This is inspired by Bill Rutherford's survey of WW2 and modern rule sets in VII/5. This seemed fair and accurate to me as the author of the oldest set covered (British edition actually June 1973, not 1975), but like all reviews tends to reflect the predelictions of the reviewer. To put things in context, after 14 years we have a new set of WRG rules for 1925-1950 now finishing testing. However, this is not intended as a blurb for the new set or even as a description of it, but as a statement of some of the philosophy behind it.
Most WW2 and Modern gamers, and rule authors even more so, are fascinated by the perforation of steel armour by projectiles. An extreme case among authors has mechanisms to decide which part of a vehicle is hit and to calculate the precise value of the armour protecting it in terms of slope and thickness. He relies on tables of penetration without knowing how these were arrived at, and revels in detailed technical descriptions of vehicles and their armour. He is less interested in what happens when penetration occurs or fails to occur, often assuming the one automatically brings destruction and that the other leaves the vehicle functional.
The professionals use a simpler approach. At a defence equipment exhibition I saw a poster produced by one of the many manufacturers of 105mm APFSDS shot. This showed the plan view of "a typical modern battle tank" (an obvious T.62), superimposed on a 3601 clock dial with a series of concentric circles marked with ranges in metres. A series of continuous contour lines were then overdrawn joining all the range and angle combinations offering first 75%, then 50%, then 25% probability of a kill (not of a penetration). Apart from being something of an eye opener as the effect of relative vehicle heading and showing much less effectiveness than implied by published penetration figures for comparable ammunition against standard targets, this was an elegant method that allowed both for increased angle of descent and the reduced probability of thinner armoured lower parts being hit as range increases.
Statistics Inconsistency
When it comes to estimating weapon performances, the post-WW2 rule writer is somewhat handicapped. He has no tables of penetration trials and only a moderate amount of help from combat histories, so is at the mercy of official handouts, and even worse, the manufacturer's brochure. Getting at the truth is tricky, but I can offer some guidance from my experience. Information from U.S. manufacturers on their products is optimistic. US Army information on Soviet weapons can be optimistic or pessimistic dependent on the impression needed on Congress. Congressional committees are pessimistic. Israeli sources are so selective as to produce a very optimistic overall effect. French sources are accurate about rival nations'weapons when these are inferior. British manufacturers are truthful - but usually not allowed to publish detailed figures. (The Centurion's armour thickness is still secret 20 years after it became obsolete, but you can go and measure it yourself on the sectioned example in the Bovigdon tank museum. The turret front is much thicker than the 150mm parrotted in all the books).
To show the perils, one of our more recent British rivals quotes the Dragon anti-tank missile (127mm calibre and 2.44 Kg warhead weight) as inferior in hitting power to the British Swingfire (170mm and 7 Kg warhead), apparently because the former's brochure quotes a penetration of 600mm, the latter's "more than 500mm". Some of you may remember the trouble the 66mm M.72 LAW (305mm penetration) had killing minimally armoured PT.76's in Vietnam. It's not the depth of the hole you drill that matters, it's what ends up getting through it.
However, all is not rosy for WW2 either. It may surprise enthusiasts to learn that there are very often several official trial results tables for a single weapon's penetration, varying by as much as 20% in their conclusions, even when gross sources of error such as ammunition differences, whether the target angle is 300 or 00, and national bias are eliminated. British trials used a quarter-worn gun and averaged a number of shots at 300 from normal. German trials were similar, and Soviet trials took the worst result. it has been suggested that US Ordnance Dept. trials used a brand new gun and eliminated poor results. Maybe some of you might like to check up now now you have a freedom of information law? The state of captured guns used in trials is anyone's guess.
There was certainly one instance of deliberate cheating. When the new US 76mm gun started replacing the 75mm gun in Shermans, American users complained that although it looked impressive, it did not seem any more capable of killing Tigers or Panthers than the old gun. After much pressure, the Ordnance Dept. arranged a public trial, but took the precaution of secretly refilling the cartridges with superior British 17pdr propellant. The ensuing triumphant vindication still left American tankers with a gun that according to their action reports could not consistently penetrate a Tiger's front beyond 200 metres.
Ammunition variation isanother potentsource of error. Gross differences, such as that between APDS and AP get picked up, but apparently lesser ones get by, with devastating effect. For example, few realise the post WW2 Soviet APHE was a huge improvement on earlier ammunition in the same gun. This has led both to wartime performance being overestimated and post war denigrated.
