Panther Shot Trap Still Not Trapping

[c3k]

Lol. Elastic vs. Inelastic physics.

If a 3" shell hits 2 1/2" up from the hull top, what happens?

[c3k]

These are not light rays hitting a convex mirror.

[Andrew H.]

So rare that I wouldn't even factor it in my decision making, I wouldn't ever bet any money against those odds, and I'm not surprised that I have never seen it in my games.

I wouldn't do either, either...but if my company of Sherms blundered into a pair of Panthers, I'd hope it was modeled. Overmodeled, even

[JasonC]

BG - the whole hit area is more like 2400, and if we adjust the peripheral area lower - the large area that is just tracks on the lower right and left - as being about half as likely to be hit per pixel as the center of mass pixels - then the area is about 2000. So the shot trap area is more like 2.5% than 1.5%. Yes that is still rare. But no, neither figure is so low we would never see it in CM games. Panthers get into tank duels from the front in which they are routinely hit 5 or 10 times. It makes a difference whether one such duel gets a one in eight or even one in four combined chance that one of those hits "finds" the shot trap area.

[Vanir Ausf B]

Okay, I went ahead and counted the number of cells for the Panther lower mantlet, and I get 60. Since the whole area is about 56x69 (I think I miscounted), that yields a probability of about 1.5% (or about 99:1 odds).

As Steve says, this Panther shot trap thing is very rare.

So rare that I wouldn't even factor it in my decision making, I wouldn't ever bet any money against those odds, and I'm not surprised that I have never seen it in my games.

60 would only be the bottom edge of the mantlet. I'm not sure what the "whole area" refers to. My testing is for hits on the mantlet only, not the whole tank so any discussion of shot distribution outside of that framework is apples to oranges. Also, while counting squares to get the relative surface area (as best we can in 2D on a 3D surface) is helpful it does not tell the whole story. Because of the center of mass aiming model used in CMx2 the lower half of the mantlet should be getting hit much more frequently than the top half even though they have the same surface area.

I have not done any shot distribution counting on various parts of the tank, but there is no doubt that under the conditions I am testing under (partial hull down behind 1 meter high berm) I would estimate around 40-50% of all hits are on the mantlet.

[JasonC]

C3k - if a 3 inch shell hits with a center point of the impact 2.5" above the hull, that means it clears the hull by one inch. 3 inch diameter means 1.5 inch radius. It would hit the turret ring, that flat front turret section below the mantlet. That spot is much less protected than they mantlet face, let alone the glacis. Even a non penetrating hit there is likely to jam the turret by plowing a plug of metal where the turret traverse mechanism needs smooth clearance and no friction, to enable a ten ton turret to rotate on the hull. So the answers to your question are maybe KO, maybe jammed turret, but either way a lot worse for the tank than a hit on the glacis or higher up on the mantlet face.

Perhaps you were instead thinking of a hit within the rounds cross section, say 1 inch above the bottom edge of the mantlet. Then the answer is, very likely a deflection downward and that is the classic shot trap hit. If instead you mean hitting the top edge of the glacis, you might get a deflection up into the mantlet, or lower angle into the turret ring area. But as deflections into basically full thickness plates, those are less likely to KO than a downward deflection. One into the turret ring might still be lethal with a strong enough round eg 17 pdr, and any hit to the turret ring might jam it.

If instead you meant to say no one can tell, I deny it. There is variation surely. But some areas are much more exposed to effective hits from marginal rounds than the glacis face is, or the mantlet face is. It is not more accurate to pretend that complexity or variety implies we can just treat all hits as flat strikes on the best protected plates on the tank.

[JasonC]

Vanir - I get 19% or call it 1 in 5 mantlet area hits as shot trap locations, based on your picture. FWIW. If in your particulars for exposed area etc, 40% are hitting the mantlet, maybe 8% should be shot trap hits.

[Vanir Ausf B]

Ok, I re-ran the same test as in the first post. Everything is the same except that the game version is now 2.12 instead of 2.11. And I got a larger sample size.

