Divisions, Designers and Doctrines – A Collection of Unmitigated Pedantry

This is the second part (I) of our long-delayed series looking at Hearts of Iron IV, Paradox Interactive’s World War II grand strategy game and also one of the most popular strategy games currently on the market.

Last week, we looked at how HoI4 simulates combat, particularly its focus on operations (rather than tactics or, as we’ll see in later weeks, strategy). Much of Hearts of Iron‘s success is due to its effectiveness as an operational simulator and its operational model is by far the most developed part of the game.

This week we’re turning to the question of how Hearts of Iron IV handles what is going on below the operational level that the player operates at. Because the player does not themselves handle – or even really see – how their armies function below that division-level operational simulator, the granularity here comes from how Hearts of Iron IV understands ‘doctrine,’ which is to say the practices, guidelines and expectations that militaries use to guide their function in the field. The player is thus notionally given some ability, by choosing doctrines and determining the composition of their divisions, to shape how their units fight in the small scale (or at least how those fights are abstractly simulated). In practice, however, as we’ll see, HoI4‘s approach to doctrine is less developed than its approach to operations and never quite succeeds in really capturing the different ways that armies in this period understood themselves and functioned.

But first, if you like what you are reading here, please share it, as I rely on word of mouth to replenish my manpower pool after poorly considered post topics. If you really like it, you can support me, my academic work and my public writing, including this project on Patreon! I can’t promise to use your support to buy a division worth of tanks, but I also will not promise not to do that. If you want updates whenever a new post appears or want to hear my more bite-sized musings on history, security affairs and current events, you can follow me on Bluesky (@bretdevereaux.bsky.social). I am also active on Threads (bretdevereaux) and maintain a de minimis presence on Twitter (@bretdevereaux).

What is Doctrine?

Now military doctrine is a concept we have been meaning to get to here on the blog for quite a while and – at last – we have a good opportunity! Military doctrine, as a concept is one of those famously tricky to define concepts, but also absolutely essential for understanding how a military in any period – but perhaps especially in the Second World War – operates in practice. Parshall and Tully offer what I think is one of the most accessible and useful definitions of doctrine, “At its simplest, doctrine is the body of formal knowledge that tells a fighting force how it is expected to fight…doctrine is bigger than tactic in that doctrine encompasses not only the means for actually handling forces in battle, but also augments it with command structure and communications procedures.”

Doctrine is thus the term for the ‘standard operating procedures,’ as it were, of an army or indeed of an entire military.

The reason doctrine is essential is because, as Clausewitz says, everything is very simple in war, but the simplest thing is difficult (drink!). The concept of a coordinated attack on an enemy position is very simple. The execution of a coordinated attack on an enemy position is very hard. After all, a real army does not actually consist of a few dozen units that, as in a video game, respond more or less instantly to the precise, clear and unambiguous commands of a single leader. Instead, a real army consists of many thousands (or hundreds of thousands or even millions) of human beings, each of whom is in their own experiential little universe – unable to read their minds and often unaware of the positions of their fellows.

If you want a large group of people to execute even the simplest of commands effectively, they need to already have some common conception of what the orders mean and how they carry them out. If you want two units, with different commanders, to coordinate effectively – often when they cannot be in direct communication – those commanders (and their soldiers) need to share some common conception of the normal or correct way to attack or defend. If I am commanding, say, a division’s artillery support (probably split into half a dozen or more batteries) and I am supposed to support a brigade’s attack in the morning, I need to have the same conception of what an attack means as the brigade commander. On the battlefield, armies cannot ‘make it up as they go along,’ they need to function according to standard procedures, a shared understanding of what an attack looks like, what a defensive line looks like, how to call out information on a radio, what orders mean and so on.

That shared conception is at the core of the concept of doctrine.

By the Second World War, this sort of doctrine had been largely formalized in nearly all of the militaries of the major participants. That body of formal knowledge consisted of training materials and best practices often laid out in an extensive catalog of field manuals, complete with their own bespoke jargon. As the war progressed, changes to doctrine could be promulgated the same way, with After Action Reports and ‘Lesson’s Learned’ summaries flowing ideally up the chain of command and new doctrine publications, updated field manuals and such flowing down the chain of command.

