Scythe accessibility teardown

Scythe (2016) – Accessibility Teardown

Game Details
NameScythe (2016)
ReviewMeeple Like Us
ComplexityMedium [3.44]
BGG Rank17 [8.14]
Player Count1-5
Designer(s)Jamey Stegmaier
Buy it!Amazon Link

Version Reviewed

English retail edition


We liked Scythe quite a bit. That’s unsurprising for a game that currently, at the time of writing, is living comfortably within the BGG top ten. That doesn’t guarantee a favourable writeup here – we are driven by our own erratic appetites more than we are those of that make up the Boardgamegeek faithful. It’s not a bad way to bet it though.

The teardown structure we use for our site means that we get a lot of flexibility when it comes to reviews. When we reviewed Scythe we could focus on the game and not as much on the tangible interfaces of play. That’s what we do here in our teardown. While the user experience of a game is an important contributor how it plays it’s usually only lightly coupled to actual enjoyment. Where the interface gets in the way of a fun time, it’s well within the scope of a review. When it doesn’t, we talk about it here.

We get to delve deeper in other words and we’re going to have plenty of cause because the Scythe user interface is one that is, in many ways, exemplary. A lot of things have been done very well here, and reduce the accessibility burdens that might otherwise be experienced. That doesn’t mean that Scythe is necessarily going to be an accessible game. It’s just a good deal more accessible than the same game would have been had it not been so well designed with regard to its affordances.

So, let’s get to it.

Colour Blindness

Each of the factions has its own mech design, its own leader mini, and even its own worker meeples.

Scythe minis with colour blind filter

Right off the bat it would be a strong contender for being completely colour blind accessible because of this, but the colours chosen for the player factions are also, by and large, reasonably distinctive. There is very limited overlap for most of the standard categories of colour blindness.

Colour blindness and worker meeples

This is important because a lot of Scythe involves identification of factional interests at a distance. Workers sometimes share a similar (although unique) silhouette so the additional disambiguation provided by the palette ensures that information is retained and enhanced where possible.

Colour blindness and the map

Individual resources have their own tokens too, and colour is not used as the sole channel of information anywhere in the game.

We strongly recommend Scythe in this category.

Visual Accessibility

While Scythe is a game with a lot of visual state, it’s also one that offers a lot of tactility to the investigation of that state. It’s not perfect, particularly when it comes to board hexes, but it goes above and beyond what I have come to assume will be the case for games.

First of all, each faction has a physical profile to its pieces that can be ascertained by touch – at least with practice and experience. Each faction has a unique leader mini, a unique mech design, and a unique worker meeple design. This means that not only presence of units on the map can be ascertained by touch, so too can their owners. That’s unusual – most games give you a way to determine presence without its meaning.

Leader minis

Each of the resources too has a different form factor, as do each of the structures that can be built in the game.

The faction boards make use of slots onto which mechs are placed – when bonuses apply they are revealed by taking a mech from its slot and placing that mech on the board. While the text of the faction powers is only provided visually, a player can tell by touch how many of the bonuses they have obtained. There are only four of these bonuses and they don’t change with the faction – they are constant and not part of the randomised setup. A player can decide on a favourite faction and learn its special powers, or memorize the four that will be relevant with regards to a randomly assigned faction.

Faction boards

The player boards come with variable costs for actions, but here there’s a genuinely excellent example of accessible design. A cube is used to cover up extra bonuses that may be obtained, and also to cover up costs that no longer pertain as a result of upgrades. This is handled through the user interface of the player board by taking a cube that is blocking a benefit and using it to cover up a cost. Physical indentations are used for this.

Indentations on the player board

As such, indentation or cube conveys important information. If a cube is present in the top row, it is blocking a bonus. If it’s present in the bottom row, it’s covering a cost. If an indentation is present on the top row, it is an additional bonus. If it’s present on the bottom row, it’s an additional cost. As long as the base values for these are known, upgrades are conveyed seamlessly to even a player with total blindness. I can’t say enough good things about this – it’s such a clever piece of design that I hope people steal it outright in future games. Sorry Jamey, but ideas this good need to become universal. It’s not just great from an accessibility perspective, it’s a massive benefit to everyone.

