Magic Maze (2017) – Accessibility Teardown

Context of Document

This is not a review of this game. You will find the review linked in the introduction.

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 please stop reading now. This will not be the blog for you, and we have no interest in debating 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.

Version Reviewed

Multilingual Edition


Our review of Magic Maze is undoubtedly stingy with its stars – a mere two shiny stellar bodies for a game that many laud as the cardboard equivalent of high grade crack.  We calls them as we sees them, and in this case we can’t see a lot of fun in Magic Maze.  As is always the case your mileage will vary and you should feel absolutely no guilt in discarding our opinions like used handkerchiefs.   If you want to play Magic Maze, our viewpoint shouldn’t hold you back.  What might hold you back though is its accessibility profile.  Will it fare better here than it did in the review?  Let’s find out.

Colour Blindness

Colour blindness is handled in a way that is best described as weird.   The game itself in terms of colour choice is a problem – the pawns have overlapping palettes that make distinguishing them more difficult than it should be.  Bear in mind here that you don’t have a pawn of your own – all pawns are communal property of the table and all have colour sensitive destinations they’re trying to reach.

Colour blind pawns

Hardcore Pawn

All the colours on the game tiles come with icons designed to resolve the problem, but it’s not entirely successful because it turns a task of ‘at a glance’ identification into one of symbolic matching and processing.  You can’t just look for the purple vortex, you need to look for the vortex with the right symbol.

Colour blindness with tiles

Deuterenopes only

The game is colour blind accessible in its strictest sense, at least in this regard, but as the game tiles begin to sprawl the task  becomes increasingly more difficult because it becomes ever more time consuming to locate necessary icons.  Since the mall changes every time you play, and game sessions last roughly fifteen minutes or so, you can’t even just commit it to memory.  There’s a constant low level processing task added on because of colour choice.

Sprawling tiles

That’s a lot of sprawl

The pawns don’t map onto these icons because they don’t come with identifying symbols, except they do because there are stickers you can attach to indicate which pawn is which.   There is then an optional layer of colour blind accessibility provided in the box.  That’s not a unique approach – Terror in Meeple City for example does something similar.   However, the problem with the stickers is that they’re absolutely crap.  They’re curved, difficult to attach to the pawns, and do nothing but curl away and off during play.


Stick one on me

As such even fresh stickers when they’re attached to the pawns do little more than create the impression your pawn is opening up a grubby mac to flash you in a secluded park.

Bad stickers

Whether this happens to you depends on how perverted your pawn is.

It’s clear though that colour blindness has been considered since it’s mentioned in the manual (which I love to see) but the solution has problems all across the board.  It would have been so much easier to have just gone for a less problematic selection of colours to begin with.

We’ll tentatively recommend Magic Maze in this category.  The sticker issue drags the grade down a bit but if you don’t mind glue or tape you can keep them in place.  Having said that, if you’re going that far you’re already potentially defacing your pieces and can avoid the stickers entirely.  As I say – weird.

Visual Accessibility

For those with total blindness, the game will be impossible to play.  Not just difficult – I’m willing to go out on a limb here and say functionality it’s simply not a game that can at all be played.   Part of that is in the graphical design of the tiles which are information dense with zero tactility.  Partially it’s because of the ever evolving nature of the map and the way your role changes every time the hourglass swaps.  Primarily though it’s because the game rules mandate that turns are played mostly in silence.  It’s hard to imagine what channel of information could be substituted for a verbal description of a board that is already difficult to verbally describe.

For those with less severe forms of visual impairment the situation doesn’t get much better.   If you’re responsible for setting up a scenario you’re dealing with tiny numbers in the corner which are then linked to specific tile stacks from which you draw.