British Sherman vs Panther A mid @ 800 meters. The Panther is partial hull down behind a 1 meter high berm. This covers nearly the entire lower hull, but not quite (I still saw an occasional lower hull hit).

Total number of hits recorded is 1096. This is for The. Mantlet. Only. Gun hits were also excluded.

I'll give the results by hit text.

RICOCHET: FORWARD TOP HULL PENETRATION: 0

FORWARD TOP HULL PENETRATION: Partial Penetration: 6

FORWARD TOP HULL PENETRATION: Penetration: 6

At this point I am seriously doubting that the old RICOCHET hit text is still working. Like MikeyD, the last time I saw it was before Charles messed with this code in an earlier patch (1.11?). It might be bugged. Certainly, if the ricochet penetrations are using the same hit text as direct hit penetration there is no way of telling them apart.

So there are two possibilities. Either the chance of ricochet penetration on hits TO THE MANTLET are 0% or 1.1%.

Note that there were no non-penetrating hits to the forward top hull. So we may have an answer to c3k's oft-repeated question...

[Vanir Ausf B]

Lol. Elastic vs. Inelastic physics.

If a 3" shell hits 2 1/2" up from the hull top, what happens?

These are not light rays hitting a convex mirror.

If we are talking in-game then it depends on whether the very few top hull hits I am getting are ricochet penetrations or direct hit penetrations. If they are direct hit penetrations then the answer to your question is irrelevant at this point. If they are ricochets the answer is that they always penetrate. So far.

[Vanir Ausf B]

Vanir - I get 19% or call it 1 in 5 mantlet area hits as shot trap locations, based on your picture. FWIW. If in your particulars for exposed area etc, 40% are hitting the mantlet, maybe 8% should be shot trap hits.

I count the gun and breech ring as taking up around 160 squares or 12% of the mantlet area. If you factor that out, as I did in my testing, the remaining area of the mantlet in my shot trap area is 17.7%

But to be fair, that shot trap area is just an estimate. Let's say for example I over-estimated its size by a factor of 2. Then it takes up 8.9% of the mantlet. Of if it is actually as tiny as BletchleyGeek guessed then it is 4.4%. You also have to factor in that not all hits bounce. About 53% appear to break up. So then you are down to about 2.3%. But this does not account for the center of mass aiming that will heavily skew the shot distribution towards the shot trap.

[c3k]

LOL. Thanks JasonC for explaining "radius" to me. That's the last time I text on a bus.

A shell will deform the mantlet. The metal will bend.

The shell will be subject to enormous forces if it changes direction rapidly. I leave it to you to determine the acceleration to deflect a shell by 45 degrees (or more) within the length of the shell.

If a shell comes in parallel to the hull roof, and just above the hull roof, then hits the mantlet, the dynamics are far different than if it comes in at the height of the barrel and hits the mantlet just below the center of the curve. Shot trap, indeed. That could be the hull roof penetrations with no ricochets.

I would be leery of trusting intuition on these matters.

There are a lot of anecdotes and very rare, if any, evidence.

The mantlet chin: why is it just on the lower 1/4 of the mantlet? Why not the lower 1/2?

Why did the Germans continue to demand zimmerit?

Momentum is lost. Metal bends, heats, breaks, and rotates.

This is an interesting subject, but there is no EVIDENCE.

Rather than hull damage, perhaps the number of driver/radio operator casualties could be used as an indicator of mantlet deflections? Or not.

This seems very similar to the claims of Typhoon rocket effectiveness.

Ken

[BletchleyGeek]

BG - the whole hit area is more like 2400, and if we adjust the peripheral area lower - the large area that is just tracks on the lower right and left - as being about half as likely to be hit per pixel as the center of mass pixels - then the area is about 2000. So the shot trap area is more like 2.5% than 1.5%. Yes that is still rare. But no, neither figure is so low we would never see it in CM games. Panthers get into tank duels from the front in which they are routinely hit 5 or 10 times. It makes a difference whether one such duel gets a one in eight or even one in four combined chance that one of those hits "finds" the shot trap area.

