Topic: Empire of the Sun
I have written on this topic before in my c3i Clio’s corner, but the conjunction of an interesting thread I found on line and the release of the 2nd Edition of Empire of the Sun moved me to want to write up some of my thoughts on ‘historical realism’ in my Empire of the Sun design. As a disclaimer I believe that it is a truism of the Internet that facts and data never change anyone’s opinion once publically stated. I write this for my own purposes as an archive of my thoughts and for the few people who read this blog. I also do not want to embarrass anyone or make these comments appear personal, so I will not state the source of various comments other than to say that I post them without edits exactly the way they were posted. They are there not to start an argument, but as a source for discussion.
First off for those who do not know me here are some of my bona fide’s as an historian. I have an undergraduate degree in History and a Masters in National Security Policy from Georgetown University that had a strong emphasis on Political-Military history. Beyond that I have taught Military Strategy and Policy for the Naval War College and Georgetown University. As an historical simulation designer with well over fifty published designs (see my Bibliography on this site), I have done decades of original historical research that I have published in books, in games, and on this website. As a gamer, historical verisimilitude is extremely important to me in any design that I play often and I play Empire of the Sun continuously. My point is if I did not think Empire of the Sun was an accurate simulation of the Pacific War I would have changed it before publication and certainly would have done something about it over the last 10 years. Truth be told, EotS meets all of my personal standards for historical accuracy that I contend are a high bar.
Empire of the Sun (hereafter EotS) is not a beginners game, so it automatically plays to a small segment of the wargame market. That said, the biggest hurtle to playing EotS is it is not a game that is easy to play well. What is the point of playing a game often if skill does not matter? As most gamers these days do not play a game more than a couple of times it can appear inaccessible unless some desire to play regularly is engendered by ones early experience with the design. This is an important point as some of the comments that follow fall into two buckets. Comments attributed to bad history but are really poor player skill with this design and those that directly challenge the game’s history based on incorrect statements of historical fact. The 2nd edition of EotS comes with a new solitaire system that I believe will act as an interactive tutorial on how to play well. Hopefully this will lower the barrier for early enjoyment of this design.
To be fair some of the commenters while tough on the design said very flattering things about me, so I know that their views are not personal and I do not take them that way, so no acrimony should be implied from my commentary.
I have attempted to integrate comments from a long thread into a shorter coherent set of comments that capture the intent, although I did not edit any of the words to include various misspellings or sentence structure, I just co-located pieces of the same point from different commentators into one place.
Comments are in italics, my response is in non-italic type.
A Cylon who attempted to learn about the Pacific War, by watching game of EOTS would learn:
- The Japanese were very careful to avoid taking more Allied objectives than absolutely necessary. This would lead to a high chance of Allied war weariness.
This first point is not an historical one as much as it is a lack of experience with the playing the game. This comment came out soon after release and continues to be discussed as if it’s a fact. My reply is based on over a decade of public online gaming and tournament play at WBC where this has been tried. This strategy with even a modicum of skill always fails. In fact the Japanese lose the game quicker. So, this comment is not an historical comment as much as it’s a lack of game knowledge comment.
The interesting historical point is the conquest of the Solomon’s and New Guinea beyond Rabaul and Lae were based on a series of Army-Navy conferences that began on March 7th and continued through June of 1942. This outer perimeter defense concept (invasion of New Hebrides, Fiji, and Samoa) was a compromise after the Army blocked the IJNs plans for the conquest of the Hawaiian Islands and occupation of Northern Australia. The main point is my research for a game has to go beyond what is written in a narrative history, but has to captured what are plausible ‘what ifs’. A ‘what if’ for me are those options that were historically discussed and not executed. The IJNs defeat at Midway caused the cancellation of the offensive into Fiji and Samoa. So, not extending the Japanese perimeter beyond Rabula was a path discussed but not chosen.
- The operational tempo of the Japanese fleet had little to do with their own fuel supplies or maintenance resources, but was instead determined by the Allied fuel and logistical capabilties.
