Home Store Products Research Design Strategy Support News

4 November, 2005

 Products Napoleon’s Triumph Design Diary Follow the Leader

Napoleon’s Triumph Design Diary: Follow the Leader

The most important change in the design since the original conception of the game is the introduction of leader pieces. Bonaparte at Marengo had a highly abstracted command system, but I had a desire to make command in Napoleon’s Triumph less abstract and more concrete. The original conception was based on printing corps designations on the pieces. The October 29 design diary entry already detailed the deficiencies of this approach, and there is no need to repeat them here.

The use of leaders in Napoleonic wargaming is hardly anything new: they’ve been around for decades. Given this, the place to start the discussion of the addition of leaders into this game is by considering why the design didn’t always have them (a question that can also apply to Bonaparte at Marengo as well, of course). The most important reason has to do with game complexity: the introduction of any new class of playing pieces always raises a game system’s complexity level: for every rule, it has to be considered whether it applies to one or both classes of pieces, and if only one class, whether some other rule needs to exist for the other class. This necessarily increases the complexity of the rules, regardless of all other considerations.

In Bonaparte at Marengo, from which Napoleon’s Triumph is derived, there was no obvious benefit from leaders. In that battle, the Austrian army was organized at the highest level by column, of which there were three (right, center, and left). The center column was much the largest, comprising about two-thirds of the Austrian army and was in fact personally commanded by Melas, the Austrian commander-in-chief. The next command level in the Austrian army was – confusingly – also the column (we’ll refer to the smaller columns here as divisions for the sake of clarity). The divisions were small – typically equal in size to about two pieces in the game. With regard to the French army, it had a corps organization but in the battle the French infantry really operated by division (only one French corps in the battle had more than one division operating together) and the French cavalry operated by individual brigade. It would have been possible to include French division commanders in the game, but French divisions weren’t that big (just one, two or three pieces each). It seemed to me that given the historical organization of the two armies, having the players control individual pieces was an acceptable from a simulation perspective and leader pieces would not be worth the additional complexity that their presence would add.

Austerlitz, of course was a much larger battle than Marengo, and the armies really were functionally organized by units that were much larger than the individual pieces represented in the game (the Allied army was organized essentially by column and the French army by corps). From the perspective of the army commanders, these were the units by which they exercised command: formations smaller than those were largely under the control of lower-level commanders. Because I always want to solve any problem at the lowest complexity price I can manage, my original idea was to represent these formations by printing command designations on the pieces. It seemed to me that adding an additional property to the existing classes of pieces (and a property whose effects would be limited to a very few rules) was a cheap solution to the problem, and that introducing leader pieces would necessarily be much more expensive in terms of game complexity.

French Leaders Allied Leaders

There was a problem with that approach: it didn’t work. The game still felt very much like the players were controlling the individual pieces without any intermediate command levels (the rules gave reason to keep the pieces in certain clusters, but that was about it). After a long, long delay I decided to try the game with leader pieces to represent the corps-level formations (treating the Allied columns and similarly-sized French corps the same). Some exceptions did have to be made: the Soult’s very large IV Corps had three large infantry divisions, which were more easily treated as ‘corps’ in their own right rather than trying to stretch the leader rules to include so large a formation. Soult himself, like the Allied commander Buxhöwden, is not included in the game. There were two other anomolous formations: the French Reserve Grenadier division (this formation stayed with the Guard and it seemed that providing a Corps commander for the Guard would allow it to be adequately treated from a command perspective), and the Allied IV Column, which was nominally under the joint command of Miloradovich and Kolowrat in the battle, but which in practice seems to have fallen under the command of Kutuzov, who accompanied it in its movements (Kutuzov himself was displaced from his nominal command by the presence of Czar Alexander). Players may of course wonder where Napoleon and Alexander are in the piece mix, but if you want to see how they are represented in the game, get a mirror.

Although I've focused on the negatives of introducing a new class of pieces, there are some positives: they give a design a new tool in addressing design problems. In this regard, the introduction of leaders has been extremely helpful in addressing a surprising (surprising to me, anyway) number of serious design problems. I will save most of these for later design diary entries, but there is one I’d like to go into right now: road movement.

