The Worker Placement Track

In this designer diary, I'm going to focus on my favorite mechanic in The Transcontinental. It's the double-sided offset linear worker placement track. 

Origins of the Mechanic

The one unifying thing in every iteration of The Transcontinental is some sort of central track in the middle of the board. But this wasn't always a worker placement track. In early iterations it was dice manipulation. Then it became a combination of worker-placement and deckbuilder, with the board becoming a central track where you reserved a space where you'd take actions, and the cards in your hand representing the resources you could then produce or unload. 

There were cards on either side of this central track, with costs there to develop the cards and earn corresponding deckbuilder cards. At some point during this design, I made these cards twice as wide as the track spaces. Offsetting the two sides from each other was the other change, which immediately seemed to present some interesting dynamics, because with two spaces for each tile, you could block another player's best option, but not all their options. 

But the final vision for The Transcontinental really took shape when I threw out the deckbuilder element and moved the actions onto the tiles themselves. Throwing out the deckbuilder element was a difficult step to take because I had put months of design and countless playtests into that version, but almost as soon as I started playtesting the new system I knew it was the big step forward. From that point forward, everything else became making the rest of the game around that mechanism. The Transcontinental isn't the first game to use a linear placement track - there are a few out there, including one of my absolute favorites, Caylus. But in my opinion linear worker placement tracks are a somewhat underused mechanic, and I hope that players find The Transcontinental to have a novel spin on this. 

How it Works

One of the things that I like about both deckbuilders and dice allocation games is comboing. This is always one of my favorite game mechanics: figuring out the best of a limited set of options, and the order to play them in. The comboing element in the deckbuilder version was fun, and the worker-placement version needed comboing as well. 

The combos are based around two properties of the action turns: firstly, on each turn you take one action on each adjacent tile; secondly, each tile has at least two actions on it: on the front, this one one develop action and one default action, and on the back of the tile (after it's developed), it's that same default action, and usually a produce action (to gain more resources), but occasionally a more powerful develop action, or sometimes both. 

So I'm going to give a few examples of how this comboing works: 

Here, Red wants to complete that Lumber Mill, but they have only one steel on the train, while the Lumber Mill requires 2 steel. Can they complete this using their one action on each side? Yes! On the tile across the tracks, there's an Acquire Action (A). Red can Acquire one Steel, and then immediately Develop (B), to unload both required steel. 

Yellow has a different opportunity: With one Navvy on the train and a tile that requires a Navvy, they could easily Develop, and unload the one Navvy necessary to finish that city. But there's a way to get even more out this turn. Instead, Yellow can use the Portage action (C) to unload any 1 of their resources anywhere, thus completing the City. Because they still haven't taken an action on that side of the tracks, they can immediately Produce using the city.  

One big advantage to this system, where each tile is actually a choice of two actions, is it presents a lot of adaptability to players. There are plenty of times that a player is thwarted by another player, but often the game leads an avenue for a player to still do something different and useful with their action. 

There are a couple other wrinkles to this. When the train reaches the West end of the track and completes the railhead phase, it travels back East. When you take an action, you remove your telegram marker from the board. But you can choose to take no actions, and leave your telegram on the board for when the train is travelling back East. Additionally, one of the actions allows you to place a 'bonus' telegram immediately... this can go ahead of the train in the direction it's currently heading, or behind the train, to reserve a space for the way back. This results in the worker placement track also being very shifting and dynamic. 

Variability

Another key element of this system is the variable setup. Because you're taking one action on each side of the track, the tiles that are across the board from each other matter immensely, and any change to the tile order drastically changes the combos available during the game. 

But each region has the same set of tiles every time (simply in a different order), so you can expect the same overall strategies will be available. But the tile placement is going to determine how those strategies interact with each other, or can be aided by the actions on the other side of the track. 

