Trying a New Resolution Engine

Last blog, I mentioned that the change over to a specific deck for Napoleon, Blown Apart resulted in a worse game, with a sort of uncanny valley of feel resulting from the new deck and decidedly less interesting gameplay. I'll resist the impulse to recap that entire blog, but my preferred path forward out of this was to look at some possible replacements for pure trick-taking for the resolution of actions.

Essentially, at the core of it, the previous versions of the game were very elaborate setups to get you to a hand of trick-taking. The rest of the surrounding armature of the game was to set up different scoring situations, have some jockeying with how you strengthen your hand, and some other positioning, but at its core, this was a trick-taking game with a couple of twists in the cardplay (the commanders and scouts). But it doesn't have to be that way. The Battle/Rout/Reserve components would function fine (probably!) with some other type of head-to-head card resolution system. I read through a bunch of card games to draw on some inspiration, and while some ideas seemed like they might have some promise (what about Guts, which is basically three card poker?), the most promising was a Russian game called Durak ("Fool").

You can read the rules to it there, but roughly speaking, Durak is played in "bouts", where the attacker plays a card, the defender plays a superior card (higher rank or trump), and the attacker can continue the attack by playing any card matching the rank of any of the played cards. It goes back and forth until one player concedes. In Durak, you then pick up those cards, and the game is lost by the last player with cards in hand. Because this is related to trick-taking, but is modeled around bouts, I thought it might be a good fit for the martial thing I was trying to put together. So I adapted Durak and created a new resolution system.

The basic idea is that each bout would represent a skirmish. The attacker (the player who selected the Battle site) would play any card, and it proceed as in Durak. However, I would have a heirarchy of cards (infanty < cavalry < artillery) allowing stronger suits to trump weaker ones, and once you move up the heirarchy, you couldn't move back down in the same skirmish. Commanders could always be played, but would reset the heirarchy, so any card could be played in response to them (respecting the restrictions about having to play a duplicate rank on attack). Once someone won the skirmish, they would take all the cards from the skirmish, and at the end of the Battle, you would count captured cards and whoever took more cards scored the difference. A Rout would be if you took cards and your opponent took none. Scouts? Well, I would figure those out later. The last thing I did was bump up the Battle sizes by two across the board, so instead of 2-8, they were 4-10, because you needed some larger hands to make a skirmish even work. I also made it so that the suits were irrelevant for Battles, at least for now. I would revisit this later if the structure worked, I wasn't worried about making the Battle sites interesting if the core of the game worked.

I started out by just dealing myself some hands and running them against each other. It seemed like it might work ok - I had to do some thinking to play the hands out, which was all I was really looking hoping to see. So, I set up a full game, dealt myself hands, and there were some details that clearly needed to be worked out. What ends the Battle? Because a player might not be able to respond to a card, you can end up with unequal hands, and the ending condition for a Battle wasn't always obvious. As a special case, what happens when both players run out of cards at the same time and there is an active skirmish? Whatever rules I applied here would make a big difference in how the game worked, especially in the smaller Battles which only featured a few cards. I tried a few variations of things here, and nothing quite felt totally right, with some real complexity in how to compute things ending. Either it was too hard to win cards and you might get a Rout with only a couple cards, or it was too easy to manipulate the Battle ending resulting in some artificial-feeling gameplay, or other strange situations that kept popping up.

Another problem cropped up after I played a few games. I wanted to try and keep things stable because I wanted to really get a feel for the new resolution system. I didn't have the same ability to do a snap evaluation of this as I do straight trick-taking, so I didn't want to judge the new rules too quickly. But as I played, it became clear to me that the defender had a huge advantage in the structure of the game. Generally, the way things were playing out was that the attacker would push as much as they could in the first skirmish, but the defender would usually win out because of the challenge of matching ranks as the skirmish wore on. That first pile of cards was often enough to win the small and medium battles on its own, with the rest of the Battle determining if it's a close win or a big win. The alternative, of the attacker conceding after one card each just to go on defense, felt even worse. Flipping things so that the Battle selector is the first defender seemed like it would help with the point imbalance, but I fundamentally couldn't let the defender have such a huge advantage, as it would distort game play.

So that's where I was: the structure seemed interesting, the boundary conditions of Battles were fuzzy and high-impact, and there was a big defender bias. All together, there was promise, but some substantial fixes were needed. I decided to think about a small problem for a bit, and maybe the big problems would have some suggestions come up. Sometimes that sort of narrowing can help. I decided to think about scouts.

The scouts were respresented in the current deck by two 2s and 3s of Spades. I kicked around several ideas but decided to focus on one of the things giving the defender advantage: it was hard to match ranks. What if the scouts helped with that? You could play a scout with another card, and adjust the rank by up to the scout's value. It would allow the attack to continue the skirmish while being basically useless to the defender. There weren't that many scouts in the deck, but it would be a positive adjustment to the balance and add some more tactical options. Seemed promising.

However, it would exacerbate the Battle ending problem, because it would cause a player to spend an extra card in more circumstances. I had to fix that. What if you also draw a card when you play a scout? It's now hand-size neutral, it potentially gives some more help to an attacker (with some luck!), and it wouldn't make the Battle problem worse. It even was kind of thematic - the scout found a way to pull in some reinforcements by getting them to the battle? It is also an improvement over having to look at your Reserve in the game, which could slow things down. Again, it seemed promising.

But that idea suggested another fix. If I'm comfortable adding cards during the Battle, what about losing cards? What if you had to discard a card whenever you passed on a play as a defender? Call it additional losses from a breakthrough, whatever, we can justify it in the game's fiction. But it would stop hands from getting imbalanced (an attacker that cannot attack doesn't need to discard a card as the number of card plays are equal at that point). It further clarifies the end of Battles: a Battle is over when both players are out of cards. If the defender played on the skirmish, nobody wins, otherwise the attacker wins it (after the discard). Suddenly, the boundary conditions would go away and there would be clarity. Nice!

The final thing that I thought about was revisiting the deck. The asymmetry in ranks which was intended to work well with the no-trump Battles in the previous iteration was, on reflection, causing problems with this version. Especially once things ratched up to artillery, the exotic ranks were extremely hard to match for the attacker, leading to some of the defender bias. It also had some strange side effects. 8s were some of the best cards for the attacker, because there were six of them in the deck, with 5-7 next best (five copies), while Ks and As were extraordinarily valuable on defense but almost useless on attack. What was intended to provide some fun texture was causing me pain. I could get rid of that pretty easily.

Overall, then, I have another major version to try. I'd keep hands at parity throughout Battles, I'd re-arrange the deck so that cards were all 2-10 for all suits, and I'd try out the new scout rules. I'm optimistic that the new rules might work out well, and perhaps would finally get me to an asymmetric deck with Napoleonic flavor, some more combat-related semantics for resolution, and a fresh feel for card play compared to pure trick-taking. That's what I'll report on next time. And if that works, I'll fine-tune the deck, maybe try and get some more Napoleonic flavor into it, and then take a hard look at making the Battles distinct.