of course the Germans would know about seven Hummels who never got to fight : ) These possibly might have taken engine and track damage but given the crew casualties and the shape of the vehicle shooting into the gun compartment has to be a big favourite. It was noticeable that the approach by the aircraft was mainly from the rear of the Hummels."

I hope this bears out my view that adequate flak makes a hell of a difference to FB effectiveness in CM.

[JasonC]

pakman1970 - you need to be much more careful about your ascription of assumptions to the argument you are examining. Nowhere did anyone assume every sortie was directed at tanks, for instance. Nor was it necessary to assume all IL-2s were lost, it is quite sufficient that they were produced and did not kill a tank by the end of the war.

Certainly the average Russian tank did not kill a German tank. Large formations of Russians tanks killed a few tanks as their edges brushed. Occasionally, large formations of Russian tanks were destroyed, or cut in half, in rapid intense fighting, in the course of which they took out a modest number of German tanks themselves, rarely their own number lost.

Nor can tanks lost several times save the image of high specific lethality. Non-combat losses are on the other side of the ledger and equally large. That is, roughly for every tank taken out twice there is another that fell out without having been KOed by direct enemy action. Sometimes after being taken out once, certainly. But total KOs is not going to exceed tanks fielded by any large factor. Percentages, sure, but none of the issues were are looking at turn on questions of 30% more.

What makes tanks effective weapon systems is not their break away exchange with enemy tanks. They would not last a year on average if that were their primary role. When a tank battalion goes weeks losing a single vehicle per week, it is earning its cost.

By hosing infantry positions with impunity from 400 yards, by crushing single batteries of PAK in engagements that outnumber those PAK by 10 to 1, by intimidating enemy formations into retreat or breaking off attacks, etc. Combat is in fact vastly more tenative than wargames show, with the sides dominating safe areas by superior force and the other side locally fearing and therefore respecting that dominance.

When instead they smash each other wholesale, one side can score higher than unity only if the other side scores less. Combined,they are always under unity.

[JasonC]

On actual exploits of aces, even if there were no skill differentials in war there would still be aces. By mere statistics. Lots of uncorrelated random events add up as bell curves not as flat lines. That's just luck. When the curve is not bell but Pareto, as it often is, it means something skill like is operating and that something is itself distributed.

Luck continues to operate. You can't remotely conclude the guy with the highest score is the best, unless the class considered is tiny. It is overwhelmingly more likely that he is just the luckiest of the good. Another man with equal or higher skill just happened to get killed sooner, before he could run his score so far, for example.

The romantic illusion is that there is safety and purpose in prowess, which men readily hope for and identify with. Nobody thinks he is in the bottom half of the prowess distribution, even though we know half are. The less skilled think they are average, the average think they are modestly above average, anyone actually appreciably above average thinks he is almost as good as the best. Which is self delusion of the most straightforward kind. (This goes for officers convince of their maneuverist genius, etc).

Most CM players have won no more games against humans than they have lost. Same story, same mathematical law. All children are not above average, as the adage goes.

[JasonC]

Consider the average effectiveness of German AFVs. They were above average weapon systems, used by the side that lost of course but in an army that inflicted higher losses than it took, particularly in Russia. But the average German AFV took out a single Russian (allied, but for the average one it was a Russian) AFV. The superior types, Panthers and Tigers, might have broken 3 apiece. The Tigers at the top of the distribution might even have managed 5.

The romantic types, and own side claims, would try to tell you it was 3-5 for the vanilla StuGs, and 10-15 for the superior types. Such claims do not withstand scrutiny.

The conclusion comes from partitioning the limited known losses among the causes of loss, and estimating the relative average effectiveness of the types. The average effectiveness of a numerous type can only be adjusted upward, if another numerous type has its average effectiveness adjusted downward. Because the total achieved is quite limited, and many other means besides tanks are "competing" in the space (mechanical losses, PAK, mines, infantry AT, artillery and air, etc).

On the side that scored more heavily, the average solid quality AFV in the middle of the distribution may have accounted for a single enemy tank over its entire operational life. (All the "more than one" weight is born by the superior types, and superior instances in the vanilla ones). On the Allied side, that is quite impossible - only above average AFVs got a single enemy AFV.

[JasonC]

"I am sure you grogs can think of many instances where big decisions, and little acts of thought or heroism, had quite a large impact"

Readily, that is not the proposition in dispute. The question is, are the romantics capable of holding their attention for more than five seconds on all the little acts of thought or heroism, that ended with a dead hero and splattered brains? Or the big decisions that resulted in bloody fiascos that threw away thousands of men for nothing? Or the incomparably more common occasions, in which a typical man did the best he could, suffered immensely but not fatally, and achieved precious little? War does not consist of its highlight reel. And if you look carefully, even in the highlight reels somebody involved is screwing up (just not the one the cameraman expects you to identify with).

[A.E.B]

Out of curiosity, how many of us here have seen photos or gun camera footage of FBs/CAS of any nation taking down armour.

