Random design lessons from the front: Contrasting views on flank attacks

During Operation Perch, after failing to push the Germans back from Tilly-sur-Seulles 7th Armoured Division attempted a “daring right hook” through a gap round the left flank of the Panzer Lehr Division. The change of direction of the attack took more than 24 hours and was characterised by a lack of knowledge about what was in front and to the flanks during the new attack. Hinde, the brigade commander, issued orders that the attack be made with all speed – this was transformed into “no time for reconnaissance”, so the advance guard of the brigade (A Company, 4th CLY, rather than the recce Stuarts) moved through Villers-Bocage to Point 213 without checking its flanks (in fact, pretty much not checking what was in Villers-Bocage either). During the engagement Hinde appeared at Villers-Bocage, but not Point 213, then went back to brigade HQ. The Divisional commander and Corps commander were nowhere near the action. Owing primarily to slow execution and lack of reconnaissance 22nd Armoured Brigade was ambushed by the Tigers of 101st Heavy Tank Battalion and after a couple of days was withdrawn from Villers-Bocage back more-or-less to its starting positions.

In contrast, Guderian’s narrative of part of his first action in the Polish campaign: “Messages from the 2nd (Motorised) Infantry Division stated that their attack on the Polish wire entanglements had bogged down. All three infantry regiments had made a frontal attack… I ordered that the regiment on the left be withdrawn during the night and moved to the right wing, from where it was to advance next day behind the 3rd Panzer Division and make an encircling movement in the direction of Tuchel… I decided…that I must visit this division the next morning… I placed myself at the head of the regiment… and led it personally as far as the crossing of the Kamionka  to the north of Gross-Klonia [about 15 miles beyond the Polish front]. The 2nd (Motorised) Infantry Division’s attack now began to make rapid progress.”

The contrast for me in these 2 narratives is striking. We have the most experienced British armoured division making an unsuccessful frontal attack, then, as ordered by Corps, changing their action to a flank attack through a known gap, but executing the attack slowly, badly and failing. The idea of the attack is characterised in accounts frequently as “daring”. Senior British commanders seem to have a very “hands off” approach to command. On the other hand, we have a German commander quite naturally and without fuss ordering one of his divisions to carry out a similar flanking manoeuvre, then personally making sure it’s carried out. The German units were all untested in battle at this stage, as was the commander.

3 aspects of this seem relevant and are borne out in some of our historical wargames: (1) Doctrine matters. (2) Reconnaissance matters. (3) Leadership matters.


Random design lessons from the front: troop representation

It’s comparatively easy to put together a vaguely credible way of representing troops at low level for a WW2 wargame. For example, with Airfix Battles we did a 1:1 representation, so each infantry figure or tank model represents 1 infantry man or real tank. As John Salt has pointed out in an earlier comment on this blog, it is not “at all easy to find out how combat really works at the lowest tactical levels”. However, for Airfix Battles, we were aiming at “credible”, not a simulation, and our approach has been well received; there are some heartening comments on Bob Cordery’s blog here: https://wargamingmiscellanybackup.wordpress.com/category/airfix-battles/, and the Airfix Battles Appreciation Group on Facebook gives us a certain seal of approval.

Modelling stuff at a higher level – by which I mean tactical representation, not making and painting figures – has needed more work, especially if I’m trying to capture a bit of the command, control and communications aspects, while ending up with a playable wargame. Taking company level as an example, a primary difficulty is the extent of articulation in a WW2 infantry company. A company might be highly concentrated in one place or spread thin in defence; it might be focused on where to place its mortars and MGs to support a neighbouring unit, or it might be focusing on all-round defence with its rifle components. Some companies might provide components as attachments to other troops, and some might be acting on their own entirely. The platoon and section/squad structure enables these sublties to be implemented. Providing a single answer to this conundrum is problematic.

Some wargame rules get around this by allowing on-the-fly creation of groups. So, you have a “centre” for a specific command function, typically representing an officer, and all or a proportion of troops within a specified command range can be used. I’m not keen on this type of solution, because it gives the player much more flexibility than the commander on the spot would have had. It also concentrates the leadership function on one area, when leadership and the command of sub-components were dispersed via officers and NCOs. Perhaps it’s more playable, but that type of solution loses some of the essence of command and control for me.