Tiny numbers

Teeny tiny

If you’re just going to be playing rather than setting up the game, it’s only a matter of minutes before you’re facing a board like this:

Starting map sprawl

Mere minutes later…

Look at all the information you need to be incorporating into play – location and direction of exits, vortex portals, escalators, special explore zones with different pawns for each, security cameras, hourglasses, and more.  The game gets more complex as you move through the scenarios too.   The biggest issue here though is that you don’t have a single part of the game on which you can focus – you need to be aware of everything.  If you’re responsible for moving pawns left, they can’t move left without your input.   Nobody can give you instructions, except in the simplest and easiest of the scenarios.  Your job is to identify points of leverage from the actions you have available and then act on them.   Close inspection of the board is possible but you’re also working to a very tight time limit that can only be adjusted by having access to single use hourglass tiles.   Time you spend investigating the board is time you’re not actually progressing towards the goal and it’s not an easy game at the best of times.

Larger map sprawl

Please say we’ve found everything

Sprawl becomes significant too because there are eight tiles that need to be found – the target tile for each pawn and the exit tile for each.   The vortex action makes traversing the board a little easier but it gets locked off when the theft is triggered.  As a result when the security shutters clang down in the mall you’re left trying to navigate the pawns from something like the configuration above so that they get to their matching exits.

Pawns on exits

Get the hell out of here

Any one part of any one pawn’s journey will involve long periods of you being involved and equally long periods of you being uninvolved.   You can’t simply wait until you’re needed – you’re needed everywhere, by different pawns each time.

Now, you can play Magic Maze without the restriction on talking.  You almost certainly shouldn’t.  The game just isn’t interesting and in any case it doesn’t become much more accessible.   At best it would become a game of someone saying ‘You can move yellow left!’ and then you doing that.   It can’t even be ‘We need you to do something!’ because again – that hourglass means you don’t have time to inspect the board to work out what that something might be.   You could also play it without the hourglass, but at that stage what’s the point of any of it?   It would be like playing Escape: Curse of the Temple in a turn-based mode.

We think Magic Maze is all but unplayable for those with visual accessibility needs.

Cognitive Accessibility

The ‘complete’ game of Magic Maze is quite cognitively challenging, but it does allow players to scale the challenge to competency through the use of scenarios that layer in additional rules as time goes by.  To begin with all a player needs to know is the set of basic actions – up, down, left, right, vortex, escalators and explore.  Depending on the number of players they may have only one of these they need to worry about at a time.   In the introductory scenario, the manual says that as a special exception talking is permitted through the entire game.   Scenario two adds in a restriction that means each pawn must exit the game through its own coloured exit.  Scenario three adds in action cycling when the hourglass is flipped.   Scenario four adds in the dwarf and elf special action, scenario five introduces the mage special power and scenario six adds in the barbarian and security cameras.   There is nothing to say anyone need progress beyond the introductory scenario, and this sets a baseline expectation of cognitive accessibility.  Everything else is optional.

Scenarios and rules


The problem is that the baseline here is either going to involve the restriction on talking or its going to lose the game’s main, key element of gameplay.   In the latter case, it’s better to simply play a cognitively accessible game in the first place – while Escape: Curse of the Temple has some accessibility considerations in this category it is broadly playable.   If on the other hand you adopt the communication restriction you ramp up the cognitive cost considerably.

Consider what not being able to talk does to the challenge of the game – every time a player move a pawn you need to reassess the applicability of your action towards the current goal.    If you don’t act on that in a timely manner you may slow down everyone else.    The current goal might change as tiles are drawn, and without the opportunity for anyone to discuss what may have happened.  For example, you might have the purple pawn exploring when someone else finds the purple loot tile.  At that point it probably makes sense for the purple pawn to make its way to the newly discovered destination.    You need an awareness of what’s happening everywhere to prioritise your actions or you may find that you and other players are actually working against each other.   That change in goal also needs a kind of adaptation for all the other pawns – perhaps orange is now best tasked for exploring as an example.

A single player can be a bottleneck for everyone, and that can pile on a lot of pressure – the big ‘do something’ pawn can be used to tell someone they need to do act.  That something may not be at all obvious and in performing an action you may do serious damage to the state of the game.  For example, a vortex action might move a pawn from one side of the board to the other and that can be difficult, if not nigh on impossible, to fix given the tight time constraints.

