|Products||| Napoleon’s Triumph||| Design Diary||| That Was the Month That Was|
On October 29 I posted a design diary entry that explained the scope and nature of the problems that had been afflicting the design. At this point, I am reasonably confident that I have solutions to all of those problems and that the design is now a matter of tuning with no significant conceptual work remaining (famous last words, no doubt). A few days ago I posted a link to the second draft of the rules, and now post a link to the third draft (click on the image to the left to download them).
This entry is intended as a point-by-point response to the October 29 entry (which you can read by clicking here if you are so inclined). By its nature, this means that this diary entry must cover some of the same ground as previous entries from the last month, but I think on balance it will have enough more than enough new material to be worthwhile.
The November diary entries cover this aspect of the design in terms of its genesis pretty well, though they do not cover the latest developments in it (a reader of the rules will find that what the rules describe differs significantly from where it was left in the last design diary entry). Rather than re-tell the story in terms of its evolutionary history, I will focus instead on the current state of the game, and how I think it solves the problems raised on 29 October.
The burden of attack was the problem that dogged this aspect of the design from the outset. The game currently has two different variants that can be played, each of which has its own victory conditions: French Attack and Allied Attack.
The French Attack variant was included because it is an interesting variant to play, and because there is nothing inherently improbable about it from a historical point of view: it simply simulates what would have happened if the Allies had gotten cold feet on the verge of attacking the French army, and decided to go on the defensive. It was what Napoleon feared they would do, what all of his deceptions had been aimed at preventing, but ultimately what he could not prevent if the Allies had simply decided to do it. I think players will find it an enjoyable experience but in the end it is something of a historical might-have-been, and I think of it as more of a side dish than the main course.
The Allied Attack variant is the version of the game I really wanted and sweated to get. What makes this variant go is that the Allies actually have two different sets of victory conditions: one set before the French reinforcements (Bernadotte and Davout) arrive and a different set afterwards. The latter requires an Allied attack; its purpose in the game is to pull the Allies forward: if they just sit on the defensive, the French will keep their reinforcements off the map and win by default. The Allies have to attack with enough force to compel the French to bring on their reinforcements, but once the French do so, the victory conditions shift to put the full burden of attack on the French: the Allies then need only backpedal and hold on to win.
I am very pleased with the playtest games I've run so far with the Allied Attack option. The French player really does get the sense that he is trying to trap the Allies, and the Allied player has the interesting problem of trying to spring the trap without getting caught by it. While I expect it will take quite a bit of testing to ensure that the play balance is good, the basic mechanism seems to work as well as I could have hoped.
Readers of previous diary entries will of course wonder what happened to the variable French reinforcement options. On reflection, I decided that by far the two most interesting options were the two that have made it into the game: the historical schedule but with an option for either side to have the burden of attack. The French Reinforcement option where they got neither Davout nor Bernadotte just didn’t seem that interesting for the French to play, and the French Reinforcement option where they got Bernadotte but not Davout was just too even in terms of strength: it was hard to design victory conditions that put the burden of attack on either side without making the game too hard for them to win. It just didn’t seem to me that those two options offered enough play value for the work that would be required to playtest and validate them, they aren’t required from a simulation perspective, and with the Allied Attack variant working well, they just aren’t needed.
In terms of space, there has been no change to the scope of the game since the first version of the game was produced. In terms of time, there are at present four versions: one starting on the day of battle, one starting 18 hours earlier, one starting 24 hours earlier, and one starting 48 hours earlier. Of these, the first is certain to make it into the game, the second is highly probable, the third is probable, and the fourth is possible.
The day of battle scenario is a given: it is the scenario that anyone buying a game on Austerlitz expects to get, and it is the shortest scenario. These are two excellent reasons for its inclusion, and I seriously doubt I would even ship a game on Austerlitz that didn’t have this scenario. As the main scenario, this is the one that will be the focus of playtesting.
The 18 hours earlier scenario is essentially a free deployment scenario. The Allies can bring their forces on anywhere with no immediate opposition from the French, although the French do have a real chance to race the Allies to the Pratzen and do what Napoleon could have done but declined to do: fight further forward. Even if the French lay back, the Allies can deploy their army very differently than they did historically. It is longer than the day of battle scenario, but should still make a manageable single evening game. playtesting it should be interesting: it has potential, and I hope that it will live up to it.
The 24 hours earlier scenario gives the armies an entire additional day. In theory this allows the Allies to attempt to get something done before Davout can show up, but it remains to be seen how well this scenario will work in practice. It is on the outer edge of a single evening game: some players on some evenings will manage it, but others may find that they have reached the end of their evening without reaching the end of their game. The big question about this scenario is whether it is really different enough from the 18 hours earlier scenario to justify its existence.