Okay, I miscounted the cells . Regarding to how often you'd see it, the math is easy.

probability of the Panther shot trap happening = 0.025

probability of the Panther shot trap not happening = 0.975

probability of the Panther shot trap not happening after x shots = 0.975^x

So, say you get 100 shots at the Panther in such conditions that enable the shot trap at all. The chances of not seeing it are an 8%. The number of trials you need to do in order to see this happening with a 99% are about 200 shots. In the situation you describe (the Panther takes 5 to 10 hits to the front) the chances of this not happening are an 88% and a 78% respectively. In order to be guaranteed to witness this, one should see 20 of those duels. I don't think I've played more than three or five scenarios featuring Panthers since 2011, and in many of those I usually kept them hull-down. So I'm not surprised to not have seen this. I am also the kind of person who thinks that OK Corral'ing Allied armour with Panthers isn't the right way to use them in the battlefield.

So even in that situation seeing the shot trap to spring would be something rare, and that assuming that the conditions that enable the shot trap to happen remain unchanged for all rounds you put on the Panther.

If somebody wants to verify this, he just needs to run a synthetic 1 minute long scenario more than 100 times in order to see this to happen. Or play Barkmann's corner ten times, as if Barkmann had Karl May's novels all over his head.

It is a very rare event, so rare that Battlefront should give a prize to the first guy to come up with a Youtube video displaying it

[MikeyD]

FORWARD TOP HULL PENETRATION: Penetration: 6

Mmmm, the ability to actually hit the hull top firing straight-on is pretty much nil. If you got six top hull penetrations you've got to wonder how the shell managed the feat. Bouncing off the mantlet perhaps?

Panther was designed with ammo stowage chests forward of the turret basket - exactly in line with the path a deflected round would be expected to take (you can actually see them in the game's Panther interior). Really bad design decision. The hope, in trying for a deflection shot off the lower mantlet, was not simply to get a penetration but to set off a catastrophic ammo bin explosion. Not only did the Germans go to the trouble of producing the expensive new chin mantlet to eliminate the risk, they also thickened the forward roof armor to 40mm.

[Battlefront.com]

RICOCHET: FORWARD TOP HULL PENETRATION: 0

FORWARD TOP HULL PENETRATION: Partial Penetration: 6

FORWARD TOP HULL PENETRATION: Penetration: 6

As I remember it, in game terms a Ricochet is something that doesn't penetrate at all. Meaning, if you see that the round has done nothing except bounce. Penetrations, on the other hand, show where the round wound up. It could be from a direct shot or from a Ricochet.

Which is to say that when you see a hit you won't see both Ricochet and Penetration, just one or the other.

Based on that it looks like your test shows about 1% certain shot trap + an unknown amount of theoretical trap shots that were not scored due to variable factors.

Remember, doing straight math on the surface area at best establishes the theoretical maximum number of shots trapped. It does not take into consideration mitigating physics factors which trump theoretical.

Steve

[Vanir Ausf B]

I would be leery of trusting intuition on these matters.

Do you suppose that Charles has access to some hard numbers the likes of which you keep asking for? That is very unlikely. More likely he is using some intuition of his own. Which is not only okay but absolutely necessary. To suggest that any opinion is invalid without perfect knowledge of all aspects of the issue is to hold us to an impossible standard.

This is an interesting subject, but there is no EVIDENCE.

After advancing 800 yards, Powers's loader, who also had his head out a hatch , realized that what he had thought was a brush pile was actually a Panther. He excitedly pointed it out to Powers, who ordered his gunner to fire. The first round hit the shot trap under the gun mantlet, ricocheted downward, killed the driver and bow gunner and set the Mark V on fire.

-- The Infantry's Armor: The U.S. Army's Separate Tank Battalions in World War II, By Harry Yeide

Source: Panther vs Sherman: Battle of the Bulge 1944 By Steven Zaloga

[Vanir Ausf B]

As I remember it, in game terms a Ricochet is something that doesn't penetrate at all. Meaning, if you see that the round has done nothing except bounce. Penetrations, on the other hand, show where the round wound up. It could be from a direct shot or from a Ricochet.