That said, as a pre-modernist, I have to note that while formal military doctrine – written out in field manuals – is a relatively modern invention, informal military doctrine is as old as warfare itself. Pre-modern armies also had ‘standard operating procedures,’ – often every bit as rigid as modern doctrine – which were learned informally, with older soldiers teaching newer ones (and young officers practically ‘apprenticing’ to more senior ones).

In short doctrine is necessary for armies (and navies, and air forces) to function. While it can be constraining, a ‘straitjacket’ that imposes standard approaches (sometimes to non-standard problems) there is simply no other way of coordination battles that might involve tens or hundreds of thousands of individuals who need to act in concert in order to be successful.

In terms of what doctrine includes, we run the real risk that because almost everything a military does (or has) interacts with doctrine, there’s a sense in which everything is doctrine and so nothing seems like doctrine in a kind of ‘the fish doesn’t know it is wet’ situation. Instead, while it is the case that doctrine tends to be evolutionary (smaller changes, iteration on iteration) rather than being designed ‘from scratch’ from first principles, it is helpful to think about building a doctrine from first principles to get a sense of what it entails.

Beginning at the operational level (we’ll put strategy aside for now), doctrine begins with a theory of victory, a sense of how an army defeats an opposing force. That theory of victory could be, for instance, attritional in nature (“we win by inflicting more casualties on the enemy until their force collapses”) or it could be based in maneuver (“we win by breaking through enemy lines and then maneuvering to cut off enemy forces or disrupt their logistics.”) Heck, in the case of some kinds of militaries waging protected war (or engaged in non-violence or terrorism), the theory of victory may be entirely abstract, focused on public opinion or popular will. Naturally armies with different theories of victory are going to construct very different doctrines for achieving those goals.

For instance, an army that has a theory of victory based on maneuver is going to need to concentrate a lot of force and coordinate it to create that breakthrough at a specific spot in the enemy line and then the ability to go fast once that breach is made, which is going to demand different command structures and procedures than an army that is focused on simply inflicting attrition. In particular, the maneuver army might want to push a lot of command authority down to lower ranks so that they can go fast in situations where communications with high command are difficult and conditions on the ground are rapidly changing. But a different army might want to pull command authority up to enable very large, well-coordinated, intricately planned ‘set piece’ battles, perhaps in order to try to shift the attritional balance in their favor (or achieve a very choreographed advance).

That theory of victory in turn is going to shape the conduct of a battle. To take our ‘maneuver’ theory of victory army, the next question becomes “what is the force structure of the sharp forward edge of your breakthrough force?” Do you want your tanks distributed widely or concentrated into a handful of units? What is the role of infantry in this attack? If the infantry and the tanks are expected to coordinate closely, they probably need to be placed together into the same divisions so that they can train together and be commanded together. What about the artillery? Do you want each small unit to have a little bit of artillery of its own, or a really big ‘grand battery’ at the disposal of the division commander, which can be unleashed in a coordinated manner, perhaps focused on a single objective in a larger battle? Doctrine is going to play into the structure of your army: what capabilities, equipment and troops does a given division (or corps or brigade or army group) have? What is its command structure? Which elements are subordinate?

Those concerns then flow downward into the fighting procedures for individual smaller units. When the divisional commander orders his armored brigade to ‘take that hill’ – what does that look like? Do the tanks have a formation to advance in? Does the infantry expect to go first in front of the tanks or to follow in their wake (or is the infantry not in the same command chain as the tanks at all and so will just struggle to mutually support each other – a persistent problem for the British Army for much of WWII). How does air power and artillery factor in here? What level of ‘fires’ support does the company commander advancing his company expect to get – and how does he expect to get it? Doctrine gets all the way down to “how does the company commander’s signal officer call for an artillery strike?” because in the chaos of battle you need standard ways to indicate a specific location and call a specific strike (grid squares, usually).