Slots on the player board

This is also carried through to recruit bonuses – upon gaining one of these you remove the cylinder covering up the bonus and so you can tell if it has been applied. You can’t tell what it is by touch, but again familiarity will make that problem less significant as time goes by.

Each of the different markers in the game has its own form factor – victory stars, popularity, power and so on – they can all be differentiated by touch. Their location on their trackers unfortunately cannot, but there is little gameplay information that is leaked by inquiring of this information.

Victory stars on the board

All of this is pretty much fantastic – without this, Scythe would have likely received nothing but a battering in this section given how much information is present in the game. Providing so much the state in this way makes it easier for all players to track what’s going on but actually brings the game into playabilty for a lot of people that might otherwise never be able to play at all. Good design deserves praise.

Unfortunately now we head into the problems, and these are considerable.

Visually busy scythe board

The board is large and visually very busy. The hexes are relatively small and will often be dense with tokens. You might have all of the following on a single hex:

• Several mechs
• A leader
• Several workers
• A structure
• A pile of metal
• A pile of wood
• A pile of wheat
• A pile of oil
• Enemy mechs
• Enemy leaders

A square doesn’t just contain what it can produce – as workers move around they bring with them any unspent resources. Resources that are produced remain in place on the map – they don’t get added to a player’s area. This means they become available for conquest for other players, and you’ll want to ensure that doesn’t happen by bringing them along with you when you move from one hex to another. This means that while a hex can have its contents investigated by touch it can become difficult to actually turn that into a coherent understanding of the state of that hex. That’s particularly true when assessed in light of tunnels and adjacent hexes. As a consequence of the large information payload a hex may convey, and because more things get added as time goes by, the board gets increasingly difficult to visually parse. While it begins by being relatively easy to identify presence, resources and factions it soon becomes a considerable chore even for sighted players. For those with visual impairments, a board which could be close inspected will soon become one where investigation can easily upset game state.

A busy hex

Another issue with the board is that the type of resource produced is indicated only with a printed icon and this icon is small and occasionally poorly contrasted – particularly wheat against the fields and scenarios against the wastelands. For the latter at least there are event tokens that are placed on the hex to indicate that there is an unspent event to encounter, but everything else needs to be assessed visually. The colour and patterns of different types of hexes are not necessarily much of a help here – while a player will likely be able to identify the difference between a lake and tundra it’s more difficult to identify villages versus fields. Given how dominance and flow of movement over the map is one of the most important aspects of the game to manage, this is a considerable problem.

Combat dial

There isn’t a huge amount of hidden information in the game, which is good. However, the hidden information that is present is explicitly that for player versus player activities. Specifically these are the combat cards (which add strength to your side in a battle) and the combat dials (which allow you to spend your power in attack or defence). Covertly playing cards and setting the dial is massively important. The dial at least can be replaced with something like a note in an envelope (or indeed, a piece of paper flipped over at the same time as the other player) but the cards are more difficult since they have differing values. It’s not enough to play a combat card – one must know its strength and also make sure that one is spending the correct cards in combat.

Event card

Event cards present information only visually, but these at least can be narrated by a sighted player at the table. There are only ever three choices, and they don’t come with any flavour text except for the options.

Overall we don’t recommend Scythe in this category, but that’s actually a remarkable achievement in a game of this nature. I can easily envisage a scenario where a different implementation of this game received an E or an F. A ‘D’ usually means ‘Look, I think this is probably going to be pretty difficult to play if you have accessibility issues but if you wanted to I’m sure you’d manage’. An ‘E’ is ‘I don’t think you can realistically expect to play this’, and an F is ‘If you can play this at all I’d be genuinely amazed’. While it’s still a ‘we don’t recommend’, it’s a very different kind of non-recommendation that would otherwise have been the likely result.