The relatively minor differences between guns pale into insignificance when one realises that the difference made by normal variations in the angle of even a frontal target was in the region of 50%. This was deliberatly exploited by German tankers, who were taught to engage over their front track horn whenever possible. Much attention is given by rule writers to the vertical slope of armour, very little to lateral Slope, and the effect of rounding is avoided likethe plague. Take say a Centurian shot at from 30 degrees to its front. Half the shots striking that flat turret hit the front at 30 degrees, half its thinner side but at 60 degrees. None hit at 0 degrees. Now compare a rounded turret like that of T62. A shot hitting dead centre is at 0 degrees, if a foot away at 15 degrees, if two feet away at 30 degrees, if three at 60 degrees. Not that much difference.
It should not come as a surprise that the theoretical penetration ranges could differ sharply from those reported as normal by fighting units. Take the case of a 17pdr anti-tank gun firing APCBC at a Tiger E. The Tiger's front armour is quoted at 100mm, its side at 80mm, both vertical for all practical purposes. The 17pdr penetrates 100mm at 200 metres at 30 degrees. Yet action reports that "the Tiger's front is at serious risk below 500 metres, and its sides can sometimes be damaged at 1,500 metres." The US 90mm gun penetrated 105mm at 2,600 metres, yet users put its effective range against a Tiger at 700 metres.
Small wonder that I choose to take action reports as my primary source and use penetration trials only as a guide when grouping guns into classes. This does not mean that the technical content of the rules is "inaccurate" as the hide-bound may claim, merely that differences in performance that cannot be substantiated by careful research and that are insignificant compared with random factors have received only the attention they deserve.
The knocking-out of AFV is not necessarilyto be equated with penetration. HE and large APHE often caused significant damage without, in extreme blowinga turretoff, and quite lightweapons could alsosucceed. The best Japanese anti-tank weapon was the 50mm mortar, and several Matilda I's were lost at Arras by 20mm flak igniting their stowage.
A penetrating solid shot usually succeeded, but the lethality of HEAT penetrations was poor. US troops in Korea complained that the WW2 2.36" bazooka could not stop the T34/85. No one asked at the time how it had managed to stop Tigers. The most likely answer is that it stopped both - some of the time. It took an average of seven HEAT penetrations to kill an M.113 in Vietnam, but the pitiful 57mm recoilless once knocked out an M.48. A fluke hit on a hull bottom weld penetrated to cut off the gas pedal, producing a mobility kill!
I now assume three partial vulnerability zones; an outer zone in which only fluke damage occurs, such as to running gear, a jammed turret ring or through the hull roof; a central zone in which penetration depends largely on target angle and chance; and an inner zone in which a glancing hit may just fail to penetrate. A certain kill requires a huge over-match of the armour.
It will be apparent that I now see the tank's behaviour pattern as at least as important as its technical characteristics. Is it motoring carefree along a road, halted at a blockage, dashing into cover, creeping stealthily to its edge, assaulting a position, attacking at speed with brief halts to fire, or dodging from one hull-down firing position behind a ridge to another? Even here, allegedly authorative sources can lead you astray. Are the T64 and T80 really racers like the T72, or are the later suggestions that they are slowerand better armoured correct? Is the M60 really capableof 15 mph across country without bending itself or its crew?
Tables of hit probability seem to take into account target behaviour, but this is largely illusory. A hull-down target is always exactly hull-down and never shrouded by vegetation, and moving targets move continuously at a steady speed on a constant heading across flat ground. in real life, your driver halts late leaving you 3/4 exposed, and the moving target jinks, changes speed or halts to fire, moves through a hollow for partial defilade, and vanishes completely as your sights come on. Vision is reduced by haze, dust, smoke or mirage, and you mistake knocked-out tanks for live ones and the bursts of your friends' shells for enemy missile launch signatures.
But surely tests of hit probability against fully exposed stationary targets supply at least a basis for comparison between ammunition types? Not always. Usually they report first round hit probability. Leaving aside the questions of whether the target appears in an expected place at an expected range, whether it is behind or in front of cover or silhouetted, a big black or white rectangle or a realistic shape, and how close subsequent targets are to the range and angle of the first, not all engagements are a single shot.
Take the case of two tanks firing at each other at 1,800 metres, one with APDS, one with HEP. If both fire one round, the APDS hits half the time, the HEP a third. If both correct and fire a second shot, both will likely hit with it. First round hit probability against artificial targets has very little in common with combat hit probability. If it had, tanks would never need to load up fresh ammo. After all, each has 30 rounds, and the enemy only outnumbers them 3 to 1!
Back to Table of Contents -- Courier Vol. VIII No. 1
To Courier List of Issues
To MagWeb Master Magazine List
© Copyright 1987 by The Courier Publishing Company.
This article appears in MagWeb (Magazine Web) on the Internet World Wide Web.
Other military history articles and gaming articles are available at http://www.magweb.com