And again, the ‘best approach’ here may differ if the theory of victory is different! We might ask for instance, during a large offensive, what is the best use of your close air support aircraft? Are they best employed to focus their attacks on the point of breakthrough, contributing to the massed firepower at the decisive point? Or perhaps it is a better idea to have them striking deeper along an enemy’s operational depth, hitting reserves, reinforcements on the move, supply depots and rear echelon units that might be softer targets? The balance between those missions (with implications for what the guys on the ground expect from their aircraft) is going to depend on what you are trying to achieve (that theory of victory) but also how you are trying to achieve it.

(As you can see, doctrine for land troops, navies and air forces are going to interact here too: these elements often need to be mutually supporting, so they need doctrines that link up effectively. Shore bombardment support gets a lot more effective if the troops ashore have a way to signal rapidly to the ships where they need the big guns.)

Mixed in with these choices are also just a ton of ‘best practices.’ Often the ‘best’ way to do something is not obvious or cannot be reasoned out easily from first principles but emerges slowly with experience. Something like, for instance, the ‘combat box‘ bomber formations, which enabled greater coverage and fields of fire from heavy bombers’ defensive guns (thus limiting – though by no means removing – the damage enemy fighters could do) had to be developed and refined based on experience from the field. Already mentioned but shore bombardment doctrine in the United States Navy evolved a lot during the war, particularly the somewhat counter-intuitive realization that often cruisers and destroyers provided the most effective shore support because their lighter (though still massive by land-warfare standards) guns could be quick, accurate and responsive, whereas the Big Guns on battleships were harder to fire ‘danger close’ and slower to get effectively on the target.

But what I want to stress is just how expansive doctrine ends up being. It isn’t just a simple tactical field manual – for one, by World War II, tactics are so complex that the ‘simple tactical field manual’ for a combined arms division with motorized, mechanized and armor elements supported by artillery and anti-aircraft along with maintenance and logistics elements would be enormous (in practice that’s dozens and dozens of field manuals). But equally command structures, communications practices, unit structures (what elements go in what division at what level), operational concepts and training procedures all fit into doctrine too.

And all of that matters because it gets to how two armies with broadly similar equipment can fight very differently, because they understand what they’re doing with that equipment differently, because they have different practices for its employment and different training to set up those practices. So doctrine is really important.

The Doctrine Tree

Sussing out how Hearts of Iron IV handles formal doctrine is actually really easy: it is a specific, discrete game mechanic called ‘doctrine,’ which is then split between land doctrine, air doctrine and naval doctrine.

That said, I think it is fairly clear that this is a set of mechanics the Hearts of Iron IV team has never been fully satisfied with. You can tell because the doctrine system has undergone more changes than most core Hearts of Iron IV systems, getting two major overhauls and a ton of minor tweaks. On release, doctrine was part of the research system and so doctrines were basically technologies which thus competed for research slots with every other kind of technology. However with the 1.11 patch (in 2021), doctrine was moved out of the research tree into its own space and used ‘army/air/navy experience’ as its currency instead of research time. Then with the 1.17 patch (in 2025) the system was massively overhauled again to drop much of the ‘tree’ structure entirely.

I want to talk about both the system as it existed post 1.11 and post 1.17, because I think both doctrine systems help understand what the developers are trying to do and also expose some weaknesses with the end result. We’re going to focus here in particular on Land Doctrine, but in each case the Air and Naval trees work basically the same way.

Let’s start with the ‘old’ doctrine tree:

Basically the structure here was a series of doctrine paths. Players first had to choose a tree (you could only develop one of the four doctrine categories) and then spent army experience (or air experience or navy experience for those trees) to ‘buy’ each progressive notch on the tree (originally this was research using technology). At a couple of points in each tree, it branched and the player chose one of the two branches (losing access to the other branch). Each stop on the tree confers a set of statistical benefits, making your divisions (or air wings, or ships) slightly better in a given way. The currency for this is that army/air/navy experience, which can be gained in peacetime slowly through having theorists in your high command or by doing military exercises, but is gained rapidly during a war simply by fighting, representing the battlefield’s role as a harsh teacher.