But all of these actions create some challenges: particularly, the number of choices! On an empty board at the beginning of the game, players have 9 spaces where they can place their telegram, and many of those spaces have two adjacent development tiles. And by late in the game, there are 25 spaces where players can place their telegram! I was often getting an reactions where first time players spent the first round getting their heads around the gameplay, and then had this 'what now?' moment. It's challenging to see how you're going to turn the gameplay into a winning strategy. One option was to say that this is simply part of the experience, working through the strategies and the way that the different developments interact. But the other option, the option I went with, was to try and pull players through that stage of discovery relatively quickly.  

Directing Choices

This is where the investment symbols come into play. I could write a whole other post about the investment symbols, which match objective cards that players get at the beginning of the game. But for now, I want to talk about how these investments solve the problem of too many choices: they draw players' focus to specific development tiles that match their investments. For example, with this Industry investment, Red could focus on the Coal Mine. That really reduces the number of choices for Red: Place one telegram marker adjacent to the Coal mine, and one on a source of iron (of which the best choice is the port supply spot). So now, choices for Red are far simpler (but still interesting). 

But these investment symbols go a little further than that in just suggesting a spot here and there, because each symbol - there's six in total - highlight a little chunk of the economy in the game. The feather encourages you to build cities that produce tourists, parks that accept tourists, and then start shipping developments down there where there are some big points available. The Beaver symbol that Red has here will lead to developing also developing Lumber Mills, but later in the game it will direct Red toward developing farms in the prairies, which need the wood Red is now able to build. 

Simply following the investment strategy you're dealt at the beginning (actually you get a choice of 2) won't be enough to win the game. You'll need to be opportunistic as well. And you'll get a second investment card midway through the game (again, a choice of 2). The thing I really like about the getting a second, mid-game investment card, as someone who plays this game a lot, is that it often requires strategy shift midgame, which I find much more interesting than simply continuing to optimize my early game strategy. 

Solo Mode

Solo mode has been one of the last design challenges with this game. It took me a while to commit that I'd include a solo mode, because I wasn't sure if I'd be able to find a style of play that captured similar decision-making to the multiplayer game. I identified three key types of interaction that I wanted to simulate in solo-mode:

1. Track-based worker-placement decisions: In multiplayer, players never have a choice of the whole track: they are always at least partially blocked by other players, so they're choosing from a limited set of spaces. As well, because bonus telegrams can be placed mid-round, players never know exactly how a round is going to play-out. So you need to prioritize: what's important, what desired outcomes have several different options., etc. 

2. Train board management: In multiplayer, players are usually keeping an eye on what other players are planning, determining if they'll be able to load their resources to the train or not. Another player loading navvies might block you from loading tourists. But if you manage to get even one tourist on the train first, that might instead block them. 

3. Railhead bidding: In the game, players blind-bid resources every round to advance the railhead and score VPs. 

Nice-to-have features: 

High replayability: keeping the game different in feel from one game to the next. 

Absolute win conditions: Personally, I'm less a fan of solo games that ask you to compete against your own high score, and prefer to have a clear win objective that I'm going for. 

Minimal new mechanics or components. I don't want solo players to need to learn a lot of new rules, or introduce new components that would only be used in solo mode. 

Throughout the game, there are 'investment' symbols on all of the developments (those building-shaped icons), as well as on ally cards. These factor into end-game scoring: players will need to build developments that match their secret investment cards to score points from these cards by the end of the game. But these investment icons also serve another purpose: they teach players the economy of the game. A player who has, for example, the industry investment, will look for that symbol on the board, and find coal mines, and also ironworks, and by building these they'll understand the synergy between those two types of developments. Meanwhile, a player focusing on the tourism investment will develop cities, and parks, and ship tourists from one to the other, scoring points and developing those investments. 

My eureka moment with the solo mode came with the realization that these symbols, already in the game, were the perfect system to hang the automa on. If the automa was using those symbols to choose its actions, then this would be create an interesting, highly replayable automa that would loosely mimic a human player. One game, it would be focused on the tourism segment of the economy, and the next game on building farms and delivering grain back to the cities. And because these symbols were also on ally cards, then the ally cards would be a great way to 'run' the automa.

