Solo play is a relatively new phenomenon in board gaming, and why shouldn’t you enjoy a game any of those times you can’t find enough people to play it? Yes, there are solo games, such as Final Girl, but there are way more multi-player games, so sticking to solo alone is rather limiting. Before I start to take a look at mechanisms involved in ‘solo-ifying’ a game, I’ll note that other reasons for playing multi-player games by yourself are practice and exploration of the game, especially handy when learning a new board game.
Let’s start with the absolute simplest approach: play two hands. This can be a chore for competitive games, but can work reasonably well for cooperative ones, unless a major component of the game is limited communication between players. For example, it’s adequate for Holding On: The Troubled Life of Billy Kerr where the small amount of extra knowledge you gain is countered by the lack of discussion, but not so good for Sky Team, in which seeing both sets of dice makes the game far too easy.
Next on the list is the ‘play a certain number of turns and beat your previous high score,’ perhaps with a side-order of ‘if you run out of time, you lose.’ As just one example, this is the default solo mode for the base game of Terraforming Mars: you play 14 generations and win if you’ve managed to hit the stops on the terraforming tracks before that. This mechanism works well for games where there’s zero interaction between players, or close to that; Terraforming Mars does have a few cards that can affect other players (taking resources from them, or gaining from some of their activities as well as your own), but I’ve never felt that to be a major component of the base game.
Your Artificial Friend
The next step up is something that behaves as another player would, and this is where we’re starting to get into automa territory.
The term, the Italian for ‘automaton,’ was popularised by Morten Monrad Pedersen, whose company, Automa Factory, has been responsible for the automas of a good many Stonemaier Games products, among others. It’s essentially an automated mechanism that affects the games state as if another player was present—a cardboard computer, if you like. (I’ll give a concrete example later on, in the section on fan automas.) Pedersen says: ‘The philosophy behind my approach is that playing the solo version of game should feel like playing the multiplayer game. This doesn’t mean that every detail should be the same, but the soul of the game should be intact. Thus you should strive to avoid cutting out important parts of the game, and the solo player should face roughly the same choices and the same win-lose criteria as the multi-players.’
The important parts here are that the automa should affect your play as human opponents would, but the thing itself needn’t play like a human to achieve that.
Some are so simple they barely qualify for the term—I mentioned in my review of Sanctuary that the solo play mechanism is pretty much as basic as you can get. In Sanctuary, the only interaction between players is taking tiles from the shared display board, and a pile of 18 numbered tokens behaves as your opponent in the game’s solo mode; after each of your turns, you take a random token and reveal its value, removing a tile from that slot on the board. If you turn over the final token before achieving any of the the standard game winning conditions, you lose. It’s not 100% like playing against another player, because tile removal doesn’t happen exactly as it would in a two-player game, but it’s pretty close, which is why I would term it an automa by Pedersen’s definition.
Forest Shuffle has a similar display space interaction between players, and the solo mode, provided in the Exploration expansion pack (shame you have to pay extra for it, but at least—unlike some—it’s not expensive, and the box does include a bunch of species and cave cards), uses a deck of 20 automa cards with actions to remove and add varying numbers of cards from/to the clearing. This time, though, you play until the normal game ending, revealing a winter card (just one rather than the usual three, but there’s no real reason to not include the others too). Forest Shuffle Exploration also includes a few ‘challenges’ to make the game a bit more strategic, e.g., in the cards at the bottom of the picture above, have a bird on every tree in your forest, or four or more bats, along with a minimum score target.
Wingspan’s automa, one that Automa Factory provided, is a big step up from those. Here the automa is much more clever, participating in almost all aspects of the game as a real player would—drawing cards, laying eggs (or rather, taking points for them since it has no player mat to lay them on), taking food resources from the feeder, etc.—and as the game progresses, some automa cards may be removed from the deck, changing its behaviour across the game. It can actually end the game as the winner (and with me, that happens more often than not). As with Forest Shuffle, the automated opponent draws automa cards to determine the action it will take, with sub-rules for which card or food to take on some turns; there are a few tweaks for dealing with when players would interact (e.g., your ‘pink powers’ are triggered every turn since there’s no notion of the automa activating a particular bird type). There’s also an intriguing multi-player cooperative mode in the Oceania Expansion, where players compete against the automa.
I did mention that solo play is a good way to explore the game, experimenting with strategies. It’s worth bearing in mind how automa behaviour differs from that of human players. For example, because the base Wingspan automa doesn’t do anything with gained food, you might blithely play cards which would give an opponent resources—perhaps a mistake in a real game.
I have to say my favourite automated opponent comes from the game Spooktacular. Here, the automa is one of the standard monsters players can choose in the game, the Killtron 3000. Killtron’s behaviour is a ‘program’ created using a trio of action cards; in normal use, the player executes this program (i.e., does whatever each card says), then disposes of the first card, shuffling the rest across and adding a new action to the end from cards in their hand. When used in automa mode, the action cards are split into three decks and on each turn, the top card of each deck is revealed and acted on, and there are a few additional rules for dealing with ambiguities and choices a real player would make. (This is an awful lot like how the bad guy behaves in Final Girl—that’s practically you vs an automa.) One of the especially nice things about Killtron is that its automa can be added to any game, so it’s not just for solo play.