And you can see what this system is trying to do. Each of the four major ‘trees’ is meant to represent the military doctrine of one or more of the major powers. The Mobile Warfare Doctrine tree is meant to reflect German doctrine through the war (generally called by its German name, bewegungskrieg, ‘maneuver warfare’), with split branches favoring offense or defense meant to reflect the potential for Germany to get in over its head and have to shift to a more defensive approach to warfare (as it did, historically). The Superior Firepower Doctrine tree is meant to reflect the American approach to the war, with split branches that honestly don’t make a whole lot of sense and mostly gesture towards late- or post- Cold War doctrinal revisions (AirLand Battle was a 1980s US military doctrine and a significant depature from what came before while the phrase ‘Shock and Awe’ is most often associated with the 2003 Invasion of Iraq). The Grand Battleplan doctrine is meant to reflect countries that stuck closer to WWI tactical approaches (France, Britain, Japan), with one line for increasingly artillery and armor heavy British doctrine and another for Japanese infantry infiltration tactics. The formal doctrine it most clearly evokes is French Methodical Battle (La Bataille Conduite). Finally, the ‘Mass Assault’ doctrine was supposed to encompass both Chinese and Soviet approaches and that’s just…a real problem for reasons we’ll get into in a moment.

The update changes this system, recognizing that it was way to rigid. So the new post-1.17 system replaced the single doctrine tree the player must choose with a ‘grand doctrine’ and then supporting modular doctrines for each combat ‘arm.’ Thus a country’s army doctrine becomes the sum of its ‘grand doctrine’ (representing overall operational concept), plus a choice of one of several infantry, support, armor and operations sub-doctrines. That said the main four ‘grand doctrines’ retain the names and framing of the doctrine trees.

A country picks a doctrine for each of these categories (spending army experience as before) and then as that unit-type (or any unit from the grand doctrine) fights, they accumulate ‘mastery’ which unlocks progressive tiers of bonuses. The idea is that you spend army experience to pick the kind of warfare you want to do and then get better at it over time as the approach is practiced and refined.

There’s clearly a second intention to this system which is to enable players to try to recreate historical doctrinal systems. That is very evidence in the original ‘four trees’ system, where each tree (or branch) is clearly themed after the doctrine of a specific combatant. But it remains true in the post-1.17 modular system. Players are going to quickly pick up on, for instance, in the ‘armor’ subdoctrine, that ‘Armored Spearhead’ shows an American tank (a Sherman), while ‘Armored Infantry Support’ shows a German tank (a Tiger) and ‘Streamlined Deployment’ shows three Soviet tanks (T-34s, I think, though so small and in profile they could also be IS-2s).

And on the one hand, I think this is a noble attempt to actually model something fairly complex and on the other hand I think quite a few problems remain.

The first problem, somewhat reduced but not removed by the new system is simply that the ‘grand doctrines’ don’t necessarily map on well to what major powers were doing or at least thought they were doing. Both the United States and the Soviet Union understood themselves to be attempting to do a kind of maneuver warfare, the USA attempting to reverse-engineer German maneuver warfare doctrine, whereas the USSR came into the war with its own maneuver warfare doctrine (Deep Operations), which it needed to refine but also actually train and implement because Stalin had purged most of the officers who knew and developed it. It is certainly the case that the United States pursued a firepower-based solution to generating maneuver, enabled by the USA’s much larger industrial base, but the goal was still breakthroughs and maneuver, not merely blowing things up for blowing things’ up sake.

Meanwhile, placing the USSR under ‘Mass Assault’ is honestly borderline insulting given that the USSR would finish WWII with what is generally now regarded as the most sophisticated maneuver warfare doctrine around. While Soviet armies often took huge casualties and Soviet commanders – especially early in the war – often didn’t seem to understand their own doctrine and engaged in under-prepared, under-resourced infantry attacks (not infrequently because Stalin demanded counter-offensives before they were ready), the USSR was not doing ‘human wave’ attacks a la Enemy at the Gates (2001) and indeed by the time we are at Operation Bagration in 1944, it is very easy to argue that the Soviets are better at maneuver warfare than the Germans (indeed one could potentially try to make that argument as early as Operation Uranus in late 1942).