Alternatively, you could implement a representation of the internal structure of the company – platoons, and so on. This has the merit of structural accuracy at the expense of greater complexity.


German infantry company deployed to attack

Our solution in Mission Command was to represent “the group” as the lowest sized unit that would be given orders, with a group in the Normandy incarnation of the game being a company or squadron – less flexible Soviets might have battalion groups. Even though our groups have multiple elements – with an element being the smallest separately movable item – the elements don’t model the internal company structure. Rather we’re modelling the combat capabilities of the whole company, and we try to reflect differences in the capabilities of groups from different armies in different periods of the war.


British infantry company deployed in defence

There are some implications for players, as you might imagine. It’s quite OK for a player handling a lot of groups to manage each company as a unit without paying unnecessary attention to the details of each element. This is particularly true with broad brush deployments. On the other hand, if you’re playing a small German kampfgruppe, where the positioning of heavy weapons is vital for defence, then you can and should focus on the individual elements and how they fit with the wider group – especially as you almost certainly haven’t got many of them. And you need enough players in your team to handle the size of your force efficiently.

Most importantly, the Mission Command framework allows us designers to focus our attention on the composition of groups within the scenario we’re designing. It’s quite rare that a force will have all its groups straight out of a standard table of organisation and equipment. Variation by scenario is vital to model that portion of reality we’ve put under the microscope. For example, a German panzergrenadier company may “normally” have 3 coherent elements (full sized elements with small arms, LMGs and panzerfausts), with a supporting HMG element and a 8cm mortar element, plus its transports, but it’s easy to vary this overall capability to a more realistic field strength. A 17SS group in Normandy would have integrated elements (just small arms and LMGs), because they weren’t issued with panzerfausts. For most scenarios a German panzergrenadier group might have only 2 coherent elements, or even only 1 with a separate command element and LMG support element, representing the normal coalescing of the infantry around their most effective weapons.

We have a lot of evidence from our games that this approach discourages micromanagement. Players (well, good players anyway) tend to focus on how the group relates to other groups at battalion level and above. There is also very much less tendency to intermingle companies, because that leads to realistic confusion, and elements that become separated from their group suffer bad morale effects. In addition, I’ve found it’s very easy to represent the particular effects of Normandy bocage terrain – simply, each element in bocage but not in a prepared position is immediately considered separated, with all the communications and morale effects that entails; this models well the sense of isolation and lack of support reported by all troops in the bocage, regardless of their company organisation.

Random design lessons from the front: figure scales

A couple of months before Salute may not be the time for this, but why do wargamers focus so much on how it all looks on the table? We’re as guilty as anyone else at our group in Frome, and it’s the same at the Huntingdonshire Wargamers too. Big miniatures, so the paint job looks good. Big scenery, so that it looks pretty. Notwithstanding that the scale of both is all wrong. 15mm figures with a typical wargame tabletop game are outlandishly large. For tanks, depths of the units are huge, even if the frontage is correct, because that’s how the models have to be. Houses and trees are gargantuan size. For Mission Command, we have a ground scale of 1mm:2m, so narrow roads are 60m wide and our narrow streams are like the Rhine in flood.

I’ve recently decided, on grounds of cost, to switch to 6mm for some of my Mission Command stuff (not Frome, because we’re committed to 15mm there). I’ve been surprised that the problem still exists here. The figures and models are better scaled, but the scenery is still massively oversized. 6mm roads are commonly 2 to 5cm in width. Just doing the maths: a popular brand has the narrowest road (called a “narrow dirt road”) with a width of 2cm plus a further 1cm of verges. The widest is the “medium [sic] metalled road” at 3.5cm plus 1 cm of verges. As 6mm is 1/300 scale, these translate to 6m carriageway for the narrow dirt road and a whopping 10.5m for the medium metalled road. Bearing in mind that modern lane widths are approximately 3.5m to 3.75m for major roads, making 7m to 7.5m for a standard 2-lane highway (an A road in the UK), these scaled versions are 50% to 100% too wide. Probably more in fact, because WW2 roads (and more so in earlier periods) were not as wide as modern highways. Just checking my own reality, the B1040 outside my house (a 2 lane medium metalled road) is less than 6m across – at a pinch this could be represented by a 2cm wide piece. But this is not a narrow dirt road.