That may well be the case now. But I am certain that as recently as last year (sometime prior to 1.11) I saw both on the same hit. In fact, I've never seen the "RICOCHET" hit text in any other context.

If ricochet penetrations are using the same hit text as direct hit penetrations this is not optimal since there is no means of telling them apart. Not a HUGE PROBLEM, but maybe something to be tweaked?

Rounds actually bounce all the time. About 47% of them in my test, but I never saw a RICOCHET hit text. Interestingly, I saw one instance of a round ricocheting high into the air, do a lazy loop over into the neighboring Panther's lane and land on it's back. The hit text said "HIT: TOP REAR HULL". It did no damage.

Based on that it looks like your test shows about 1% certain shot trap + an unknown amount of theoretical trap shots that were not scored due to variable factors.

47% bounced somewhere other than onto the top hull. The other 53% just disappeared. I assume those rounds are breaking up but it's not clear what exactly happens to them.

Remember, doing straight math on the surface area at best establishes the theoretical maximum number of shots trapped. It does not take into consideration mitigating physics factors which trump theoretical.

Right. It also does not take into account how the center of mass aiming affects shot distribution.

[Battlefront.com]

Do you suppose that Charles has access to some hard numbers the likes of which you keep asking for? That is very unlikely. More likely he is using some intuition of his own.

Or the other alternative... he's making no presumptions at all and is simply leaving it to physics.

BTW, I don't think there's any reason to say the shot trap didn't exist. Although the Germans did do some things proactively despite no immediate problem (Zimmerit), I doubt they would have added the chin armor to later Model G's if there was no problem at all.

However, there is a corollary to that. Panther D, A, and most G's did not have chin armor at all. That means it was nearly 2/3rds through it's lifetime before the Germans decided they should stick the chin armor on the mantlet. If Panthers were getting routinely knocked out by shot trap hits I'm sure they would have done it much, much earlier.

Therefore, the evidence suggests such hits did happen but were extremely rare on a real battlefield.

Steve

[Vanir Ausf B]

Or the other alternative... he's making no presumptions at all and is simply leaving it to physics.

I don't think there is any question about that. But we need to keep in mind that these are simulated physics. In addition to the possibility of programing bugs there is the need to assign physical properties to objects in the game. As the recent change to Jpz IV armor quality ratings shows, there is inevitably some educated guesswork involved with this.

However, there is a corollary to that. Panther D, A, and most G's did not have chin armor at all. That means it was nearly 2/3rds through it's lifetime before the Germans decided they should stick the chin armor on the mantlet. If Panthers were getting routinely knocked out by shot trap hits I'm sure they would have done it much, much earlier.

Therefore, the evidence suggests such hits did happen but were extremely rare on a real battlefield.

It's a fair point, but it should also be pointed out that the mantlet chin wasn't the only change made. All G's (not just some of them as with the chin) had their top hull armor increased from 16mm to 40mm thick. The CMBN manual says this was for the entire deck, but according to Germany's Panther Tank: The Quest for Combat Supremacy by Thomas L. Jentz (pg 86), the increase in top hull armor thickness was only done on the front deck. Hmm...

I don't think shot trap penetrations were either extremely rare or routine, but somewhere in between (now I'm rhyming!).

[JasonC]

I don't have a problem with not reporting "ricochet", as long as there are the right number of partial penetrations and penetrations. From Vanir's reports, it sounds like we see -

12 out of 1096 hits specifically on the mantlet

2 out of 2234 hits anywhere.

That means 1.1% of mantlet hits where by area considerations we would expect more like some high portion of 20% - an order of magnitude too infrequent, in other words.

And it means 0.09% of overall hits where by area considerations we would expect more like some high portion of 2.5% - again an order of magnitude too infrequent.

It is great that the game does allow it to happen. But the frequency seems low by at least a factor of 5 and possibly a factor of 10.

Overall I'd say Vanir has spotted and documented the issue well, and that he presents a strong case for dialing the frequency of shot trap hits upward, significantly. One man's opinion...