Likewise, ‘Grand Battleplan’ as the umbrella for France, Britain and Japan is equally odd. The British Army struggled a fair bit with infantry-and-armor coordination through the war and generally preferred carefully planned ‘set piece’ battles, with a pretty clear legacy reaching back to the lessons of the First World War (alongside the French Methodical Battle approach), but the Imperial Japanese Army was devoted to the cult of the offensive and favored quick marches, surprise and tenacity in offense, rather than a slow operational tempo characterized by intensive planning. If I were to place the IJA anywhere on the doctrine tree, it would be as a kind of alternate branch of the Mobile Warfare tree: Japan was trying to do speed, infiltration and maneuver, just with infantry (for the most part) rather than with armor (for the most part). The limitations of a ‘light and fast’ approach become pretty shockingly clear as one gets into the back half of 1942 and the IJA starts tangling with well-supplied Americans and Australians who have a lot of firepower at their disposal. In short, the Imperial Japanese Navy may have favored baroque, overly complex and choreographed battleplans, but the IJA was very much a ‘cult of the offensive’ organization that prioritized speed and surprise – albeit with a lot less trucks and tanks than what European armies were using.

So in the notional layout, these doctrines are often a weak match for the historical countries they are supposed to represent. The bonuses also often do not line up clearly either. Admittedly, the new doctrine system is significantly better in this regard. One of the classic balancing goof-ups of the old system was the ‘Mass Mobilization’ branch the Mass Assault Doctrine, which offered bonuses to manpower and reduced frontage for infantry brigades (enabling a country to get more infantry in the fight on a given section of front), which was meant to represent Nationalist China and (to an extent) the USSR reaching deep to mobilize enormous amounts of manpower – a ‘muscle not metal’ approach – but actually was basically useless for those countries (they have plenty of manpower anyway) and extremely useful for small highly industrialized countries which desperately needed the huge boost to recruitable manpower.

The new tree avoids these kinds of errors, but in playing I have generally found the connections between doctrine choices and actual operational concept to be fairly weak. Part of the problem is that doctrines are presented as a confusing collection of tiered bonuses to a bewildering array of things. Which is better for building armored divisions arming for decisive breakthroughs: +5% breakthrough, +10% soft attack and a flat +5 organization for medium and heavy tanks or +10% breakthrough, +2 organization and +0.05 recovery rate? I have six hundred hours in this game and I would still have to consult a guide to figure out which was the better option for that specific concept.

Meanwhile, balance remains a problem. Ideally, this would be a case where different countries would find that different doctrines supported their own mix of industry, terrain and strategic needs. In practice, each balancing patch tends to just alter the array of ‘meta’ choices with entire grand doctrines almost invariably assigned to the scrap heap of ‘never pick.‘ This is a problem that, as we’ll see, is intensified greatly by the division-designer. From my perspective, this system could actually do to be a lot simpler, with the player being tasked to think about their intended operational concept and then guided into choices that fit that concept, rather than having to confront a bewildering array of stat bonuses which often don’t do what they are supposed to historically model in any event (and do almost nothing to really tell the player what they are supposed to do).

The final issue here is very strong path dependence in doctrine choices. Historically, armies frequently tried things, realized they didn’t work and then tried new things. American amphibious operations doctrine – which by 1944 was an absolute marvel of efficiency in managing enormous and extremely complex operations – largely emerged by getting it wrong a whole bunch. The marine landings at Gaudalcanal in August of 1942, for instance, failed to load supplies so that they would be unloaded in the order of the most essential to the least, leaving the marines in the lurch when Japan contested the waters around the island, stopping unloading prematurely. That problem was fixed for the Operation Torch landings in North Africa in November of 1942, but the major takeaway of allied commanders from the Torch landings is that if they had been facing truly fierce, determined resistance, they probably would have failed. In short, no one set out in 1939 to plan D-Day: the doctrine for Operationa Overlord emerged out of a process of trial and error.

