Posted on

Anatomy of a Line: Keyflower, Part One — The Original Game

Figure 1-1: Arriving in the Keyflower boats

Richard Breese has been designing and publishing games since Chamelequin (1989), but he’s better known for the “key” games that began with Keywood (1995). His best received and most popular game is actually a co-design, Keyflower (2012), with Sebastian Bleasdale. It’s no surprise that it’s led to two supplements and two reimaginations of the game, the newest of which, Key Flow (2018), finally got to American backers in December. As is usually the case, looking through a series of designs like this can offer intriguing lessons about the art of game design.

This first article focuses on the original game, while a follow-up in two weeks investigates how the expansions and reimaginations have changed the game.

Keyflower (2012)

remains a popular game in part because it does a masterful job of combining three major mechanics: worker placement, auctions, and resource management. But it did more than that, also innovating all three mechanics in notable ways.

Of the three major systems in Keyflower, the worker placement is the closest to the norm: you place workers in action spaces, you get rewards, and you (sort of) block other players from using those spaces. However, it still has several (small) enhancements.

Figure 1-2: Keyflower workers

Worker Placement: Escalation. The workers in Keyflower can be escalated. Rather than blocking an action space entirely, a worker makes it more expensive for the next player to place workers. Usually three worker placements can occur, costing 1, 2, and 3 workers respectively. This escalation variation has also shown up in other recent worker-placement games, but it’s still relatively rare.

Worker Placement: Restrictions. Those escalated worker placements are also restricted, which is a more innovative mechanic. Each worker comes in one of four colors: red, yellow, blue, or (rarely) green. Once a worker of a specific color has been placed, all escalated workers at that same action have to be that same color. This introduces intriguing possibilities for both blocking and brinksmanship, as a player wants to ensure that his favorite actions are linked to his favorite colors and he can try to lock up an action by placing a worker in a color that’s available to him and rare to the other players. (Green is particularly popular for this purpose, due to its scarcity.)

Worker Placement: Ownership. Players don’t own their meeples, like they do in most worker-placement games (which is why Keyflower can use meeple colors for other purposes). So who gets the workers after they’re placed? They go back to the owner of the building. This is a particularly clever mechanic because it takes a trope that was traditionally codified by rules and instead makes it organic: where many worker-placement games give a player a reward for having a valuable building, by requiring payment of some type, in Keyflower the workers themselves become the payment; there’s then no need to remember to pay, because a player will just naturally scoop them up at the end of a season.

When worker-placement games became popular starting in 2005, they put the nail in the coffin of two mechanics that had been central to the young eurogame genre of the ’90s and early ’00s: auctions and majority control. Certainly, both mechanics still exist, but they’re much less common than they used to be. Despite that, Keyflower isn’t afraid to use auctions, and it isn’t afraid to innovate them either. 

Figure 1-3: Keyflower auctions

Auctions: Simultaneous.The most notable element of the Keyflower auctions is that there are actually multiple auctions running simultaneously: one for each tile that’s available for purchase in a round. Each turn, a player can make a bid or increase a bid on a single item, and play will continue around the table until everyone is satisfied with their final bids. In fact, with the bids scattered around the table in so many places, it looks an awful lot like majority control, underlying how similar these two mechanics are. The big thing that keeps this mechanic on the auction side of things is that players get their bids back if they lose a bid (and can even rebid with them, if they do so carefully).

Auctions: Restricted. The bidding is actually done with those blue, yellow, red, and (rarely) green meeples. And, the same restrictions apply as with the worker placement: once one player has bid on a hex with a certain color, all future bids on that hex and raises must be done with that same color.

Auctions: Integrated. Here’s the true heart of Keyflower, and what makes it brilliant: the worker placement and the auctions are entirely integrated. You’ve already noticed that they use the same meeples. That’s because on any turn, a player can use meeples to activate a worker-placement action or to bid on a new tile (and thus ownership of its worker-placement actions). In fact, players can even place workers on the tiles being bid for! A less cleverly designed game might have contained both auctions and worker placement, but wouldn’t have figured out how to make them seamless parts of a whole. The fact that they are in Keyflower is what makes everything work.

Finally, Keyflower contains some resource management, and as with everything else in the game, it features just enough innovation to make it an original. 

Resource Management: Geographical. Keyflower contains two different types of resources, both of which are used to upgrade tiles. Skill tiles are the sort of resource that you’d find in other games. They’re handed out to the players, who then hand them back in when they’re ready to use them. The octagonal resource tokens, though, are something else. When they’re generated, they’re placed on the worker-placement spaces. They then have to be moved to the correct action-hexes, following the geography of the game board, to enable upgrades. Though there certainly are other games that maintain resources entirely on the board, forcing players to move them to the correct place, such as Klaus-Jürgen Wrede’s Mesopotamia (2005),  it’s another mechanic that’s unusual enough that it stands out in Keyflower.

Figure 1-4: Keyflower game elements

And that’s not even everything. Since the worker-placement actions are all on hexes that players arrange to form a village, you have a bit of city building too. Since the upgrades of those hexes improve their capability, you have a bit of technology. These elements deserve a bit of additional comment:

City Building: Limited. The city building of Keyflower is minor yet integral to the game. It’s integral because it exists at the intersection of all the other elements in the game: you win city hexes in auctions; you activate them in worker placement; and you manage resources to upgrade them. It’s minor because despite the city building being geographically based, there’s not much to that geography, other than the movement of resources. In some way, the city building of Keyflower feels like the game’s missed opportunity … but how much can you complain when the game has so many other intricate, complex elements?

Finally, since those worker-placement spaces can be complementary, you have some engine building. And finally, since some of the end-game victory points require specific combinations (or collections) of tokens, that’s set collection


Keyflower revisits a lot of classic eurogame mechanics, particularly three favorites: worker placement, resource management, and auctions. However, it’s not even close to a retread of anything before it. Instead it achieves its success in two ways: by innovating how those mechanics are used and by integrating them together into a whole that’s so cohesive that it’s a bit hard to see the individual elements.

So how do you improve on such a well-designed game system? In my opinion, some of the expansions and reimaginations have managed to do so and some haven’t, but that’ll be the topic of my next article in this series.

Liked it? Take a second to support Shannon Appelcline on Patreon!

The original article can be found on the great Mechanics & Meeples