Shadowrift is a cooperative deck-building game.
It has a few interesting mechanics.
One is the monster power mechanic, which operates similarly to a japanese deer-scarer, or a relaxation oscillator. There is a deck of monsters, which is placed face up. Each turn, the monster faction gains as many tokens as heros are playing. Then, if the monster faction has enough tokens (equal to or greater than a number marked on the card), the next monster comes off the top of the deck, and starts attacking the town. However, the monster faction loses that number of tokens. If there are remaining tokens, more than one monster can come of the top of the deck on the same turn. If there are remaining tokens, but not enough to purchase the next monster, the tokens carry over to the next turn.
There are several interesting aspects to this.
- Inter-arrival times correlated with difficulty.
- A single tunable parameter.
Telegraphing is a word in game design borrowed from boxing. The corresponding word from poker is "a tell". In "Mike Tyson's Punch Out" Piston Honda is programmed to wiggle his eyebrows immediately before punching; the other fighters telegraph in other, generally more subtle ways. Adding telegraphing to a game allows the player to see deeper into the future, and gives the player a richer landscape to learn.
Inter-arrival times correlated with difficulty creates a mean-reversion effect, where after a relatively easy span of play, a more-difficult-than-usual challenge occurs.
The mechanism that typically creates a mean-reversion effect is a deck. Mean-reversion-via-a-deck is how card-counters in blackjack get their advantage; after a sequence of high cards, the remaining cards in the deck are mostly low cards. When the remaining cards in the deck are mostly low cards (the deck is "hot"), the card-counter raises their bet, and they can win back enough to pay for all the watching and waiting while losing minimum-size bets.
This mean-reversion-via-a-deck occurs in Shadowrift too (and deck-building games in general). Because the player cycles through their deck over and over again, a hand that is rich in one particular type of card (say "prowess") is more likely followed by a hand that has few of that type of card.
However, having more mechanisms to create mean reversion is great.
The single tunable parameter is great. Of course, there is a whole deck that contains the monsters, and and each card in the monster deck has not just the number that controls how much monster power it takes to bring it out of the deck, but also other numbers and text that controls how problematic that monster is, so saying "this mechanic only has one parameter!" is a little deceptive. It's used, as I mentioned before, to adjust the opposition strength to fit different numbers of players. However, in some scenarios, the amount of monster power increase per turn is also adjusted by in-game events - ratchetting the overall pace of the opposition up or down.
The town deck as health
Shadowrift is a cooperative game. The town deck acts as hit points for the team - when a monster attacks, one of the people in the town deck is removed, and replaced with a corpse card. If the entire town deck is corpse cards, then the players have lost.
The interesting part is that each turn, a shared hand of cards from the town deck are layed out. This gives players additional verbs, such as using the town blacksmith to do additional damage, or using the town stonemason to help build a wall. However, when the town is mostly corpses, the shared hand is mostly corpses, which means that few of these "get help from a villager" types of verbs are available. This is a self-reinforcing loop, where initial failure leads to more failure.
In this diagram, I only have two different kinds of townspeople, but in the real game there are lots of types.
The player deck
The player deck mechanic is broadly similar to deckbuilding games (Dominion-likes) generally, and trading card games such as Magic: the Gathering before that. I am not a M:tG expert, but I understand that the ratio of land to other cards is the starting point of Magic deckbuilding. In Dominion-likes, you have several mana-like currencies, and so you have several ratios, and during play you can add and occasionally drop cards from your deck, so that your deck's ratios will change during the game.
The initial ratio is fixed, and it is very prowess-heavy. I am not sure what the correct long-term ratio is, but I believe it is far more balanced than the initial ratio, or maybe even extreme in one of the other directions. This gives the beginning game a different quality than the middle game.
Also initially, you draw a hand of five cards each turn, but some purchaseable cards have text something like "you get effect Foo and also draw another card". As you add those cards to your deck, your expected hand size gradually increases, and they self-synergize. If ten percent of your deck has these "and draw another card" effect, then you don't have 5 + 5 * 0.10 = 5.5, it's more like 5.55, because you might, in the course of drawing another card, draw a "and draw another card" card. The self-synergy can get much stronger. When half of your deck consists of these "and draw another card" cards, then you expect to draw, on average, 10 cards per turn, and the occasional explosive, cascading, enormously successful hand size, makes it feel like anything is possible.
There are two ways to get these "and draw another card" effects - heroism cards and might cards. Heroism cards are awarded for killing monsters, and so they create a "success leads to success" virtuous cycle effect.
In general this very successfully achieves an "engine building" feel, which is of course characteristic of deckbuilding games in general. More specifically, the typical dynamics of the ratios creates characteristic differences in feel between beginning-game, middle-game, and end-game. This is how the game mechanics can tell a story, or more precisely, a family of stories, including some where the player's initial success leads to more success, eventually leading to winning, and others where the player's initial failure leads to more failure, eventually leading to losing.
I do not have a diagram I am pleased with capturing this aspect of the game. This diagram that I do have is flawed in many ways.