And while the current doctrine system allows for lessons learned and experience gaining through the ‘mastery’ system (which ‘levels up’ existing doctrines to unlock new bonuses), it heavily penalizes shifting doctrine in response to changing conditions. A player that goes in with an ‘Armored Spearhead’ oops-all-tanks armored division concept only to have to realize they need to back up to a more combined arms ‘Armored Infantry Support’ doctrine – which is exactly the experience Britain and the United States had with their armored doctrine! – is basically forced to reset to zero as if they hadn’t learned anything at all. The player is aggressively penalized for trying an operational concept, realizing it doesn’t work and trying to shift tracks (because it resets their bonuses and costs more currency), but this is exactly what the most successful armies in World War II did.

However none of these issues are, in my view, the biggest issue for Hearts of Iron IV‘s concept of doctrine. The biggest issue for HoI4‘s doctrine is that the most important – for the game itself, at least – parts of doctrine do not take place in the doctrine system.

The All-Powerful Division Designer

As noted in the first part of this series, the basic unit of Hearts of Iron IV‘s combat is the division. Now, as we’ll see, divisions can vary a lot in size – indeed, divisions can be as small as a single brigade or practically as large as an entire corps. But the game treats all of these units as divisions, which cannot either be split or merged.

This is actually a significant change from previous entries in the series. In Hearts of Iron III, countries actually produced brigades (each of a single type) which were then combined into a division (which could have 2-5 brigades in it). A player could thus pull a brigade out of a division and attach it to another, or cobble together a division from available spare brigades, or produce new brigades and attach them to existing divisions and so on.

By contrast divisions in Hearts of Iron IV are produced to templates. Each division has a template which determines what its components will be and it will automatically attempt to pull the necessary manpower and equipment to bring those components to full strength. In essence, the actual ‘brigades’ in the division structure are really mostly notional: a division is actually a container for battalions (the sub-units of brigades in the division-designer). In addition to the core ‘line’ battalions (which determine a division’s combat width – the amount of frontage it takes up), a division also can have support battalions, which boost its capability (and raise its equipment and manpower cost) without adding frontage, but it is limited to five of them.

Divisions go into training as whole units (the entire division is trained together before deployment), but once deployed, they remain ‘attached’ to their template. If the player changes a division’s template, that instantly changes the expected composition of every division using that template. All of those divisions will immediately request or return the necessary equipment and manpower from (or to) the national stockpile to reinforce up to their new expected components. In practice, because being under-strength for a given composition can impose pretty significant penalties, it is a wiser idea to copy an existing template, upgrade it, and then swap divisions over to the new template when they’re out of combat.

Now this needn’t necessarily be a problem if you could have infinite division templates, but you generally can’t. Modifying a division template requires army experience, the same currency used to buy doctrines (and also to modify vehicle templates – so designing a new tank or creating an upgraded version of an old one burns this currency too). In practice, of the three ‘experience’ currencies, army experience remains precious the longest in the game, because it is used for the most things. That means that unnecessary division template changes burn a real and limited resource.

That will matter to a different degree to different countries. France and Britain, in particular, tend to be ‘starved’ for army experience prior to the start of the war where they then suddenly have to face Germany. You can see the resulting incentives play out in the UK run I have been showing in the screenshots: my army experience was very carefully ‘budgeted’ until well into 1942. I needed enough army experience to have two effective infantry division templates (one for UK infantry, one for units recruited from India) ready in 1939, plus enough to create templates for one armored and one motorized division, to design the tanks and self-propelled artillery to go in those divisions and finally to buy all of my sub-doctrines. That leaves functionally no space for specialized division templates.

Now what the AI does is it ‘versions’ its division templates (because it costs nothing to copy an existing template you already have), causing the AI to end up with a dozen or more templates, which are all incremental changes on each other. But my sense is most players do what I do: they maintain a handful of division templates which they work hard to get to a ‘final form’ before deployment and then subsequently make on minimal changes to.

The result are, frankly, ‘cookie-cutter armies.’ In that UK run, by the summer of 1944, my army consisted of 305 divisions (roughly 3 million men) using just seven templates (two of which are actually identical ‘colonial’ templates to pull manpower from India and Malaya; they stay unique even after those countries are absorbed, so it is really five templates). My main force in Europe thus consists of 130 exactly identical infantry divisions, backed by 13 exactly identical mechanized infantry divisions and 13 exactly identical armored divisions.