Oversized terrain in 15mm. The road is supposed to be a narrow road, but the infantry element has a frontage of 100 metres. Also the men will have trouble getting into that church, which is far too small for these figures, though it is about 150 metres long (Notre Dame is 128m long for comparison).

Other scenery in 6mm is not much better from many manufacturers. One leading company I investigated advertised 6mm scenery, but the size was effectively correct for 15mm figures, not 6mm.

Why is this a design problem? In my view, it heavily distorts the wargamers’ perception of scale when playing the game. There’s a tendency to assume a tank model represents a single tank or a single figure represents a single soldier – even if we know, intellectually, that the model represents more than one thing (unless it’s a skirmish game with 1:1 representation). So, shooting at a tank model might “knock out” the tank; but it may represent more than 1 vehicle, so you haven’t actually KOed all those tanks. Similarly, eliminating an element doesn’t represent causing all those guys to be casualties – some may have been killed, some wounded, maybe some captured, but many will have run off, helped the wounded back to safety, got lost, and so on. In fact, looking at tank losses during large engagements – Goodwood springs to mind, as I’ve been delving extensively into Normandy campaign materials – it’s clear that a tank unit can be rendered entirely combat ineffective without having all its tanks destroyed. When 1 of our tank models in Mission Command, representing say 4 vehicles, is removed, this might mean that 1 tank was burnt out, another was seriously damaged (maybe requiring 3rd line workshop repairs out of theatre), another maybe was repairable within 24 hours, and another was pretty much fine, except the crew bailed out, or it made tracks away from the scene. In a later loss report, these might go down as 1 or 2 losses only, depending on how that army recorded such events.

That road on the tabletop also skews our perception of distance. The position in front of my troops can’t be very far, because this (overly wide) road my guys are on is only a foot or so away from it! But a foot may not be close at all – with Mission Command, a foot on the table represents about 600 metres on real ground, and in Normandy an advance of 600 metres could take 3 hours of intense fighting, or even more in the bocage.

For an appreciation of what our toy soldiers are doing, if we’re reflecting reality, we need to be aware of the distortions of scale that our “pretty modelling” portrays.

Rather better 6mm terrain. That single storey farm is about right against this Panzer IV.

Random design lessons from the front: air stuff in a land battle

If your focus is on the land battle, keep the air stuff simple! But on the other hand, do include it!


In our fictional Russian assault on Pleskau/Pskov in June ’44 (April 2017), a bunch of Sturmoviks attacked the inevitable Tractor Factory defended by the Germans. It was quickly apparent that actually using 8.8cm FlaK as AA guns was effective – but how many times do we see flak used entirely in AT role in wargames? In this situation, fields of fire for AT were restricted, so 88s in AT would have been very vulnerable (high silhouette and suchlike), but very good for AA, and later for counter-mortar fire.

Our AA rules are pretty straightforward. All air attacks come at the start of the turn, and have dive, low, medium or high altitude. 88s fire out to horizontal range of 150cm and all altitudes (not interested in high flying heavies!). As standard in Mission Command, roll d20 to hit, if hit then roll d20 for effect. Any aircraft not damaged or destroyed complete their mission. Ground attack uses the same templates as guns, but oriented portrait rather than landscape. Roll for deviation, which is riskier the higher up the aircraft. It’s area fire, so for anything with majority of base under the template roll a d20 for effect.

For simplicity we don’t differentiate between aircraft models, just fighter, fighter-bomber, dive bomber, medium bomber, heavy bomber (though really medium and heavies don’t show). A whole air attack rarely takes more than 5 minutes to do, but can be quite exciting and certainly adds a realistic tension, especially if the deviation is a bit wild; blue on blue *has* happened.

The only difficulty we’ve had is reconciling the feeling that Typhoons should be effective against tanks with the reality that they weren’t as effective as the pilots reported. We’ve settled on using values at the edge between player expectation and actual stats – bearing in mind that German tankers were often more scared of Typhoons than they needed to be, we’ve factored in the fact that some crews abandoned their tanks when under air attack, even if the tanks themselves survived.

Uncombined arms

Dateline: 1 Feb 2018. A Mission Command scenario to test a strong infantry attack against a (weak?) combined defence.