A quick check of my books/videos turned up trucks (in one case a whole german supply column lost to IL2s near Kursk - although it appeared to be a chain-reaction of exploding trucks), trains, infantry, planes on the ground, ships, and even horses and carts.

The only example of an actual attack on armour was from Korea, with Starfighters rocketing T34s (with limited effect).

Gun camera footage and stills from planes do exist, but I cannot recall seeing tanks in them, let alone tanks that are being KOed.

So has anyone here actually seen such photos?

Note: I have many photos that claim to be the handy work of FBs/CAS. For instance an abandoned T34 with a large crater nearby claimed as the handy work of a stuka. But I have no photos of stukas actually bombing tanks.

A.E.B

[undead reindeer cavalry]

Originally posted by JasonC:

The romantic types, and own side claims, would try to tell you it was 3-5 for the vanilla StuGs, and 10-15 for the superior types. Such claims do not withstand scrutiny.

for what it's worth, here's a word about Finnish StuGs, how they were used and what kind of results they got.

Finnish StuGs were at the front for three weeks, including about one week of actual combat.

it was the first time the StuGs were used, so the crews weren't elite hard-core veterans.

the tactical way the StuGs were used was very infantile and certainly inferior to how e.g. Germans used them.

terrain was rather unsuitable for the use of turretless assault guns.

Finns faced better-than-average Guards units, armed with top quality equipment (T-34/85, IS-2, ISU-152 etc).

Soviets had massive numerical superiority and almost total air-superiority (most StuGs that were lost were lost to enemy air activity).

the results:

in average, a Finnish StuG scored 4+ tank kills (+ loads of other enemy equipment) during that period.

in average, for one lost StuG, Finnish StuGS destroyed over 10 Soviet tanks.

if Finns would have had proper tactics, more experienced crews, less outnumbered, more suitable terrain, faced more typical Soviet units, had seen combat for more than just one week, or just one of these kind of things, the 4+/1:10+ ratios could have easily been dramatically higher.

[pamak1970]

Originally posted by JasonC:

pakman1970 - you need to be much more careful about your ascription of assumptions to the argument you are examining. Nowhere did anyone assume every sortie was directed at tanks, for instance. Nor was it necessary to assume all IL-2s were lost, it is quite sufficient that they were produced and did not kill a tank by the end of the war.

Certainly the average Russian tank did not kill a German tank. Large formations of Russians tanks killed a few tanks as their edges brushed. Occasionally, large formations of Russian tanks were destroyed, or cut in half, in rapid intense fighting, in the course of which they took out a modest number of German tanks themselves, rarely their own number lost.

Nor can tanks lost several times save the image of high specific lethality. Non-combat losses are on the other side of the ledger and equally large. That is, roughly for every tank taken out twice there is another that fell out without having been KOed by direct enemy action. Sometimes after being taken out once, certainly. But total KOs is not going to exceed tanks fielded by any large factor. Percentages, sure, but none of the issues were are looking at turn on questions of 30% more.

What makes tanks effective weapon systems is not their break away exchange with enemy tanks. They would not last a year on average if that were their primary role. When a tank battalion goes weeks losing a single vehicle per week, it is earning its cost.

By hosing infantry positions with impunity from 400 yards, by crushing single batteries of PAK in engagements that outnumber those PAK by 10 to 1, by intimidating enemy formations into retreat or breaking off attacks, etc. Combat is in fact vastly more tenative than wargames show, with the sides dominating safe areas by superior force and the other side locally fearing and therefore respecting that dominance.

When instead they smash each other wholesale, one side can score higher than unity only if the other side scores less. Combined,they are always under unity.

If you did not assume that every sortie was against tanks then why you mentioned the 30000 IL-2 multiplied by 20-30 missions in order to describe the size of the denominator?

You do assume in this case that all sorties of IL-2 were aimed against tanks.

Farthermore if you choose to calculate and multiply all 30000 planes with the average number of 20-30 sorties which is the duration of their average life of an assault plane,you basicaly imply that these 30000 planes "spent" their average operational life, which is like saying that they were destroyed.

If you do not want to assume this ,then you can not multiply the whole number of planes with the average number of "sorties per operational life".

Regarding that the average Soviet tank did not kill a German tank , i do not have any objections.

My point is not to prove that statistics of strategic level should point that a certain system did K.O an enemy one.

My point that these statistics are not indication of the effectiveness of the weapon in the tactical level.

Does the fact that a Soviet tank did not KO a German one make us claim that Soviet tanks were not effective to fight enemy tanks in the battlefield?

To me it is irrelevant if we can say at strategic level that an average Soviet tank did not KO a german tank,or if it did KO five.

I accept whatever you present in this case without challenging the number itself.

I focus on the conclusions.

Regarding your comment about tanks ,i do not see what is the point you try to make.

First if a tank battallion goes for weeks without loosing more than a tank or two , is because the tank battallion goes for weeks without participating in an armor battle.