This is, to put it lightly, not how actual divisions generally worked. Of course divisions might be of different sizes and strengths due to combat losses (something HoI4 considers), but while most countries did have a standard ‘template’ sense of what an infantry division or an armored division should look like, that was no guarantee at all that every such division did look like that. The division template, for instance, treats light, medium and heavy tanks as no-substitution battalion types, but actual Panzer Divisions varied quite widely in how many light, medium or heavy tanks they had. Pretty much everyone threw divisions forward that were effectively missing elements, either because those elements hadn’t been stood up yet or were still in transit or had been detached for some other job. Hearts of Iron IV has no way to represent that: you either have all of a division or you have nothing. A division can be damaged, but it cannot be incomplete and it can never be ‘non-standard.’

The larger problem is the degree to which the division designer – the screen where you decide what battalions go into your division templates – largely usurps the role of doctrine. In the Second World War, having the right doctrine and training to get tanks and infantry to cooperate together effectively what substantially more important than the simple question of deciding to have armor divisions – command structures, communications, tactics all mattered. But the lion’s share of the work in Hearts of Iron IV is simply accomplished by placing motorized infantry and armor in the correct quantities into the division designer in a single division. The job of learning how to coordinate armor and infantry is now mostly complete; having the right doctrine provides a noticeable but decidedly smaller bonus to that ‘correct’ division template’s performance. But in fact, as was repeatedly demonstrated in World War II, having the ‘right’ division structure was not enough to produce victory, not without soldiers and officers who understood how to use that structure.

The importance of the division designer is, of course, most visible when it ‘breaks.’ For the first few years after release, for instance, due to the way frontage was calculated, divisions at exactly 10, 20 and 40 width functioned much better than any other configuration. That has long since been patched, but the example (especially in single-player) that remains is what the community refers to as ‘space marines:’ a leg infantry division with a single armored battalion in it (typically heavy armor, because who cares if it goes fast). Because battalions do not have stats, divisions do, the one battalion of armor has its armor value averaged across the entire division, which can push it just enough enough to get over the armor-piercing break-points for regular infantry or infantry with just support anti-tank, which triggers a massive reduction in the damage that division takes. Thus the label ‘space marines’ because the division fights as if its infantry were wearing powered armor. ‘Space marines’ can thus easily outmatch basically any division the AI will produce, although in multiplayer, players are well aware of this trick and can counter it.

The irony is, of course, that this idea – distributing heavier tanks across the entire infantry force more or less evenly – was tried in World War II, but it was famously ineffective. This was the French and British approach to armor, the ‘Infantry Tank’ and the actual observed effect was that there were never enough tanks at the critical point when they were ‘sprinkled’ like this.

As a game design element, the division designer gives the player so many buttons and knobs to twist and turn that it is almost inevitable that players are going to min-max divisions out of the system that grossly outperform the more-or-less historical division templates the AI has to work with. On the flipside, the division designer has some trouble really representing historical differences in unit composition. If you want an ‘artillery heavy’ army, for instance, you have to add ‘line artillery’ (artillery used as a main ‘line’ battalion rather than a support company) as a co-equal element with the infantry brigades – you cannot just expand the ‘support artillery,’ but in practice American and British divisions remained ‘triangular’ (three infantry brigades) in structure, with division artillery (‘div arty’) simply expanded to include more batteries. The line/support distinction in the division designer doesn’t seem to have a real world parallel (unless there is one in the Red Army? I am admittedly less knowledgeable about the Soviet military).

Meanwhile the system is again incapable of representing Soviet practice well. While Soviet divisions might notionally be close to the same size as German, British or American divisions, in practice they tended to be substantially smaller. Importantly, a lot of the heavier supporting elements (like artillery) were concentrated not at the division level but at the corps level, with the result that a Soviet corps almost functioned like a jumbo-sized western division. You can sort-of-kind-of simulate this in Hearts of Iron IV by using relatively large division templates for the USSR and simply pretending they represent corps instead of divisions, but HoI4 simply does not have a level of organization for a corps.