Cognitive Accessibility

Scythe is a reasonably complicated and very complex game of actions that don’t quite align with your planning. Scythe requires you to constantly be thinking at least one move out of sync with your current intention because for most factions you can’t take the same action twice in a row. As such, you need to be considering what this action permits you to do for your next and what it prohibits. The actual impact of your choice likely won’t be felt until the next turn and there’s a significant difficulty curve that needs to be navigated. Initial turns are incredibly important and their impact is massively greater than those of later turns. Scythe presents a ‘change the headwaters’ circumstance – if you get a mech quickly it’ll enable a lot more than if you struggle to get that mech in comparison to everyone else. Faction design strikes me as reasonably balanced but not in terms of cognitive affordance – some factions are clearly going to be more challenging than others, made more so by the variable player mats. Your faction/player mat combination is a puzzle of optimisation that needs to be worked out at the time it’s assigned. Familiarity with a single combination is possible to build, and people will likely have favourites. This may not be possible to easily support in the kind of game group that would be most appreciate of Scythe. What do you when multiple players want the Rusviet / Industrial combination?

Coupled to this, visiting the factory gives a player a random action and understanding its power and opportunities must be considered in conjunction with the rest of a player board. Where it optimally fits into a player’s strategy may not be obvious even for a board that is otherwise well understood. Players also have hidden victory conditions that will incentivise, or disincentivise, particular courses of action. Simply being familiar with a faction and player combination does not mean that the game becomes straightforward. There is always an extent to which players are going to have to puzzle out their best strategy.

There’s a little literacy required to play Scythe, but most of it is contained in the scenario cards and those can easily be narrated by an able player. The vocabularly used in these is not excessively difficult, although the thematic trappings of play can be somewhat tricky to navigate.

Numeracy is a much larger problem, because scoring is difficult to calculate on the fly and yet important to appreciate on an intuitive level. The game offers a variant that heavily penalizes people that try to bring the game to a halt while they calculate out their actions but knowing where everyone is likely to be in the scoring is important because it’s a predictor for when the game will end. Technically the game ends when someone claims their sixth victory star. Really the game ends when someone claims their sixth victory star when they think it gets them the win. However, score is based partially on a simple calculation but also on the ebb and flow of popularity. As such it’s not just a case of knowing how well someone is doing but interpreting that through the lens of their popularity values. Someone can be dominating the board but still lose and being able to crunch and recrunch the numbers gives someone a considerable advantage.

Game flow is reasonably straightforward – pick an action, do its top part, optionally do its bottom part, move on to the next player. However, the manual also recommends a degree of interleaving here to speed up play and I think that’s unlikely to be appropriate for accessibility in this section. The manual points out that bottom actions take longer to do than top actions so when a bottom action is being worked out the next player should do their top action. That would hugely limit auditability of the game state for a cognitively impaired player and also likely create circumstances of confusion because they were distracted and didn’t see important game state changes.

There are a lot of different cards, tokens and parts to Scythe and it’s easy to be overwhelmed. Importantly these all interrelate because the mechanisms are highly synergistic. The synergy is relatively shallow – there are no exploding or self-referential chains – but in order to do one thing you need to do all the other things. Your popularity relates to your power relates to your income relates to your board position and so on. It’s hugely complex – a game of deep decision making where there’s a lot of subtlety. Knowing what the different tokens mean and what part of your faction they represent is vital.

So, that’s all the bad stuff, and Scythe is complicated enough as a game that it was never going to be a strong contender for a high fluid intelligence grade in this category. Again though let’s look at some of the features we’ve discussed and how they contribute to a game that is more accessible than we might reasonably expect it to be.