When they get the chance to participate as a front line unit ,losses are inflicted in a matter of hours-or they inflict to the enemy.

At least that is my understanding of battlefield experience.

The high rate of tank non combat losses is related mainly to mechanical problems .These tanks are not KO since they do return in the battlefield

but i am not sure why you think it is so important to mention it.

If your point is that this number of non-combat casualties implies a farther diminishing of the "effect" of tank shooting against another tank, i will not agree for an additional reason i have not mentioned.

Non-combat casualties does not affect only the number of tank losses by enemy armor.

It affects also the number of tanks engaging and shooting at enemy armor.

You see only the effect on the numerator and ignore completely the effect on the denominator here.

Regarding the effectiveness of the tank cause it destroys other key facilities and weapons, i will agree but this is a seperate observation which does not point anything about the effectiveness of the same platform against other tanks.

From what i see, you seem to change your position from FB not being effective against tanks ,to tanks not being effective against tanks either.

[pamak1970]

Originally posted by JasonC:

Consider the average effectiveness of German AFVs. They were above average weapon systems, used by the side that lost of course but in an army that inflicted higher losses than it took, particularly in Russia. But the average German AFV took out a single Russian (allied, but for the average one it was a Russian) AFV. The superior types, Panthers and Tigers, might have broken 3 apiece. The Tigers at the top of the distribution might even have managed 5.

The romantic types, and own side claims, would try to tell you it was 3-5 for the vanilla StuGs, and 10-15 for the superior types. Such claims do not withstand scrutiny.

The conclusion comes from partitioning the limited known losses among the causes of loss, and estimating the relative average effectiveness of the types. The average effectiveness of a numerous type can only be adjusted upward, if another numerous type has its average effectiveness adjusted downward. Because the total achieved is quite limited, and many other means besides tanks are "competing" in the space (mechanical losses, PAK, mines, infantry AT, artillery and air, etc).

On the side that scored more heavily, the average solid quality AFV in the middle of the distribution may have accounted for a single enemy tank over its entire operational life. (All the "more than one" weight is born by the superior types, and superior instances in the vanilla ones). On the Allied side, that is quite impossible - only above average AFVs got a single enemy AFV.

Why you use statistics about losses to make conclusions about "average" weapon systems?

There are many things that affect losses besides the quality of a weapon system.

Think for example that case where Russians were using the exact same type of armor like germans did.

Now at the end of the war ,how would you calculate the effeciveness of this common AFV.

Would you use the number of Soviet industrial production and relate it with the German AFV losses cause of Soviet armor action?

Would you use the number of German AFV production compared to the Soviet AFV losses which would give a totally different result ?

P.S i assume you agree that the above results would be different and not claim that because both used the same AFV ,these resukts of "AFV effectiveness" should be about the same nomatter which numbers we use.

[dieseltaylor]

So there we have it then. Those who wish to average a weapons effectiveness over the length of a war against those who measure its effectiveness on what it did when it was in action.

If we look at time and the Matilda Queen of the Desert we would say pretty useful. Let us assume that we built thousands but when we came to invade mainland Europe the basic German tank was the Panther..... The result is self-evident but I am curious what happens to all these ineffective and obsolescent tanks. Are they included in these large figures quoted or is there some mechanism where only tanks actively in the front line count towards the figures.

And of course active in the front line but mean excluding the second rate armour in Yugoslavia, or in Norway and Denmark where it is relegated to airfield security. Technically a tank but its survival or destruction is irrelevant other than it provides " averages" for alll Axis tanks?

[Michael Dorosh]

Originally posted by dieseltaylor:

So there we have it then. Those who wish to average a weapons effectiveness over the length of a war against those who measure its effectiveness on what it did when it was in action.

I think it is more like looking at a weapon's effectiveness over the course of the war in order to arrive at its "true" relative effectiveness while in action....

[dieseltaylor]

JasonC

My theory of war is not that disimilar to yours in that vast amounts of logistics occur to put numbers of men opposite each other, hopefully equally well armed, to sit there holding each other in check whilst someone somewhere works on a cunning plan to win the war. This normally involves getting a lot of men killed at a particular point.

The average expectancy of survival for everything here diminishes drastically.

I was mulling over Taranto as being 21 planes creating a major war effect for relatively little effort. Of course it should not have worked and they were probably flying one of the most obscelent planes around but at the time thay did it it created a major effect. Of course if they had sunk the ships 4 years later the war average for Swordfish strikes would have been the same but the strategic effect not nearly as important. Makes you ponder.

The German paradrop on Crete if Maleme had not been taken and held. Alter the course of the war? Majorly I say.

[John D Salt]

Originally posted by dieseltaylor:

[snips]

The German paradrop on Crete if Maleme had not been taken and held. Alter the course of the war? Majorly I say.

Really? OK, we've wandered a fair way from the original point here, but would you care to state how exactly the German occupation of Crete caused major inconvenience to the Allies in the prosecution of the Mediterranean campaign?

All the best,

John.