Game flow isn’t just variable, it’s basically a batter that is made up of a number of ingredients that change on a second by second basis.  The game is played in real-time which hides this to an extent but we’re not talking here about the order of turns being malleable but rather the weight of activity.   You might be very, very involved or not involved at all for completely variable chunks of the game.  You might have everyone waiting on you, or you might not be able to do anything without upsetting other plans.   You might be waiting on other people to do something while they’re simultaneously waiting on you.  There’s no obvious shift from one phase to the other than what you pick up on yourself.

Elf symbol

You’ve heard of elf on the shelf, but what about…

For those with memory impairments only there are fewer areas of concern.  When playing with the action tile switching rule players need to remember that their actions will change on a regular basis, and incrementally the rules add in a degree of complexity that comes along with playing accurately.   The elf explore action for example permits people to speak, and the security cameras add a condition of movement on to sand timers.  Remembering the set of these and how to apply them is made easier by symbols on the tiles but it’s still a memory burden since everyone is focusing on their own tasks.  By and large your mistakes are going to go unobserved since everyone has their own difficulties.

On top of this we need to layer the real time aspect and the time limit that adds an extra dash of additional cognitive complexity onto everything since it means the collegiate element of a co-operative game is made more burdensome.  You can’t simply stop to explain or help someone with their task because you’re all too busy with your own.

We don’t recommend Magic Maze for those for whom fluid intelligence impairments may be a factor, although we can tentatively recommend it for memory impairments only.

Emotional Accessibility

Mmm!  Well, here’s the key consideration for you – how well do you handle being nagged to do something when you have no idea what it is you need to do?   That’s what the ‘do something’ pawn is – a way for a table to repeatedly and regularly prod you to activity without any other clarification of action.  It’s not ‘You need to move the yellow pawn, take this!’.  It’s ‘You need to do something, and we can’t say what it is’.   And then, how well do you handle that when there’s a time pressure?  And when the enjoyment of the entire table might be, to an extent, on your shoulders?

Magic Maze is a co-operative game where it’s absolutely possible for one person to lose the game for the rest.    For example, you might have a pawn on an escalator and two pawns blocking each other and the first at the top.

Blocking pawns

Pardon me, can I get past?

It’s not difficult to resolve it but the exact way you do that might need different people to work together in different ways.  You could trigger the orange explore option and then move the orange pawn either through the new tile or left and down through the orange door.  You could move orange to the door, green to where orange is, orange through the door, green to the newly vacated space, and then purple up the escalator.  Or you could move purple down, use the escalator to get green to the next floor and then orange can please itself.  So many options, all of them very easy.

Except it’s different people that need to work together at different times with different actions for each scenario, and none of them can work without multiple players interacting.

So what happens if the player that can trigger the green pawn is busy on another part of the board, or is working for one strategy and everyone else is working for the other?    What if the player that can trigger the escalator action is waiting for green to move left, and the player than can trigger the left action is waiting for orange to explore?    In general, nothing happens here except the do something pawn starts making its way around the table like weaponised passive-aggression.

While everyone might be to blame in the above scenario, sometimes the problem is that you just took an action that caused more damage than anyone could resolve – often a portal related activity.  Or the problem might be that you as an individual weren’t picking up on the signals the table was sending out.   At the end, when it’s all over, recrimination is possible and it can be highly targeted especially since people, as a rule, are not great at taking personal responsibility for collaborative failure.

Magic Maze can be stressful.  It can be chaotic.  It can be funny but it can also be very frustrating.  That frustration has to be expressed non-verbally except at certain pre-defined points in the game.  That’s not a great combination of things in this category.    All of this might be okay if it wasn’t also very difficult, at least until you’ve developed a kind of lexicon of secret communicative regimes.  Familiarity will dull the difficulty, as with any game, but in the mean-time expect to lose and expect the burden for losing to fall disproportionately on identifiable players.

We don’t recommend Magic Maze in this category.