The road movement rules from Bonaparte at Marengo worked in that game because almost all road movement (apart from the odd cavalry flank march) was along the primary roads. This made the compexity of the system bearable, if not welcome. In Napoleon’s Triumph, where there are many more pieces and much, much more use of secondary roads, the complexity of the Marengo system becomes unbearable. Part of my original idea for road movement in Napoleon’s Triumph was that road movement would be by “road columns” where one piece would be designated the head of the column, and the other pieces would be moved one a time after it down the same stretch of road. It helped some, but not nearly enough. The patterns of movement were still simply too difficult to manage and multiple road columns could coordinate criss-crossing moves of vastly greater complexity than the historical formations could ever have managed. Thus, the rules were failures from both a simulation and playability point of view.

It may not be obvious how leader pieces could solve a road movement problem, but in fact they helped enormously (this was one of the most important things in confirming me in the decision of introducing such pieces into the design). Before I can go into the details of how this would work, I need to give some background on how command is to work in the game. In Bonaparte at Marengo, the players had three commands to expend a turn, which could be used to move a piece or group of pieces (pieces could move as a group if they were starting and ending their move in the same place). Expending commands were the only way pieces could be moved, apart from road movement along primary roads. In Napoleon’s Triumph, “commands” are replaced by “orders” (the terminology change is necessary because in the new game, a “command” is a set of pieces under the command of a leader). Each player gets three direct orders a turn, which can be used to move any individual piece (but not groups). In addition, each leader can issue one order, but only to pieces in his command (which must be co-located with him – meaning either in reserve in the same locale, or blocking the same approach). A leader can either: (a) order all of his command to move with him to a different position, (b) detach one or more pieces and move them to a different position, after which they are no longer in his command, or (c) attach a single piece with which he is co-located to bring it into his command (attaching a piece also requires the use of one of the three direct orders).

Ironically, at first I thought of the co-location rule as being a major problem for handling road movement, since the pieces in the road column couldn’t be co-located with the leader. I considered various special rules to cover that case, but then I got a better idea: to abstract the road column instead. The pieces under a leader would be physically co-located with him, even when moving by road. Also, rather than have the fiddly Marengo rule restricting the number and order of pieces that could cross an approach by road, I would put down a blanket fiat: if a leader moved his command by road, no other friendly piece could move into the locale in which it started, ended or through which it moved that turn, as the road column would have been occupying all of those locales through the entire move. While there are some fine points in all of this that I haven’t mentioned, there were three huge advantages of this approach:

  1. It was mechanically much easier for players. Rather than move individual pieces one at a time, each tracing its own route and with its own start and end point, entire groups of pieces would be moved at once, from a single start point to a single end point. As a result, the game will play significantly faster than it would otherwise.
  2. The rules would be much easier to communicate and understand (I don’t know if I’ve proven that at this point, but if not I do believe I will before all is said and done), both in terms of reading the rules and in executing those rules in play.
  3. The rule blocking pieces from entering locales used for the road movement route of another group would eliminate the problem of the players being able to make road moves of extraordinary sophistication that would not have been possible historically: groups marching in parallel would need clear and distinct roads on which to move; only simple road movements would be possible.

The illustration below compares moving a force by road, with the old way on the left and the new way on the right. The old way is more compex to execute (tracking those numbers for the last move across an approach isn’t always easy to do in your head) but it does have the virtue of showing where each piece would end up at the end. The new way is much easier to execute, but the presence of the tail of the road column is abstracted rather than literally represented. One way isn’t necessarily “better” than the other – they are different solutions to the same problem, with different strengths and weaknesses whose acceptability is situation-dependent. While I don’t think that the new way would work particularly well for Bonaparte at Marengo, I think it might work very well for Austerlitz.

Moving a Force by Road

The Old Way The New Way

before the first move

before the move

the first move

the second move

the move

the third move

after the moves

after the move

–––––

To wrap up, one of the interesting points of comparison between Bonaparte at Marengo and Napoleon’s Triumph is that the former game benefited from an abstracted command system but a literal road movement system (I will not argue that the road system in that game is the best one that I could have designed, but I will argue that a literal system of some kind is a better fit for the game), while Napoleon’s Triumph is benefiting from a literal command system but an abstracted road movement system. This difference points out one of the difficulties of “system” designs (those in which almost identical rules are applied to a variety of subjects). While that approach does have advantages, it is not necessarily true that a single system will yield the best results for each game: there are design trade-offs whose merits can and do depend on the particulars of the subject matter, which are difficult to satisfactorily handle in designs that are constrained to have “universal” application.