The scene was an area 75cm x 100cm, so quite small, merely 1.5km x 2km. The scenario was designed to take 2 – 3 hours with 1-2 German players versus 1-2 British players. In this game Pete was Brits, I was Germans.


British attacking from the north (bottom of pic). Mission: push in the German outpost in and around the village, so that the area can be used to assemble troops for a major attack on the main German position to the SW – the large slope in the top right leads to the main German position. The stream is fordable along its entire length, the orange patch is a small hill, and the woods are open to the south, but dense on a rocky outcropping to the north. There’s a sizeable patch of bocage before we reach the village.

It’s a couple of hours before dusk, and the Divisional commander wants this outpost cleared before nightfall. The Brits have an infantry battalion and (off-table) a couple of batteries of 25 pounders. British recce suggests the Germans have only a company, but probably with some limited supports, possibly including AFVs.

fire in the bocage.JPG

Fire in the bocage!

Rather inaccurate British artillery opened the engagement to cover the advance of the troops. It was quickly corrected by Forward Observation Officers and was moved forward to the crest over a couple of turns.


B Company advanced – rightmost 7 elements, with C Company to the left. Each of the 4 companies had 2x integrated (rifle+LMG) elements, a command element with jeep, plus light supports of PIAT, LMG and 2″ mortar. 2 companies have an additional PIAT element attached from Support Company. 3″ mortars are giving support with the artillery from off-map. 2″ mortars of each company are dishing out limited smoke.


D Company advanced on the other side of the stream. Half out of shot is a Sherman with the FOO for the off-map battery of Sextons.


The full battalion en avant! Note the 6 pounders from Support Company deployed in the centre.


On turn 3 German artillery picked on the 6 pounder position, having been easily spotted by Germans on the crest of the bocage ridge before the smoke and barrage intervened. It took a while for the transmission of orders to the battery of Wespes off table, and it was to an extent a lucky shot (1/3 chance of being on target using predicted fire). 1 6 pounder model destroyed, the other moved away.


B Company (nearest) continued its advance protecting the left flank of C Company attacking directly into the bocage, C level with A Company on the right. Things were very murky in the bocage at this point, because the barrage reduced visibility by one state – partially obscuring terrain becomes obscuring, so it was very hard for the attackers to see what was in front.

However, the 3″ mortar fire in front of B Company wasn’t enough to prevent Germans not in the bocage from seeing them coming.


A dug in StuG Zug used opportunity fire on the lead element of B Company, then overran it. Surprised, having taken a few casualties and with only relatively distant PIATs immediately available to deal with the assault guns, B Company reeled back, many of their riflemen being captured.

You can also see at the top of the picture that C Company were taking fire from panzergrenadiers around the ridge line in the bocage. The Germans were suffering greatly from the artillery, so the effect of their fire was keeping British heads down rather than causing casualties.


The aftermath of the overrun was that the British left wing had gone. The remaining 6 pounders – still limbered up from the earlier move away from the Wespe fire – was hastily unlimbered, but (shoot then move!) the StuGs had the initiative and quickly shot them up. The StuG’s orders did not include a lone Zug attacking a battalion, so they disappeared back to their secondary position out of sight.bocage_cleared.JPG

Meanwhile A and C Companies’ fire and the supporting indirect fire had driven the German defenders out of the bocage with considerable losses. D Company were established on the undefended ridge on the far side of the stream, ready to push on towards the village from the north west.

This was the situation after 90 minutes of play and game time (the objective for Mission Command is that real time and game time should be about the same). With the StuGs somewhere around the ridges at the bottom of this picture, further Germans undoubtedly not yet discovered directly defending the village, and only 30 minutes of daylight left, it would be a tall order for the British to clear the village before nightfall. Unfortunately we had run out of time – our Thursday sessions are only 2 hours at the moment. I would have liked to have run the remaining bits, but real life can get in the way!

The purpose of this brief scenario was to investigate the difficulty of attacking a combined arms force without armoured support in the late war period. Although this was not a scientific approach and was only one game, I think it is an example of how a few AFVs in a defensive position can strengthen a numerically weakly held position, if the attackers have no armoured support themselves.

