In the last design post, I gave up (on the Durak-derived resolution system I had been tinkering with). I ended that post by declaring that I was just going to pop away and do some reading about Napoleon's battles and see what kind of categories I could place them in. Well, it's been three weeks now, and I've done some reading, some very simple analysis, and I think I've reached some conclusions.

I went through all of Napoleon's battles and read a little about each of them, seeking to first capture a notion of the size of each of those battles, and then at least a little bit about what might make the battle out of the ordinary. Every battle, of course, had unique circumstances associated with it, be it the location, the strategic situation, the terrain, the forces involved, or whatever else. But I was really trying to see if there were large factors that might allow me to sort battles into some buckets. I took some quick notes on what I thought was the most exceptional thing for each battle, and wrote that down for each along with the size of the battle in a spreadsheet.

I then went back and read through all of those notes and there were a couple of categories that seemed pretty obvious. One is that the truly massive battles were kind of their own beast. Of course, the line between the really big ones and the regular battles was somewhat arbitrary, but obviously Leipzig, Dresden, and Wagram were in. I counted Borodino among the group by virtue of being such a bloodbath. Aspern-Essling seemed like a more normal battle to me, so the question was what to do with Bautzen and Waterloo, the two battles inbetween Borodino and Aspern-Essling in size. Waterloo is certainly famous, but it doesn't strike me as exceptional in how it played out, nor did Bautzen, so that's my line. Four massive battles that had their own rules.

With that line determined, the next obvious category was sieges. There were a fair number throughout Napoleon's career, with some called sieges and some that revolved around fortifications that might not have been called sieges. In all, I roughly categorized 15 of the 77 battles in my list as either sieges or siege-ish. Basically, battles where artillery was center stage, infantry did some grunt work, and cavalry was really not very important.

The remaining battles, 58 of them, needed to be broken down a little bit more. Having at least one more category of significant seemed like it would be important. Reading through my notes, what stood out was that there were a number of battles where cavalry took center stage. These ranged from pursuits, to meeting engagements, to some of the battles in Egypt where the Mamluk side was mostly cavalry, to rearguard action, to some battles where things were decided by large cavalry charges. There's a little less tying all these battles together compared to the sieges, but grouping them together pulled 19 of the 77 battles together. That left the remaining 39 battles, basically half, as sort of the "standard" battles.

With my categories more or less figured out, at least for now, I needed to express them in game terms. The commonality among the categories is that each of them had differences in the importance of cavalry, ranging from the most important arm (in the pursuit/mobile battles) to the least important (the sieges), with the standard battles in-between. That translates nicely to just changing the ordering of trump suits. Standard would be my standard artillery/cavalry/infantry ordering, mobile battles would be cavalry/artillery/infantry, and sieges would be artillery/infantry/cavalry. There's kind of a neat little thing there, where artillery always beats infantry, which just kind of feels right.

As a digression, there's some real "design for effect" going on here in wargame terms. That is, in Eylau (for example), did we know ahead of time that it would feature one of the great cavalry charges of history? The initial conditions of the battle weren't necessarily ones where we would have expected that to happen. The causes of why cavalry played such a big role in that battle aren't probed at all in this design. The game captures the effect, that Eylau was dominated by cavalry, without really explaining any of the factors that resulted in that. It's an approach to designing historical games that I would normally shy away from, but this is only barely a historical game. I think it's totally fine for something this abstract, but I thought it was worth mentioning.

So, finally, what to do with the massive battles? My instinct is that they were such big affairs that each arm had its role to play, and so maybe those battles should be played at no-trump. Why not? The other nice thing is that the counts of each category, turned into percentages, come out to roughly 5% massive, 20% sieges, 25% mobile/pursuit, and 50% standard. That can easily turn into a deck of a multiple of 20 cards. I'm going to go with 40, because I want to preserve the possibility of two massive battles in the layout. And so: 2 massive, 8 sieges, 10 mobile/pursuit, and 20 standard.

The last thing was looking at the sizes of battles. If I assume 8 different sizes of battle and bucket the battles into those sizes, the buckets are pretty asymmetric. The first 8th and second 8th of sizes comprise more than half of my list (20 and 23 battles), so a straight mapping probably wasn't going to work out that well. But it does suggest skewing the sizes down some. In particular, the sieges tended to be small, with the sieges falling mostly into the two smallest buckets, three in the third-smallest, and only one in the fourth-smallest. No sieges were above the median. The pursuit/mobile battles were larger, with Eylau and La Rothière in the second largest bucket. That suggests some asymmetry in the battle types with regard to sizes. The sieges will mostly be the smallest battle types, the pursuit/mobile battles a little bigger, and the remaining battles will skew larger. And, of course, the massive ones will be, uh, massive.

And after that long digression, I arrive at the next iteration of the design. The main deck will be the same as last time: infantry cards (2-10 x 3), cavalry cards (2-10 x 2), artillery cards (2-10), four commanders, and four scouts (2 x 2, 3 x 2), for a total of 62 cards. The battle site deck will be 2 massive cards (2 x 12, played at no-trump), 8 siege cards (2 x 2, 3 x 2, 4 x 2, 5, 6), 10 pursuit/mobile cards (2, 3, 4, 5 x 2, 6 x 2, 7, 8, 9), and 20 standard cards (2, 3, 4, 5 x 2, 6 x 2, 7 x 3, 8 x 3, 9 x 3, 10 x 4). Commanders count as the lowest card of the highest suit for the battle type, scouts can either add to the rank of a card (and then draw) or be discarded to draw cards equal to their rank (choosing one and discarding the rest). Finally, the number of battles is 8, you win if you score 9 Rout points, and the number of cards in the reserve is 6. If all that is confusing, well, I'm mostly recording it here for my own memory. But if this works, I'm going to put out a full set of rules for the two-player version and maybe it will make sense.

Coming up next, testing this thing out, updating the formal rules to reflect all these changes, and then starting on the four-player partnership version. I actually think that that version might be the most interesting version of the game, I'm excited to start on it.