The 48 hours earlier scenario is the most problematic because of its length. It has an estimated playing time in the 6 to 10 hours range, which marks it as an all-day rather than an evening game. I don’t object to games of that length on principal, although they are obviously more difficult for players to actually play than shorter games. My main concern about this scenario is the drain it will pose for playtest time. Nothing would pain me more than the thought that in reaching for this scenario I let a serious problem with one of the shorter scenarios go unnoticed. We’ll see, but as I think about the schedule more and more, I grow less and less optimistic about this scenario, and with the main scenario working so well so far, I feel less need for multi-scenario insurance against its limitations.
There is no bigger difference in my mind between Napoleon’s Triumph and Bonaparte at Marengo than that Napoleon’s Triumph has commanders and Bonaparte at Marengo doesn’t. If I could change Bonaparte at Marengo, I wouldn’t: for reasons I’ve explained before, I don’t think commanders would make that game better (in fact I’m pretty confident they would make it worse). That said, the scale and nature of Austerlitz made it really, really need commanders in order to recreate a sense of the battle.
Commanders don’t affect a localized part of the design, they affect all of it, but they really function to split the game into two different levels: a macro-level where the player is managing the battlefield, and a micro-level where the player is managing a particular corps.
At the battlefield management level, each player is operating with eight corps, and that’s it. The importance of this can’t be overemphasized. Players used to Bonaparte at Marengo may see the additional units in Napoleon’s Triumph and think that there are a lot more pieces to move, but on the scale of the battlefield as a whole, that’s an illusion: there are actually a lot less pieces to move. With every unit detached, the entire game would not be long enough for a player to move his army a single locale. It is only when moving by corps that the army has mobility.
The paradox is that the army can only move by corps, but as it starts to fight, it starts to detach. Eight corps cannot cover the entire front without doing so: some units need to be detached to protect the flanks of the army, and tactical requirements for individual corps will require them to detach units for purely local considerations. A corps can attack as a whole, but if it attacks, it had better win: the defeat of an entire corps is a serious matter: all of its units will get detached, an event from which it is unlikely that the corps will ever recover. This means that attacks by detachments become necessary. Also necessary are detachments for defense: a corps that remains intact can only block a single approach, and eight approaches is too narrow a front for security (not to mention that eight approaches assumes that the army has no reserves). This problem of when, where, and how much to detach from a corps is the heart of the corps-level (as opposed to battlefield-level) game.
The road movement rules in Napoleon’s Triumph are different from those of Bonaparte at Marengo in many ways, but are like them in one way: they are good enough, and that’s about all. Napoleon’s Triumph actually has two conceptually different sets of road movement rules which are awkwardly spliced together: one for individual units and one for corps.
The rules for individual units are a refined version of the road rules for Bonaparte at Marengo. In fact, they could easily be the basis for a revised set of rules for that game: they are not quite as realistic, but they are much easier to manage. Bonaparte at Marengo required that players track the step at which individual units crossed individual approaches while Napoleon’s Triumph only requires that players track for how many units have moved by road out of the locale where other units start their move. This is actually quite a bit easier, and in most situations yields identical results to the more complex rules of Bonaparte at Marengo. Really, though, the rules for individual units are a side show for Napoleon’s Triumph: without the command limit exemption for primary roads, the number of road moves by individual units in Napoleon’s Triumph is sharply limited and they will tend to be spread out over a large area, making complex moves by them in close quarters quite rare.
The real action for road movement in Napoleon’s Triumph is in the corps road movement rules. Road columns are abstracted, sacrificing quite a bit of realism in order to bring the complexity of road movement under control. Their general purpose is to allow the players to get corps around the board in an acceptably realistic way with a minimum of fuss. The real design weakness in the Napoleon’s Triumph road rules isn’t in the individual unit rules or the corps rules individually, since they both actually work pretty well for their respective jobs: the problem is that the two have ended up co-existing in the same game.
This problem, however vexing it may be from the standpoint of design purity, does not even come close to sinking the game in my opinion. I hope that I will get some bright idea to reconcile the two more smoothly so that they form a more cohesive design, but I’m not very optimistic about it. I think that on the whole, they are much less of a problem for players of Napoleon’s Triumph than the Bonaparte at Marengo rules are for players of that game, which is all I really need to make this game work, and so, I fully intend to declare victory over this problem and leave it pretty much alone from here on out.