[John Kettler]

Vanir Ausf B,

A most excellent and useful picture of the RAC Tiger 1. The account of its capture never made all that much sense to me, since the photo shown merely depicted the gun collar ding, not the gouges ahead of it or what the ricochet did to the the turret base. I now fully understand why the turret jammed in traverse, but I don't understand why the tank didn't back out of the fight, rather than being abandoned by the crew.

Regards,

John Kettler

[c3k]

Do you suppose that Charles has access to some hard numbers the likes of which you keep asking for? That is very unlikely. More likely he is using some intuition of his own. Which is not only okay but absolutely necessary. To suggest that any opinion is invalid without perfect knowledge of all aspects of the issue is to hold us to an impossible standard.

-- The Infantry's Armor: The U.S. Army's Separate Tank Battalions in World War II, By Harry Yeide

Source: Panther vs Sherman: Battle of the Bulge 1944 By Steven Zaloga

Nice.

(The pedant in me would ask to see the hole in the roof of the hull! )

I think there's a difference between a shot trap and the ricochet mechanics.

A trap, in this case, guides the round in. A ricochet would be more...dynamic.

The chin would, to me, stop the trap, not the ricochet.

VaB, I know you've done a LOT of drudge work. How about replacing the Panther in your test with a version with the chin? That would be interesting. I'd expect the number of roof penetrations would drop to zero if the chin were effective and modeled accurately. (Ref your results with the 6 penetrations.)

(Further, there was a reason why the Germans were working on the schmallturm (sp?): they wanted to shrink the mantlet even more.)

Ken

[Offshoot]

(The pedant in me would ask to see the hole in the roof of the hull! )

There is this picture, supposedly of a panther, though I can't make out which part of the hull has been hit - http://www.armorama.com/photos/review/8369/panther_osprey_review_006.jpg

From the Osprey book "Panther: Germany’s quest for combat dominance"

[Vanir Ausf B]

Nice.

I think there's a difference between a shot trap and the ricochet mechanics.

A trap, in this case, guides the round in. A ricochet would be more...dynamic.

The chin would, to me, stop the trap, not the ricochet.

True. As I mentioned earlier, 47% of all the hits on the mantlet I recorded ricocheted in some direction (this was a bitch to count, BTW). The question I have is why did only 12 of them out of over 500 ricochet down onto the top hull?

VaB, I know you've done a LOT of drudge work. How about replacing the Panther in your test with a version with the chin? That would be interesting. I'd expect the number of roof penetrations would drop to zero if the chin were effective and modeled accurately. (Ref your results with the 6 penetrations.)

Here I was thinking I was done and you go have an idea that makes too much sense not to try

(Further, there was a reason why the Germans were working on the schmallturm (sp?): they wanted to shrink the mantlet even more.)

Yep. Shrink it so they could add armor without adding weight. And eliminating downward ricochets into the driver compartment.

(The pedant in me would ask to see the hole in the roof of the hull!

Regiment stole my thunder. It's not the same tank, but here's a larger version of the same pic

[Sgt Joch]

I don't think shot trap penetrations were either extremely rare or routine, but somewhere in between (now I'm rhyming!).

Do you have empirical test data? Otherwise you are just guessing.

[JasonC]

Sgt Joch - asked and answered - the empirical test data is that it happened one out of eight times in their penetration tests of whether the US 76mm could defeat the Panther front. A hypothesis test p value of getting 1 out of 8 trials as a success in a geometric distribution, from a hypothesis of a prior probability of only 2 in 2243, is: 6.2 times 10 to the minus 32 - we can reject that hypothesis.

With a 2.5% prior probability that same test comes back with a p value of .07 - entirely possible by random chance by ordinary statistical standards. There is about a 1% chance the test result could have happened randomly with a prior probability as low as 1.5%.

Would it be preferable to have a much larger sample? Of course. But that doesn't mean you haven't been shown evidence, and evidence that is statistically inconsistent with the hypothesis that it only happened one time in a thousand. It might well only happen one time in 50, at a stretch one time in 100.