Physical Accessibility

There’s a lot of reaching over a game map that is going to grow and expand as time goes by, but all you’re doing most of the time is moving a pawn from one place to another.  You only move in one direction (or perhaps more at lower player counts) and as such your actions are mostly possible to verbalise.  The problem is that… yeah, you may have guessed it, you’re not supposed to talk during the game.   It seems reasonable that an exception be permitted in this case though since all you’re doing is saying what you’d otherwise do physically – it doesn’t include any other kind of gameplay information that would otherwise be hidden.   Saying ‘Do something, Michael!’ would be another solution to the problems that might come with the do something pawn.

However, it is still a real-time game and everyone is going to have their own things to be worrying about.   It’s also something that would require a very strict rule of refusing to countermand orders given.  Nobody should be permitted to question the order on the grounds that it’s going to interfere with other plans.   For the most part this would be sufficient, although it’s going to be an issue for portals.  Strictly speaking you can always identify these such as ‘Green portal on tile ten’ but the numerical identifiers for tiles are too small to be useful in this regard.   As the board gets bigger, the harder they’ll be to see and the harder it will be for people to find them when you want them referenced.

If dealing with the physical interaction personally, there’s quite a lot of arm movement required as pawns distribute to various tiles.  You’re working with all of them constantly, so you might be grabbing and moving a pawn up in the top left and then back to the bottom right and then back up to the top left before moving off to the top right.  The interaction can be quite intense, although it doesn’t need to be particularly fine grained.  As long as pawns don’t go in directions they’re not supposed to go they don’t need to end up precisely in the squares they’re intended to occupy.  If you’re responsible for the explore action you’ll also need to grab tiles from the central stack and place them in the maze – that’s a task of orientation and working within occasionally quite tight constraints.    The tiles don’t fit together in a standard, deterministic square – they’re all slightly off centre so they create some interesting patterns where they could overlap if this isn’t done neatly.  Not an insurmountable problem, but certainly a potential issue.

We’ll tentatively recommend Magic Maze in this category.

Socioeconomic Accessibility

The wizard is a woman, judging by the cover.   We’re given no indication of the elf since he, or she, is way off in the distance heading down an escalator.    The manual doesn’t default to masculinity.  The other two characters prominently on the cover are a dwarf and a barbarian.   All are white, as is often the case in the ‘lazier’ kinds of generic fantasy.  Recent versions of RPG sourcebooks and fantasy books have done a wonderful job in widening diversity, but it still isn’t reflected in the default theme we see in many games.     It’s a bit of a shame – there’s plenty of easy scope, and precedent, for it.

Magic Maze box

Concept art from the Hobbit movies

Magic Maze has an RRP of around £25 and it’s hard to fault at that price.  It plays reasonable well as a solo game.  Indeed solo is my favourite mode if such a thing could be said to exist in a game that got such a poor reception in our review.  It goes all the way up to eight players, and while I’ve never played it at more than four the collective wisdom of BoardgameGeek gives it a recommendation at all counts between two and seven.   Given the breakdown there it looks like it probably suffers as more players get layered in, but only at eight does the playability seem to become seriously compromised.  Having said that, BGG doesn’t recommend it at one player either so take this with a pinch of salt.

We’ll recommend Magic Maze in this category.


Paradoxically this is a game that is about not communicating but still has massive problems in this category.  It’s not, after all, that you can never communicate – it’s that you need to compress communication into intensely information dense bursts when the sand-timer is flipped or when the elf takes an explore action.   Depending on the player count this might be a quick, purposeful discussion or a rollicking argument with everyone talking over everyone else.

It’s a feast or famine model of communication and it’s going to be an issue – those with hearing difficulties may find it troublesome to pick up on subtleties of the discussion if it’s happening at speed, and the hourglass means that there isn’t enough time for everyone to fully articulate everything they say without it having a serious game impact.  Those with articulation difficulties may find it problematic to express their viewpoint and have it understood, especially since it’s going to be happening under stressful conditions with everyone mindful of the time.  You could pause the game at this point, to an extent, but that’s a violation of the spirit of the game – it’s designed around the idea of these transient, occasional windows of planning and something would suffer if their duration was flexible.