The combat rules of Bonaparte at Marengo were both a challenge and opportunity in the design of Napoleon’s Triumph. In the early development versions of Bonaparte at Marengo, there were two different kinds of attacks (bombardment and assaults) and there were not separate phases: all combat occurred during movement. The phases were added later because it proved difficult to track which units had moved and which hadn’t when trying to figure out which units could attack, and splitting assaults and bombardments into their own phases helped matters. Maneuver attacks were relatively late additions to the design to solve the problem of making pursuit and retreat work.
A distinguishing characteristic of Bonaparte at Marengo is the sharp divide between maneuver attacks, which were never caused attacker losses and assaults, which always caused losses. Bonaparte at Marengo is often said to be like an abstract game, and it is chiefly maneuver attacks that people have in mind when they say this (if you check user comments at BoardGameGeek, one thing you will find is that some people see this as a good thing and others a bad thing, but there is a general agreement that it is the case).
This sharp distinction is blurred in Napoleon’s Triumph. There is only one type of attack (bombardments are also folded into it), which sometimes causes losses and sometimes not, and which always occurs during movement: the turns of Napoleon’s Triumph are without phases. One motivation was to make the game more fluid. The assault rules in Bonaparte at Marengo were designed primarily with the prolonged conflict at the Fontanone in mind: the French army held a fixed position which was repeatedly assaulted and bombarded by the Austrians for about four hours. The maneuver attack rules were designed with the subsequent pursuit and withdrawal in mind. Both work reasonably well at their assigned tasks (and in fact, I think the pursuit rules are something quite new in this period and at this scale, which previously had not been simulated very well), but neither could represent something between the two: a more fluid fight, such as occurred at Austerlitz.
Complicating matters was the game’s complexity budget. Napoleon’s Triumph had a number of game elements that Bonaparte at Marengo did not (commanders and ‘super-elite’ units being the most important), and it still needed to simulate something that Bonaparte at Marengo did not. In order to keep the complexity down, some aspects of Napoleon’s Triumph would have to be simpler than those of Bonaparte at Marengo. Most of Bonaparte at Marengo’s complexity was in its three-pronged combat rules and corresponding phases. It really wouldn’t be possible to cut as much complexity as was needed in any area other than there. And so it was decided first to get rid of the phases if at all possible. This would both save rules and was inherently a step towards a more fluid game. It was also decided to try to collapse the three kinds of combat together, which would also save rules and also contribute towards a more fluid game.
Another combat problem was the balance between attack and defense. Bonaparte at Marengo is tilted strongly towards the defense, but the tilt gets especially strong as the number of defenders multiplies. This is not a severe problem in Bonaparte at Marengo because of the small armies, although it isn’t ideal even in that game. The assault rules in Bonaparte at Marengo originally included a complex counter-attack system which was dropped when maneuver attacks were added. The result simplified the system, but in order to keep the balance between the attack and defense, other rules were changed to make the defense stronger. The result was a defense that is very strong right up to the moment when it breaks. With its greater unit density, Austerlitz needed assault rules that could not only simulate a more fluid fight, but which also would soften defenses, so that it was possible to do a defensive position more damage without necessarily breaking it.
The result of this process is a heavily re-arranged combat system. Some procedures that strongly favored the defense in Bonaparte at Marengo (such as making the attacker reveal his leading units before the defender) have been reversed (in Napoleon’s Triumph, the defender reveals his leading units before the attacker). At the same time, the defense has been strengthened by adding back a simplified version of the counter-attack rules that were dropped from Bonaparte at Marengo. Another critical change is the ability to inflict losses with an attack from reserve. Still another is the stepped-up losses when combat occurs on wide approaches (arguably losses on wide approaches should be heavier in Bonaparte at Marengo as well). The net effect of these changes is that combat is generally faster and more violent in Napoleon’s Triumph than it is in Bonaparte at Marengo, which has an almost eighteenth century feel by comparison. Interestingly, I think each matches its own battle pretty well: it may be that a grand unification is possible, but I would not be tempted to try it if I thought the result would be a more complicated game. To my mind, it is better that the rules for each game serve it well, than that each is cluttered with complications that the other needs but it doesn't.
After four months of banging my head against the wall, it is really surprising how quickly the design has come together in the last month. The writing of the rules is really a major step: playtesting a game in order to tune it is a process that urgently needs formal written rules, but such rules are a hindrance earlier on, when the major questions in the design are all up in the air. That a week ago I felt that I really needed to write up the rules before I could make any more progress says a great many good things about the status of the design. Sadly, though, I missed the bicentennial of the battle, which I was originally hoping to hit. Of course, better a good game late than a bad game on time, and while I don’t regret trying to force it into print before it was ready, I can’t help but wish that I had been smart enough to think of all this months ago...