EotS is not my first Pacific rodeo. Some may recall my earlier Pacific War game where I also did extensive fuel and operations tempo analysis. That experience gave me a rich body of data and experience upon which to draw from when I designed EotS. What may not be well known is the first thing I did when I began designing this game was to play my third Pacific War campaign game (the first two occurred when I did the original). Based on that play through I reviewed all of my original research and supplemented it with additional data based on new books and material that had been published over 15 years, not yet available in the early 1980’s.
The Japanese fought six major offensives with high operations tempo in the Bismarck barrier throughout 1942 and 1943. In 1944 the Japanese were prepared to throw the fleet at any USN attack on their inner perimeter. They had developed numerous options, but the key-planning feature was a shortage of carrier air not fuel. This resulted in two major sorties (Philippine Sea and Leyte Gulf) that resulted in the destruction of the IJN. I also discovered through research (Prados) that there were several fleet sorties, to include the BB Yamato that did not lead to battle. In these situation the fleet put to sea but it was too far away from the invasion to intervene, hence the operation was cancelled, but not before lots of fuel was used. All of these factors went into a model.
One of the interesting facts is while the Japanese suffered extreme fuel shortages; the major impact was felt in the civilian economy. The military always had first call on fuel resources and while they were always a consideration in their planning I can find no instance where operations, especially counteroffensives, were curtailed or cancelled due to fuel shortages. The reality is there was a fuel shortage in Japan due to the loss of oilers to US submarines, but not in the amount of fuel available at its source. This is why the Japanese fleet eventually moves closer to the source of fuel in the Dutch East Indies, something that occurs naturally in the game as augmented by an event card. While using unrefined crude oil levies a maintenance penalty on the ships, the Allied air-naval die roll modifiers for the late war in some part account for this performance penalty.
The result of this analysis was a model of historical operations tempo as a collective metric for sortie rate, historical fuel consumption and maintenance capability. I ran this model through a Monte Carlo set of simulation runs and the result was that the Japanese consistently ran out of forces before they could exceed the bounds of their fuel constraints or historical sortie rate. This was confirmed by game testing and a decade of matches that demonstrated if the Japanese player attempted a higher than historical tempo of operation they ran out of navy faster than occurred historically naturally capping the number of sorties and fuel required. If the Japanese player played more of a guerre de course strategy they were using fuel at a lower than historical rate across a larger number of small sorties. If they husbanded their assets for a late war showdown, again they were using less fuel than they did historically. My point is the EotS model in the aggregate captures the Japanese sortie rate bounded by combat attrition and historical operations tempo.
These points aside, some folks like to micro manage fuel points and the like. My design goal was to create a strategic model of the Pacific War that integrated most of the key operational factors in a more streamlined manner. For some this is too abstract and for them I designed Pacific War where you keep track of this stuff.
- Japanese military capabilities included seizure of the Hawaiian Islands, Northern Australian, and Ceylon as well as creating enough political unrest in India to cause them to successfully revolt against the British
One area of WWII history that continues to fascinate me are the minutes and commentary on the big strategy meetings such as the one held in Japan on March 7th, 1942. The Japanese navy wanted to invade the Hawaiian Islands and Northern Australia. Plans had been drawn up that were logistically challenging, but at least on paper were considered feasible. The Army would not support either of these operations, but that was a matter of inter-service politics not capability. For an extensive history of Japan’s plans for invading the Hawaiian Islands a recent book not available to me when I did this game confirms my earlier conclusions (John J. Stephens, Hawaii Under the Rising Sun). So to make the point that this was beyond Japanese capabilities stands in opposition to the view of elements of the Navy staff. Hayashi’s ‘Kogun’ makes the case that an Australian campaign required more logistics than the Navy thought was required and the Army could provide. The fact was those logistics were available to the Army, but they would not release them from what they considered higher priorities. This makes my point that this was more an issue of military politics than logistic potential.