Another player board

Most of Scythe’s player board mechanisms work something like a tech tree – you invest effort into improving base level functionality and capability. Often this is handled with cards that are accumulated or tokens that get sprinkled on to other things. Consider Terraforming Mars for example where you end up with a cardboard spreadsheet of baffling complexity and are forced to work out how it all relates. Now compare that to the cube based console of Scythe and see just how smooth it makes the experience. They’re obviously not directly compatible games but I could easily imagine Scythe with card driven technology and it would have been a disaster.

The cube console system offers numerous affordances. The first is that cost and benefit are all displayed on the board exactly like you’d expect, they’re not offloaded onto other parts of the game interface. The board tells you everything you need to know without cross-referencing. It also heavily hints at paths of accomplishment. You have four mechs on your faction board. When all four are gone, take a victory star. When all your enlistment tokens have been removed, take a victory star. When all your victory stars are gone, you’ve won the game. Note how the workers on the player mat reveal ever increasing costs as they are placed on the board.

Scythe workers on the board

As such, while I can’t at all recommend Scythe for players with fluid intelligence impairments I’m willing to recommend it for those with memory impairments alone and that’s entirely a consequence of design. Seriously, this is what accessible design does – it doesn’t necessarily make a game playable by everyone, but it makes it playable by more people than would otherwise have been possible.

Emotional Accessibility

There’s a lot of very pointed aggression in Scythe and it can hugely undermine work that you’ve put into play. Resources that are gathered are left on the player board where they originated and other players can claim them through combat (if you have defences in place) or simply by moving a mech or leader into your workers. If you’re not careful here it’s easy to essentially hand an opponent a massive victory. In one game I played a player had managed to pick up about eight metal resource tokens and then moved their workers and mech to an adjacent hex. They didn’t need any more metal to finish their mech upgrades. But I did. So I moved two of my mechs and my leader into their hex, defeated them in combat, and sent their workers and mech back to headquarters. And then I spent all their metal as quickly as I could to prevent them attempting to retaliate. By the time they were in a position to get their metal back it was already gone. That was a two player game so it was easy for me to risk leaving myself otherwise undefended, but it was undoubtedly what comfortably won me the session. That’s quite emotionally risky as a design choice. You never lose mechs or workers, but you can lose the results of their labour and the territory they were controlling.

It’s also very possible here to experience an overwhelming sense of asymmetrical progress. If one player has a good faction / mat combination they can make rapid advances while you may struggle to make any. Early decisions are hugely impactful and someone having two of three good early turns might lead to them forever being ahead of you in the game. More than this, success tends to breed success. Mistakes tend to be costly and difficult (even impossible) to rectify.

If I have more mechs than you it gives me more flexibility while simultaneously making you more fearful. I can even lock you out of getting anything if things work out well for me – I can move my mechs into your metal producing areas and leave you almost helpless – stuck with the inefficiencies of alternate routes to resources. It’s possible for one player to end up with two mechs within a few turns with the right combination of faction, player mat and event luck. From that point on everyone is largely at their mercy. Players can absolutely bully each other, and the victory point system encourages it for everyone while positively incentivising it for some (Saxony). Everyone gets a victory star for winning a combat and six stars ends the game. If you’re gunning for victory you’re best to go for the weakest opponent, and it’s a good idea to keep them weak lest they seek revenge later.

Score disparities thus can be very significant, made more so by the popularity system that hugely increases the points received for game accomplishments. A player that is blocked out of access to events and resources may find themselves hugely behind everyone else when the final reckoning arrives.

We don’t at all recommend Scythe in this category.

Physical Accessibility

Here’s another area where the accessible design of Scythe yields benefits. While it might be difficult for those with fine grained motor control issues to slot cubes into place, when they are fixed in location they are difficult to dislodge. There’s no randomisation of game state with a disruption of a player board like you see in Terraforming Mars. It takes a very significant amount of unwanted agitation to change the settings of your console. It also yields itself well to cube placement being done on a player’s behalf because you can easily move player mats around the table without worrying about the impact.