With only towed AT guns, rather than armoured tank destroyers or tanks, it’s difficult to co-ordinate against a potential limited counter-attack, while maintaining a decent pace to the attack. With 20-20 hindsight it might have been better for the British to deploy as follows:

  1. Set up the 6 pounders as 2 batteries, 1 on each flank, in overwatch, so they could deal with any armoured forays from the village, from either ridge or the bocage, then move them up to the slopes on each side when captured.
  2. Put PIATs and LMGs on overwatch during the advance, moving forward by bounds, rather than continuously. Then, if there’s a counter-attack or indeed German op fire, the British have an immediate response.

Having said that, it’s still difficult to co-ordinate, because the movement forward of the AT guns will require time, and that’s very limited in this scenario. The British have enough artillery and mortars to suppress the German infantry and thereby support their own infantry onto the position and through to the village. But the German armour changes the nature of the engagement completely. It’s no longer a classic fire and movement situation, but contains a more complex set of problems coordinating anti-tank weapons against armour as well.

Many thanks to Pete P for accepting the short straw of being the attacker!



Someone mentioned that the previous few blog posts have been a bit long and with no pictures. So here’s some explosions we’re going to be using in Mission Command…


These will be used in the Area Fire templates that we’ll make available for download. We will include Mike and Uncle templates, but don’t expect to use them very often!

I’ve been working on the Reference Cards as well. Example here:


Finally, some pictures to illustrate some aspects of playing.

p11-Inverted Wedge-c

A company of Panzer IVs in “breitkeil” (inverted wedge). Note the space this formation covers, roughly 500m x 500m. This enables the rear elements space to manoeuvre against a threat without the whole company being engaged simultaneously. Each model represents an inverted V formation of (usually) 4 tanks.


British infantry company in defence. It has 2 integrated infantry elements (large elements), a company HQ element (back from ruin) and 3 light supports, 1 LMG element (left), 1 PIAT element (in ruin) and 1 2” mortar element (centre). It occupies a frontage of 50cm (1km). It is entrenched in position, and each element would be connected by field telephone land lines, so all its elements can communicate. The left element (LMG) can give flanking fire to support the main central position. The 2” mortar can support the whole position, or retire if attacked. Ideally it would be supported by a further position to the rear!


This German panzergrenadier company has 3 coherent infantry elements and 2 heavy support elements, 1 HMG and 1 8cm mortar. It physically occupies a frontage of 15cm (300m), but its small arms fire allows it to dominate a further 5cm (100m) each side, while its fire still remains effective out to 15cm (300m). There are many alternative formations, including echeloning elements back from either flank, attacking with 1 element leading, and deploying heavy weapons to either flank.

Credits: Vicki Dalton for the explosions; Neil Ford for the pix.

The burden and rigour of battle – Part 7

Concluding Remarks

One of the good things about wargaming is the lack of real danger. Unfortunately for simulating war, it’s also a bad thing, because danger is a major determinant in how people react. In short, there can be a tendency to hurry in wargames, where caution would be the watchword in real life, because of the danger of hurrying. Reconnaissance and planning save lives, but wargamers (or at least hobby wargamers) want to “get on with it”, even to “get on with the game”, as if reconnaissance and planning aren’t vital parts of the event. Real danger focuses the mind. Direct personal risk of injury or death was present at all levels in World War Two, so I don’t believe that it’s a significant factor per se in the difficulties of battlegroup wargaming compared with higher or lower command levels. However, it does affect playing battlegroup wargames significantly.

I will use my wargaming experience with a tabletop miniatures wargame called Mission Command to illustrate some of my conclusions. Mission Command is a World War 2 simulation wargame that I’ve been designing and playing for over 10 years. It’s a co-design with fellow enthusiast Peter Connew. We design, develop and play scenarios with the Abbeywood Irregulars wargamers in Frome, Somerset, a group of ex-military bods and experienced amateur wargamers (it is an all-male group, unfortunately). As we state in the introduction to the wargame:

“Mission Command attempts to capture the essence of combat command from roughly company level to division level without the bloodshed, fear, death and destruction normally associated with actual warfare. The rules concentrate on helping players to learn more about the effectiveness (or otherwise) of a national army’s way of fighting during the Second World War using tabletop miniatures. The focus is primarily on tactical implementation within an overall operational context; games generally reflect up to a day or two of real combat involving up to a division or two on each side.”