So I started with some standard deck-builder-type mechanics around the automa. It wasn't enough to use just the ally cards. I needed three different cards (themed around the syndicate) as well. With this sort of deck approach, the automa's actions became predictable, but not too predictable, in the way that playing against a real opponent is: you can know that your opponent is prioritizing a couple strategies, like tourists and grain, but you won't be able to predict exactly how it will execute that strategy, or whether it will do something that feels kinda dumb or really smart. These deckbuilder mechanics determine everything from what actions the automa performs, to where it places its telegram, to what it unloads at the railhead. 

Choosing victory conditions for beating the syndicate was another challenge. Simply scoring more points than the syndicate tended to be too dependent on the speed at which the railhead advanced: slow the railhead's progress down, and you had lots of time to come up with a strategy that would outpace the syndicate. So I ended up going with three victory conditions, all of which needed to be met in order to win: 

1. Score more points than the Syndicate (by collecting more VPs through the course of the game). 

2. Do not let the Last Spike progress into the prairies (which means keeping the railhead moving forward at a steady pace). 

3. Complete all of your investment cards (by completing developments and acquiring allies that match). I can tweak the difficulty a bit by having the player finish with 2 investment cards for an easier game, or 3 investment cards for a standard difficulty. 

With three win conditions, this raises the question of what should happen if the player fails on #2 midway through the game (the Last Spike is the only win condition that can be failed before the end of the game). All 3 win conditions must be met for a victory, satisfying 2 of 3 is an unremarkable defeat, and satisfying 1 or 0 win conditions is an ignominous defeat. So if players do fail the Last Spike condition early, they can either accept an ignominous defeat, or attempt to play for a forgettable defeat. My win rate against my automa is less than 50%, Although most of my defeats have been of the unremarkable variety. 

I usually feel pulled in a lot of different directions, trying to prevent the Syndicate from getting to far ahead of me in points, keeping the railhead moving forward, and checking off my investment objectives. 

I'm going to be revealing the artwork for the syndicate cards during some solo game demos at awSHUX between Oct 16 and 18. If you're interested, you should sign up for this free online event and come hang out with me while I attempt to defeat the Hugh Allen Syndicate. 

Tuesday, Sept 21: Origins

In this post, I'm going to talk about the early history of this game. It's been about 3 and a half years of development in all, and it was fun to look through archives of old versions, most of which have been harvested for other designs and ideas. I'll focus less on how it got to the final iteration, but moreso what the early process was like. 

Designers often ask one-another what comes first for them: theme or mechanics. But I think for a lot of designers, we're constantly coming up with both and it's finding a matter of fit. I keep a list of themes that interest me, and a list of mechanics I'd like to explore, and occasionally I find matches between them. 

But The Transcontinental started out as mechanics-first. I had this idea for a dice-allocation game, in which there was a central track, along which little containers of dice would move, and players could take actions that would manipulate the values, and add or remove dice. I knocked out a really rough protoype, and took it down to Sentry Box, where our local design community gathers (or did gather prior to the pandemic). 

Paul Saxberg sat down with me and we played through the mechanics I had, and we both agreed that there were some interesting things there worth exploring. Then he said, "You know, if you're looking for a theme, this would be great as a Canadian railway game." I saw immediately what he was talking about: the central track, with cargo moving up and down it, perfectly fit Canada's railway story, which, unlike the dueling routes and companies in other countries, comes down to the building of a single line. 

So the earliest versions of the game were based around these dice allocation mechanics. In my archives I can find only one physical card from this generation of the game:

But eventually the dice-allocation idea fell by the wayside. But a few ideas that remained constant from those earliest iterations was that there was a central track that expanding westward, while also developing things beside the track. 