Another busy hex

However, the board gets very busy and there’s sometimes a lot of manipulation needed in the midst of complicated setups. Imagine I have three mechs, four workers, eight metal, four oil and two wood on a hex. Then I need to spend three metal for another mech – I need to extract those tokens from very tight hex constraints. Currency like this is always present on the board because it can be stolen away from you and thus it’s not straightforward to replace it with written variants or other options. There are multiplier tokens that can be used in extreme circumstances but it’s still not an ideal state of affairs, especially given that the board of Scythe is very large – it’s an eight-fold board that opens into something like the large print edition of your favourite Sunday paper.

There’s also a lot of manipulation of the various trackers and scores and victory point stars but there’s nothing that says this need be done on a player by player basis. An abled player, should one be available, will be sufficient for this.

Token track on the board

Verbalisation is not well supported because none of the hexes have identifiable names, although a workable notation can be employed. It’s cumbersome though. ‘The first wheat west of the Rusviet starting location’ or ‘The lake north of my third mech, counting clockwise’. Map co-ordinates or landmarks would have alleviated this to a considerable extent. There’s nothing to stop players coming up with their own landmarks though. Given how rich the thematic framing of Scythe I can’t help but think ‘official’ names for parts of the map would have been very evocative. It’s not going to prevent anyone playing in any case, but some work needs to go into making it function.

There are hidden hands in the game, comprising of combat cards. There is also a combat dial which should be set to the amount of power a player wishes to spend on a battle. There are easy alternatives to the dial (writing on a piece of paper and simultaneously revealing for example, or having one player write and another verbally reveal) but the combat cards will need to be used in conjunction with a card holder.

We recommend Scythe in this category, provided players don’t mind playing with support from the table. If independent agency is an important consideration, we wouldn’t recommend.

Socioeconomic Accessibility

The leaders in Scythe are made up of a blend of personalities, genders and body types and come with a range of interesting animal pets. Given the theme of the game, they are drawn from a particular region of the world where the ethnicity trends heavily towards white European and Slavic – Germany, Poland, Crimea, Norway, Russia. These are not 100% correct real-world parallels for the factions in Scythe in the same way as talking about Germany doesn’t map cleanly onto the historical regions of Prussia. They are fictional parallels. It would be nice to see a wider range of ethnicities make their way into the game, but it requires an expansion before you get to add the Japan/Scotland analogues and I’m never a fan of diversity being locked away behind expansions. Still, men and women both get a fair crack of the leadership whip and everyone looks absolutely awesome. The manual makes use of second person perspective throughout.

Leader minis

Scythe is absolutely not a cheap game, with an RRP of about £70. However, it’s a game where the value for money is obvious in the lavish production values. It feels like a luxury product, and that’s exactly what it is in the end. It supports one to five players and does so well at all player counts. However, as much as I dislike the term, this is a ‘gamer’s game’ and it’s not one out of which you can necessarily make an easy family night. I genuinely think though the right group could play this hundreds of times without ever really getting tired of it. I’ve played it dozens of times myself, although mostly on the digital version, and I still occasionally crack it out. In that respect I think it has tremendous value for money but there’s no denying that £70 is an eye-watering sum for a frugal buyer to spend on a single board game. All I can say is that I think the cost is proportionate to the quality.

We’ll recommend, just, Scythe in this category.


There is a considerable amount of literacy required for play – not necessarily in terms of reading but in terms of perhaps playing in a second language. There are numerous event cards that must be dealt with. The sophistication required of a vocabularly is not high, but it’s also sometimes making use of words that are unlikely to be used regularly in a day to day context. This is going to be something of an issue too if players are communicating over an articulation or hearing divide. The framing of the game makes use of complex names for each of the factions. However, the factions can also be greatly simplified by simply referencing them by colours.

The game manual says that it’s possible to make ‘informal agreements’, bribes and alliances but it puts few rules on them. As such, it’s largely going to be up to the table to decide what those negotiations would involve and how permissive they should be with regards to confrontation, covert information and so on.

We’ll tentatively recommend Scythe in this category.