As far as India is concerned there are a host of books that I read that go into great detail on how the Army viewed the value of Bose and Indian unrest in achieving their goal of closing the last supply route supporting Chungking. So, not much to say on this point other than there is a significant body of literature that I read in the Georgetown library that is the source of how this is treated in the game.
- Japanese ability to strategically deploy hundreds of thousands of troops and their equipment was not hindered by the fact that virtually their entire merchant marine was destroyed by Allied airpower and submarines.
I have a table taken from the US Strategic Bomber survey that shows with some precision the amount of merchant shipping possessed by the Japanese for every month of the war. Broadly the Japanese started the war with 6 million tons of merchant shipping which was augmented by 1.2 million tons of captured Allied shipping. The Japanese had sufficient merchant shipping to move their ground forces around until 1945 and even then they were able to reinforce Okinawa, Luzon, and Formosa. In one month a merchant ship can generally get from anywhere on the map to another location. In a four month turn, even accounting for attrition there is sufficient shipping to administratively move forces around. This does not say that the civilian economy was unaffected, but the Army controlled their own shipping and they set the priorities. The short pole in the Japanese tent was less merchant shipping but oil tankers covered in my earlier remarks.
- The pre-war Japanese military was able to establish very large, high capacity, navel and air bases throughout the Pacific from which to combat Allied efforts to seize the islands upon which those bases were to be built by the Allies once the Allies seized them.
One thing I avoided in this design was extensive engineering rules. Even the Japanese could build an airfield within a four month game turn and the Seebees do not even break a sweat to accomplish the same thing. Another consideration is I integrated the US mobile fleet train into my infrastructure model. So, while the capture of Ulithi gives the US a base it is more a function of the anchorage in combination with the large number of auxiliary ships that make this possible. As far as the Japanese side goes, a large amount of the Japanese bases existed before the war or were captured fairly intact (Rabaul). In the end I chose to simplify this dimension of the game. The key is how I treat the auxiliary ships and US underway replenishment groups, floating dry docks, etc.
- Really? It is exactly my issue with the game. In all the games I've seen, the US routinely suffers failed amphibious invasions, and the Japanese fleet retains operational and strategic mobility until the end of the war.
This issue has two dimensions, game play and scale. It is more or less mathematically impossible to fail an amphibious ground combat in EotS. If it occurs it is poor play and has nothing to do with the design. The simple rule of thumb is one Marine division will defeat a brigade, two Marine divisions will defeat a division, and three corps plus any size unit will defeat 2 divisions (full strength army). The only issue is not success or failure but how bloody the battle will be. So, the notion that this is a regular feature of the game is just not true if you have any idea how to play the game. So when somebody says that the majority of invasions succeeded in the Pacific this is true once you establish forces ashore and is fully supported by the design.
However, EotS is a strategic simulation. At the strategic level and the way EotS models offensives there were several failed invasions during the war to include Coral Sea and Midway. In both cases the intent of the attack was to culminate in an invasion, but in these two cases the transports were turned back when the naval battle was lost. These were major strategic failures and the rule whereby losing the air-naval battle precludes a landing captures the strategic nature of these failed invasion offensives.
- If the major Allied operations cards come out wrong, the Allied player may have to run more risk, because simple math will not allow sufficient operational intensity. It is this part that I find worst about EOTS, because it simply makes no sense. What does it mean that "the major Allied operations cards come out wrong"? Those major operations were planned because they were seen as the key stepping stones on the path to victory, and therefore they were given the resources needed to succeed. The cards we find in the game were the things that were built to fit the order in which the cards were needed historically, not the other way round. (I'm aware that the official explanation is that "the cards you get represent the constraints that someone like Nimitz worked under" but having read extensively on the strategic planning in the Pacific, the constraints imposed by the cards do not seem to reflect that style of planning at all.)