Very early on, I realized that I wanted the economy of the game to be about more than just strictly the resources necessary to develop the railway: the settlement of the West and its grain-production for the cities back east seemed to be an important part of the railway story. So too did the development of the National Parks. So I had three different types of people: Navvies (navigational engineers), settlers, and tourists. Eventually, settlers were removed from the game as a resource. 

One feature of these cards that ended up staying around was this: each space had a unique cost that would need to be paid before the railway could advance through there. As well, this development would cross various regions, with the costs and and opportunities changing to reflect the region of the country. That seemed to be a critical narrative point. 

One early system had a number of cars that could hold resource cubes, and then had spaces for players to place their coloured pegs. These coloured pegs would essentially be a system of reserving actions: putting a yellow peg in the fifth space on a train car, for example, would allow yellow to take an action when the train got to the fifth space on the track. Looking at it now, I'm actually surprised at how many similarities there are between that system and the final system, as I diverged a long, long way from that system before coming back to something kinda similar. 

At this time, overall I was trying to simply not get locked in to any single idea, and just take anything I liked and throw out everything I didn't like. 

I've got a great group of local designers and playtesters, and with a weekly playtesting session (sometimes twice a week), I could come home from a session on Monday night, make some harsh decisions about cutting stuff, spend some time prototyping, and by Monday have something pretty different from the last week. Not happy with the system overall, I went in an entirely different direction at that point, which was to make the game a deckbuilder. In the deckbuilder version of the game, players would acquire cards when they completed a development. These cards could be used for a corresponding reward. For example, developing a lumber camp in the Shield region would allow you to acquire a card that you could play on future rounds to load more wood in that region. Cards were double-edged so each could be used in a couple ways. This double-sided approach definitely influenced the 'choice within a single tile' system that made it into the finished game. Players were making interesting choices between developing tiles, or other actions. It existed as a deckbuilder game for at least six months. But it still wasn't playing the way I wanted. 

Now this is starting to look a little familiar: when I moved away from the deckbuilder, I combined it with some of the earlier ideas and had something kinda similar to the final product. The double-sided track emerged here. So too did different VP rewards for contributing the most, or a lesser amount. 

I developed ally cards around the same time, though they came out of a different design challenge: I knew that there were parts of the Canadian railway story that I couldn't entirely ignore: changes to Indigenous and Metis lifestyles and rights, and the treatment of railway workers, especially Chinese workers. And I tried a few different mechanics, such as having treaty negotiations as a mechanic in the game. But that never felt comfortable. I decided that what was needed was to give these groups a presence in the game, without editorializing on it. So ally cards were a way to do that: to show the vast range of people affected and involved in the railway story: from the company presidents, and politicians, through the surveyors and workers, to people outside the railway who were affected: those Indigenous and Metis groups, early ranchers, entrepreneurs, and others. 

Even early on, I knew that Ally Cards needed to factor into the end-game scoring, but it took a long time to find the right fit: this version had the concept of 'retiring': when you completed a development with a matching symbol, you could retire the ally for Victory Points, at which point you could no longer take any actions. Those early ally-cards were multi-use, so it was an interesting choice between losing future actions to score VPs. The retiring system ended up getting removed, but I kept an element of trying to pair a development with an ally, but now it happens during end-game scoring, rather than during the game. 

At the same time, I was still working through how the railhead needed to function. It always felt good that every round, the railway would advance west, opening up more of the board. The version on the left required a certain amount of each of the three railhead resources: wood/steel/navvies. But I replaced that with something close to the version on the left, where there's just a flat cost for a given tile, and instead introduced the priority resource, where each round one resource-type is doubled. (railhead tiles)

I also went through a lot of different approaches for how resources were stored on the train. Some versions had train-cars with a grid of cube spaces; players could unload one line or column. And I explored having different, thematic approaches to different types of cars. A car where you stacked resources, for example, would be last-in, first out. A hopper that unloads from the bottom would be first in, first out. 

But the simplest, best approach to that was to have them all work like a column. You could remove any cube from the column, and everything above it could be unloaded, at its owner's choice. This is very similar to the final mechanic, though the columns have been turned into rows, but function the same: cargo always slides forward, and any resources behind the first resource removed if their owner chooses. 