Intersectional Accessibility

We have been very complimentary about Scythe’s design in a number of these sections, but we still can’t recommend it in many. That limits the impact of intersectional issues because it means individual impairments are already a sufficient barrier to play without thinking about how much more problematic they become in combination.

For those with colour blindness and a communication impairment (wow, that’s a new one) it would perhaps be necessary to adopt a convention for discussing factions that didn’t focus on their colours (since people may not be using the same identifiers), but there’s already plenty about the factions that would serve as a suitable base for this. The tiger faction, or the bear faction, or whatever. A physical impairment intersecting with a communication impairment would make verbalisation more difficult, and in that circumstance we might recommend players consider a game that is more accommodating to unambiguous referencing.

Scythe tends to run quite long with larger player counts, and it absolutely can drag out at the end when it comes to everyone trying to calculate their way through optimal actions. The box suggests two hours of play-time but I think that’s optimistic at the five player level. It’s easily a game long enough to exacerbate issues of discomfort, and it’s not one that lends itself well to ‘saving’ because there is so much game state spread over so many locations. It also doesn’t cleanly support players dropping in and out. There is an automata but it doesn’t serve as a suitable replacement since it works in a different way to a simulated player. There’s no easy or sensible way to permit a player to stop playing without it impacting hugely on the rest of the table. Carving up their empire, removing them from play or skipping over is almost certainly going to result in one player being more advantaged than another.


Well, Scythe doesn’t escape from this with a particularly encouraging set of grades but it’s important to note that in many ways this is the best that a game of this nature could manage. There’s room for improvement, because there always is, but in many categories I would have put an E or an F had the design not been so good.

Scythe, Meeple Like Us, [CC-BY 4.0]
Colour BlindnessA
Visual AccessibilityD
Fluid IntelligenceD
Physical AccessibilityB-
Emotional AccessibilityE
Socioeconomic AccessibilityB-

These recommendation grades are not objective – they are judgement calls that broadly capture the humming and hawing I’d do if someone asked me directly. A ‘D’ doesn’t mean ‘this game does a bad job’ although it’s often hard to escape that conclusion. It means ‘On the balance of everything I think you’d probably find it less of a pain in the arse to play something else’. A ‘not recommended’ grade doesn’t mean ‘unplayable’ and I think Scythe is considerably more playable than it could have been in another universe. The difference between a D and an E might be tens of millions of people in the world being able to play, and the same number not. While Scythe didn’t get a lot of recommendations that doesn’t mean its design is a failure from an accessibility perspective. That’s important to bear in mind here.

We liked Scythe a lot, and gave it four stars in our review. When you multiply that by its popularity I think it probably wins at boardgames forever. It’s not a game that we can enthusiastically endorse for a lot of our categories but there is an intentionality here that is inspiring. Many games we look at, if they are accessible, are only accidentally so. Scythe is one of only a handful where if the publisher said ‘We took accessibility seriously and designed for it where we could’, I’d believe without question. There are a lot of really good lessons other designers can learn from Scythe, and I hope they do.

A Disclaimer About Teardowns

Meeple Like Us is engaged in mapping out the accessibility landscape of tabletop games. Teardowns like this are data points. Games are not necessarily bad if they are scored poorly in any given section. They are not necessarily good if they score highly. The rating of a game in terms of its accessibility is not an indication as to its quality as a recreational product. These teardowns though however allow those with physical, cognitive and visual accessibility impairments to make an informed decision as to their ability to play.

Not all sections of this document will be relevant to every person. We consider matters of diversity, representation and inclusion to be important accessibility issues. If this offends you, then this will not be the blog for you. We will not debate with anyone whether these issues are worthy of discussion. You can check out our common response to common objections.

Teardowns are provided under a CC-BY 4.0 license. However, recommendation grades in teardowns are usually subjective and based primarily on heuristic analysis rather than embodied experience. No guarantee is made as to their correctness. Bear that in mind if adopting them.