In our Mission Command simulation wargames we often present tabletop situations with no visible enemies, so our players have a lot of experience of not being able to see things to shoot at, or that shoot at them. For this reason, we now have much more realistically cautious players, in planning, reconnaissance and in simulated combat. In place of “I’ve rolled to spot into that piece of terrain, so I know there’s nothing there”, we now have “I haven’t searched physically through that piece of terrain, so there might be something in it.” And in place of “That AT gun shot at my tank, therefore I can quickly knock it out before it gets more shots off”, we now have “that piece of hedgerow might contain an AT gun, so I’d better use smoke or suppressive fire.” However, this does raise the serious practical difficulties I’ve mentioned earlier, and it’s only with the use of information technology – specifically very easy digital photography and printing – that we’ve engineering a relatively slick method of handling this issue in a manual wargame without recourse to poorer proxy methods such as dummy units or rolling dice. Of course, the handling of this aspect is one of the advantages of computer wargaming.

Modelling the complexity of command and control at the battlegroup level is difficult, more difficult than at higher and lower levels of command. This level of command presents a set of complex, interlinked communications problems, so mechanical solutions like command points are tricky – rolling few PIPs on a d6 is a crude reflection of command problems, as is rolling a dice to see if you get artillery support. Sometimes the effect may work, but the impact of randomising away the issue is profound, if part of what we’re trying to do is to learn the nature of the problems. For example, with a randomising mechanic, it may be worth carrying out a “suicidal” attack, hoping that the opponent’s dice will fail; in the wargaming environment no harm done, but also no lessons learned. Somewhat worse, many wargaming systems will “work” using tactics that, history shows, would almost certainly fail if used in reality.

In Mission Command, we attempt to model the constraints on command and communications, by organising forces using realistic information about the command structures of different national armies, by imposing appropriate delays in the transmission of information and new orders, and by reflecting tactical circumstances. But, as our players know from our early play tests, communications systems are hard to model and still have a playable game.

Similarly, modelling the co-ordination of all the multitudinous different weapons systems available to the battlegroup commander is difficult, more difficult than at higher and lower levels of command. This is particularly so, because analysts and military historians are still discussing and revising our understanding of the nature of WW2 tactical combat at this command level.

In Mission Command, we decided that we had to condense or abstract out much of the detail, in order to retain a sense of the battlegroup scale aimed at, but without losing what we considered to be essential elements. For example, we believed it was important to retain relatively fine-grained definition of AT weapons, lest we lose the evidentially certain impact of more advanced weapons, such as the German 7.5cm L70 on the Panther, as compared with the 75mm L48 on the Panzer IV, while we also believed that it was not necessary to include fine detail of the armour on different areas of individual tanks; our armour classes run from 1 (worst) to 10 (best). We don’t include details of whereabouts any individual tank was hit, but we have retained the basic notion that it’s harder to destroy a tank from the front.

Credibility of the model is also important, in a very popular wargaming period where players can be incredibly knowledgeable. A specific problem for example was in relation to the effectiveness or otherwise of air attacks, particularly by rockets. Here we saw a direct clash between what we now know and what was thought at the time. After action operations analysis of rocket attacks, particularly during the destruction of forces in the Falaise Pocket and during the Ardennes counter-attack, demonstrated that, contrary to the claims at the time, a very small percentage of tanks was destroyed by such attacks. However, rocket-firing Typhoons are often a stalwart and highly effective air asset in World War Two wargames, and there is an expectation amongst players that they should be more effective against armour than they actually were. The situation is complicated by the tendency of inexperienced German tank crews late in the war to abandon their tanks in the face of this type of attack. For our Mission Command implementation, we have adjusted and re-adjusted values in our model, until we have a solution that maintains reasonable historical accuracy, but does not render the rocket-firing Typhoon ineffective. This satisfies the players and the designers, but it has meant balancing opposing viewpoints.

When wargaming at battlegroup level, we can present our players with highly complex situations that were very challenging even for the trained, experienced and supported commanders facing them in World War Two. By doing this through good quality wargames, designers can, I believe, provide a means for gaining insights into the nature of battlegroup level warfare in World War Two. These insights can be gained through all the processes of the game – design, research and development, play and post-action analysis. I’ve been struck by the willingness of players and umpires to engage seriously over long periods with these complexities, to try out ideas both historic and less so, providing more material to aid our understanding, often through failure, which is one of the best ways to learn.