This was around the time that I submitted the game to the Canadian Game Design Awards, which it won. Despite the win, there were some things I still wasn't happy with, the largest of which was the way that new cars get added to the line. The CGDA version had a system where players could take an action to contribute coal towards a particular type of car, and when enough coal was added for that type, then a car of that type would be added. But this seemed like two steps too many. I introduced the Coal Tender, which was just a special car at the front of the train which, when full, would automatically trigger a car getting added. 

Once I made that change, the game felt pretty complete, except for balancing values and ally powers, and of course diving into the iconography and visual design of the game. 

Tuesday, Sept 14: Iconography

The iconography for The Transcontinental has gone through a lot of iterations, and in this post I'm going to talk about my choices in designing the iconography for the game. 

The action iconography needed to serve two purposes: show actions available to players on a track space, and show actions available when playing a card. 

Each track tile gives a player a choice: Choose one of 2 or sometimes 3 actions available on a card. 

Putting the elements together in what looks like a single, unified graphic, instead of disparate parts, separated by background, means that a player will never see one without the other, and should never miss that this is an 'OR' decision. 

Resources are almost always used in the game by placing cubes on slots with matching symbols, so identifying resources as square was an obvious decision, and from there it was pretty natural to make all actions circular: circles invite us to interact with them in a way that few other shapes do, like a button that says 'press me'. A variant of this is the Develop action, which has a circular base, but then also has 'wings' to hold whatever resources are required for development. 

Less obvious was the design and colouring of these icons, but they emerged as I began to put together the colours of the rest of the game. My approach was to keep all of the colours on the background relatively muddy... colours that were neither high nor low in value (meaning not too black nor too white), nor high or low in saturation (meaning not very bright or very grey). By doing so, this left me to use high contrast, and bright colours on the iconography, and know that they would pop against the muddy background. (I'm going to have another post specifically about artwork, where I talk more about the colour-palette of the game. Watch for that soon.)

I also went through a lot of different iterations for what each action should depict, but I eventually settled on each depicting an object: a cash register for Acquiring, a telegraph knob for Telegram, a canoe for Portaging, a timber-frame for Developing, a coal shovel for Resupplying, a block and tackle for Developing, a surveyor's scope for First Player, and a scale for Trading. The most difficult object to choose was one for the ally action: in previous versions I had a handshake, but to my mind this broke the otherwise consistent 'object' approach to the iconography. In the end, I settled on a point blanket. The presentation of a point blanket was traditionally, and in some circles, still is, an important gesture of fellowship; and while I'll concede this iconography is a little esoteric, it doesn't matter within the context of the game, because this is the most obvious of any of the actions: it simply requires a player to take a card with that matching symbol on it.  

Adapting these symbols for the ally cards took a little more creativity. I wanted to keep the stack approach, but a lot of the ally cards describe actions that are exceptions to how that action would normally work. This led to the development of an 'arch' form of each icon, that could contain additional information below the standard icon. As well, additional keywords were needed in addition to 'OR', such as 'THEN', and 'IN', as well as a way to represent the regions in the game. Critically, each icon still fits into a vertical stack, so they are read in the same way as the icons on the development tiles. 

Victory Points were the easiest icon to decide on. It just always felt natural that maple leaves should be used for VPs. In fact, the earliest implementation of this, before I had a score track on the board, consisted of players using Canadian pennies with their maple leaves to count score. In earlier iterations, I had a number of different maple leaf icons, with different colours, so that low-value victory point icons would be yellow, then orange and red for increasingly higher points, and finally a purplish crimson for the highest point-value icons in the game, like those found at the Western end of the railhead. Ultimately though, I determined that having one consistent approach to the colours on the icon made more sense, especially since the icons already occur in a lot of different locations around the board. As well, earlier versions were backed by a circle, but that went against the idea that only actions use circles in their iconography.