On that note, the Root ‘clockwork’ expansions add a host of rules and heuristics to enable any of the standard factions to be automated and added to a game. I also know there’s a Terraforming Mars: Automa expansion, but I’ve yet to try it—and I’ll explain why later.
Now, what if there’s no automa for your favourite game?
Fan Automas
Take a look at the BoardGameGeek—you might be surprised at what you can find…
Jan Van Regenmortel’s nicely designed automa for London can be found in the game’s BGG pages. Called ‘AutoMartin’ (after designer Martin Wallace), this offers a deck of automa cards to print out and a guide for their use. Take note of the way this automaton chooses which card to take—this is essentially how most automas work, by simulating the decision process a human opponent would make via a series of if-then choices. The block of text in the middle of the first card above details in shorthand how AutoMartin chooses a card to pick up: first, look at what’s on the development board and if there’s only one card, take that; else look at the maximum prestige gain of all cards on the board and if there’s a single winning card, take it; else look at the poverty gain of those cards that tied for maximum prestige and if there’s a single minimum poverty gain one, take it; otherwise fall back to drawing from the deck. In contrast, the last card above merely indicates that AutoMartin takes the top card off the deck. Besides this decision flow, the card also shows at the top of the card what AutoMartin gains, and at the bottom what other action it takes. This automa makes for a pretty good opponent, and though I’ve come close, I’ve yet to beat it.
One of the more sophisticated automas is Andre K’s A.R.N.O. for Ark Nova, which provides a much more interesting game than the default solo mode of playing six rounds; additionally, multiple A.R.N.O.s can be added to a game, too, solo or otherwise. More complex automas like this do come with additional overhead to manage them, though. There are a couple of other unofficial solo modes for Ark Nova which seem a little simpler: Haven and Hanna, though I’ve not taken a close look at them yet.
Third party solo modes exist for many other games but I’ll leave you to explore to avoid this article devolving into a series of BGG links. Well, apart from mentioning one more: the official Forest Shuffle Exploration automa is more or less Anna Triebert’s solo mode, so it seems Lookout Games liked it a lot, so much they’ve even called the automa Anna. It’s gratifying to see publishers engaging with fans in this way. (Incidentally, if you’re a tightwad or want to explore before purchase, you can still grab the Print and Play version from BGG.)
Automated Automation
One small downside of these alternative automas is having to print out the cards, but some of the card drawing mechanisms have been implemented as web apps, such as Phil Range’s AutoMartin; and Stefan Seifert’s A.R.N.O helper looks after dice rolling and various aspects of the automa state as well as card drawing. These are how I use both these automas all the time as it’s so much more convenient than fiddling with cards.
There are many more automated implementations for various automas, official and unofficial, but as with automas themselves, I’ll leave it to you to hunt them down for the games that interest you.
Having mentioned apps, now I’ll admit why I haven’t got Terraforming Mars: Automa yet (besides its cost, that is). The PC implementation of the game is actually pretty good, and when I want some solo play, I’ll fire that up instead of cracking open the physical box. While the game’s AI opponents aren’t great strategists, they’re good enough for my Mars quick fix. Having said that, the downside of this particular implementation is the low number of digital expansions, so I’m stuck with the base game and Venus Next.
Similarly, the PC version of Wingspan, offers AI opponents, but it also includes the automa. I do still tend to use the former, since it’s more like playing with real opponents.
As a counter example, I love Marine Worlds, and the Ark Nova app, gorgeous as it is, doesn’t include that… yet. So, if I want my reef dwellers, it’s the physical game for me, and A.R.N.O. is way better than the official solo mode.
Conclusion
Solo play is a feature many games offer, and something a lot more should. Sometimes, a single player mode is bolted on as an afterthought, and sometimes it’s well-designed. Either way, talented fans may have created other solo play mechanisms.
The more complex of these mechanisms involve a cardboard computer, referred to as an automa, and while some of these can be very nuanced, I think it’s still generally the case that other human beings are better—it’d have to be something very special to make me buy a multi-player game just for a solo mode. Having said that, the occasional stand-in for a real opponent can be useful, and the ability to increase the number of players of any game with an additional robot is interesting.
And finally, I do find that where decent software implementations of the boardgame exist, their AI players can be good alternatives to automas.
About the author
When not playing boardgames or blogging about them, L.N. Hunter keeps himself occupied writing fiction: a comic fantasy novel, The Feather and the Lamp, sits alongside close to 100 short stories in various magazines and anthologies, and on websites and podcasts (see https://linktr.ee/L.N.Hunter for a full list). L.N. occasionally masquerades as a software developer or can be found unwinding in a disorganised home in Carlisle, UK, along with two cats and a soulmate.