There’s no need of literacy, and the largest burden of communication in the main portion of the game is going to be implied – grabbing the ‘do something’ pawn and meaningfully placing it in front of another player.  Or banging it like a gavel.  Or staring deeply into their eyes until they get creeped out and do something out of sheer discomfort.

The main meat of the game, where communication is restricted, is likely to pose few problems for players in this category.  The occasional planning windows that open will be more of an issue.   Whether this is likely to be a deal-breaker is going to depend on the size of the group, the extent of the planning, and any combinatorial communication issues that need to be taken into account.  We’ll tentatively recommend Magic Maze in this category, but only just.

Intersectional Accessibility

A number of the recommendations here are very tentative, and they don’t become less so when intersections must be taken into account.   Colour blindness and memory impairments would be enough to rescind both recommendations because the solution to the colour issue is through symbols and those symbols will need located and parsed in every action a player takes.  It’s not onerous in and of itself, but adds a low-grade memory burden to every single action a player performs.  For those where physical and memory impairments may intersect, the degree of multitasking required may be sufficient to cause problems, especially given how difficult it can be for a plan of action to be held in mind as it changes according to circumstance.

For the intersection of communication and physical impairments, there will be a significant impact on verbalisation but also on the implied conversational tool of the ‘do something’ pawn.  When everyone is on equal footing during the quiet phases the pawn can be used to indicate that someone needs to act.  However, if that pawn can’t be used and a player cannot clearly indicate that a player should act, and which player it is, the only real prod available is removed from the game.

The time constraints we have discussed throughout are a major issue in every category, and it’s best to think of them as an accessibility tax on every interaction performed with the game.  As such they should be multiplied when dealing with any intersection.   Timed scenarios are rarely compatible with accessibility compensations because accessibility takes time and that essentially means everyone is always playing on a level of difficulty higher than the game itself is designed around.

Magic Maze simultaneously makes players irrelevant as individuals and yet ties their presence so tightly to the board that they can’t drop out.  You don’t control a character, you control a direction or action and as such you could be cleanly replaced by any other player by simply passing your tile to them.  The problem is that it’s not just about the ability to perform actions, it’s about ensuring those that can perform actions are separated.  Being able to control up and down, for example, might be incredibly over-powered in a game balanced around everyone having a different direction.   The availability and composition of action tiles is tightly bound to player numbers and you can’t simply double these up without game impact.  That’s a shame, because this is also a game that is likely to be very intense during its relatively short play-time.  You don’t get much of a chance to relax between actions – there’s no such thing as downtime during Magic Maze.  The game begins, and you’re on – and you stay on until the game is over.    The time issues, the intensity of interactions, and the stresses that come from having a table focus their attention on you when you don’t know what they want – it can all result in a draining fifteen minutes and it’s important to take that into account when setting up for play.


Well, it perhaps isn’t a huge surprise that Magic Maze doesn’t come out of this teardown in a particularly great state.   Real-time games are rarely strong candidates for accessibility, and while Magic Maze has a number of very positive accessibility features they don’t cohere in a way that makes it easy to recommend.

Colour Blindness C+
Visual Impairment F
Fluid Intelligence D
Memory C
Physical C
Emotiveness D
Socioeconomic B
Communication C-

It’s not entirely a lost cause though – some of the problems emerge through slightly weird design choices.  Stickers that don’t stick, tiny numbers that don’t differentiate tiles, and a few other easy fixes that a later edition could address.   Most of it though is fundamental to the nature of a game like this – one where you’re being deliberately set up to make mistakes and given all the tools to optimise your own descent into failure.

We didn’t get on very well with Magic Maze which is why it got only two stars in our review.   We don’t recommend it as a game, and we’re not able to recommend it particularly strongly as an accessible product.   For now, the hunt for the perfectly accessible and fun real-time game remains active and I suspect we may never truly strike gold.

If you like what we're doing with Meeple Like Us, please consider liking us on Facebook, following us on Twitter, and sharing our content. If you want to support us financially, you could make use of our Amazon affiliate link to send a bit of cash our way - it doesn't cost you a penny! Drop us a comment too in the discussion section below! We appreciate every thing you do to help us get the word out!