And while division-size may seem relatively minor – what is the difference between two twenty-width divisions vs. one forty-width division – the way that Hearts of Iron IV calculates damage reduction via defense (when defending) or breakthrough (when attacking) has typically meant that oversized divisions are better, particularly for attacking. So whereas in the war itself, the Germany, Britain the USA and the USSR all settled on a similar size for an armored division (or for the Soviets, a tank corps), with around 200 tanks and 8,000-10,000 personnel, the current HoI4 ideal tank division is a monster with nearly 20,000 personnel and around 450 tanks. There’s a reason no one in the actual war made their divisions regularly that large.

The end result is that a lot of what in the real world we would term doctrine is expressed through the division designer, but that frankly the division designer feels like one of the ‘gamiest’ elements of Hearts of Iron IV that doesn’t do a great job of expressing differences in doctrinal approaches between countries (‘meta’ division templates are often used quite regardless of the country or the doctrine) or in modeling the divisions that, by war’s end, most of the combatants had settled on as ideal.

No Obligation to Follow Doctrine

A serious problem in planning against American doctrine is that the Americans do not read their manuals, nor do they feel any obligation to follow their doctrine

(An apocryphal quote generally attributed to an unnamed Soviet observer during the Cold War)

On the one hand, Hearts of Iron IV makes a real effort to try to model the way different countries approached doctrine. I’ve focused here on land warfare doctrine, which really at the time mostly meant maneuver warfare doctrine, as for the major combatants in Europe the defining question was how ought the tank be used – but there is a matching set of questions for air and naval doctrine.

On the other hand, I don’t think this model succeeds particularly well.

By the war’s end, the maneuver warfare doctrines apparent in Europe really were visibly different. German maneuver warfare doctrine emphasized devolving command authority downward to create rapid moving, agile forces (a concept that regularly goes by its German name, Auftragstaktik, “mission-tactics”), which aimed to achieve breakthrough by focusing all of its firepower on a key point in the enemy front line. Soviet maneuver warfare doctrine (Deep Operations), instead pulled command authority up, focusing support materiel up at the corps and even army level. It also aimed to produce mobile breakthrough, but by engaging the enemy across their operational depth, using infantry, armor and air power to simultaneously hit front lines, reserves and rear areas to induce a broader frontal collapse. American doctrine was, in some ways less developed, but – motivated by the limits of how many divisions the United States could realistically deploy in Europe – focused on packing as much firepower as possible into the smallest frontage and divisions possible, often using absolutely smothering quantities of airpower and artillery.

And Hearts of Iron IV doesn’t really draw that out. Infantry divisions with line artillery isn’t a cute trick for just Superior Firepower doctrine armies to pack a ton of firepower into a small unit, but a fairly standard template choice you’ll find used by players running a lot of countries. Deep Operations doctrine doesn’t actually provide a way for players to strike across the operational depth of an enemy: damage accrues to enemy divisions reinforced into the front lines regardless and there isn’t any simulation of collapsing rear areas. Even the basic question of ‘should the infantry or the tanks go first in an assault’ (‘armored spearhead’ (USA, Germany) vs. ‘armor as exploitation force’ (USSR)) isn’t really modeled in terms of which divisions take damage.

Meanwhile at the level most visible to the player – at the operational level – the player is under no obligation to follow their own doctrine. Playing as the United Kingdom on the Grand Battleplan doctrine which should mean a strong preference for exhaustively planned set-piece battles, I find often launch far quicker, less planned operations which ought to be something out of the Mobile Warfare doctrine tree. Sure, the bonuses from Grand Battleplan make my planning cap somewhat higher, but the difference isn’t large enough to offset other momentary advantages.

The result here is a valiant attempt to model the way doctrine shapes how militaries conduct themselves in war that feels like it falls short, in part because it is too afraid of actually constraining the player, to force them to actually follow their doctrine. In particular, the game is afraid of ever having units simply function badly because their doctrine is poor or because they’re being used in a way not suited to it, but of course that absolutely was the experience of various countries which tried to use armor without the necessary coordination and communications doctrines. As a result, the doctrine trees end up mostly as a set of bonuses which may or may not be optimal rather than something more closely approaching playstyles.