One thing that I wrestled with was whether or not to name the cards for historical operations instead of labeling them large offensive, medium offensive, and small offensive. In the end leaving off the historical source of the card is like cooking without spices. As I have said innumerable times the cards are logistics and you can ignore the names if it bothers you, although that is like telling a jury to ignore the inadmissible evidence they just accidently heard. This is one of those if you do not want to see past the titles there is nothing I can say or do at this point that can change it other than to say I like it.
On the other hand the card decks are mathematically constructed. The order of the cards has no impact on strategy except in a fun way. There are times you may have to solve a military puzzle, but the Allies can have the cards come out in just about any order and they will have sufficient logistic resources to do anything they could do historically. The same goes for the Japanese in the opening. This has been demonstrated in numerous public online games under what are considered the worse conditions imagined by novices, such as over half a game with the Allies under ISR and the A-bomb still got dropped.
There are over 5 billion seven card hands possible in the 84 card Allied deck, so no amount of play testing could ever confirm this so I had to calculate all of this mathematically. The acid test that I got it right is I have played this game for over a decade online with a large crowd of participants, collected data on 11 years of WBC tournaments, ran one two-year online tournament and at no time has an experienced Allied player had insufficient resources over the course of any game turn to accomplish what needed to be accomplish. If you have not played this game more than once or twice you likely ran into a problem, but that is inexperience not a design issue. This comment is made often despite its inaccuracy, so it is what it is.
- An EotS abstraction that threw me for a loop is the possibility of the Japanese shipping an entire army by sea to reinforce an invasion target during the invasion operation. IIRC, Mark explained it as US intelligence underestimating the size of the garrison. Well, um... I don't count that as an abstraction, but a distortion, in particular since the Japanese ship that army when they know the operation is already happening. Mark explained it as US intelligence underestimating the size of the garrison. Which, not to belabor a point, but as far as I know never happened during the actual War..... I mean, really, missing a Corps+ (!) on a island <10 mi^2?
There are two pieces to this comment. First, you can only react with one division (~15,000 soldiers), never a Corps or Army. This is an important factual error for what follows.
One of the important features of the EotS design and my earlier Pacific War is I treat intelligence as a central feature in both games. While most games usually only look at the Midway phenomena that is seen as an Allied advantage there was a flip side of notable failures. I researched how many Japanese defenders were believed to be present for all of the major invasions during the war, such as Bougainville, Peleliu, Saipan, Iwo Jima, Okinawa, and the Philippines. In all of these cases the intelligence estimate was off by at least 15,000 soldiers, one division. In the case of the Philippines the estimate was off by 250,000 soldiers. This data is not hard to find, so it is easy to validate.
The Japanese were very competent and it was not hard for them to anticipate the likely locations for the next US offensive. In most cases the reason that the intelligence estimate was off is the Japanese reinforced the island proximate to the invasion. The Allies for the most part relied on photographic intelligence to estimate troop strengths based on a formula of visual clues. The problem was the Japanese got very good at overhead cover discipline and they were excellent tunnel engineers. This made estimating the size of the garrison an art more than a science. There is a reason that Peleliu and Saipan were estimated to be 3 day operations and instead took almost a month.
The one fly in this ointment is the method by which I chose to incorporate this intelligence feature. The reaction move of a division while the invasion force is already at that location is a mechanistic abstraction that can be visually tough to reconcile. I did not want to have people writing stuff down on paper, etc. and the method chosen works quite smoothly with the other elements of the design. This is one of those that you either buy the abstraction or play another game.
I think that covers the main points I wanted to cover. I want to reiterate I appreciate the commentary and the fact that the individuals played the game to form an opinion and stated it so I could continue to refine my thinking on this topic. I hope that I have protected the anonymity of the commentators for it is not my intention to start a feud. They are only saying what others have said before, but I had the time and the inclination to write my thoughts down. As I stated in the opening this is being posted in my blog so I can archive my own thoughts and what I have written will not change any opinions that already exist either for or against this design of mine. All I can say is I am really looking forward to continuing to play this design with the new 2nd edition.