Deception: Murder in Hong Kong (2014) – Accessibility Teardown

Game Details
NameDeception: Murder in Hong Kong (2014)
ReviewMeeple Like Us
ComplexityMedium Light [1.60]
BGG Rank161 [7.54]
Player Count (recommended)4-12 (5-12)
Designer(s)Tobey Ho
Artist(s)Marcin Adamski, Ben Carre, Tommy Ng and Ari Wong
Buy it!Amazon Link

Version Reviewed

Jolly Thinkers English/German edition


Deception: Murder in Hong Kong is seasoned through with flavourings of Dixit and big lumps of Mysterium. A little soupçon of One Night Ultimate Werewolf is sprinkled on the top like a fine garnish. Is a soupçon something you can garnish? I honestly don’t know. I often use words without knowing what they mean.

Anyway, in our view it is a genuinely marvellous game and one that I have no hesitation in recommending to people who have always wanted to see what a miscarriage of justice looks like from the inside of a police disciplinary hearing. That’s not why you’re checking out this post though – you want to know about whether it’s playable at all. The only clues we have to go on are this claw hammer and this outsized novelty clown nose. Let’s see if we can’t apply them to a more productive teardown.

Colour Blindness

Colour blindness isn’t a serious issue. Really the only places that colour is used for informational content is in the special evidence tiles (location and cause of death) and differentiating between murder weapons and clues. For the former, some careful pre-organisation of the tiles will alleviate the problem. In any case, each tile has appropriate text displayed along the top.

Colour blindness tiles

For the clues they’re differentiated by card back but in a way that will be appropriate for the majority of players will colour blindness. It would have been fully colour blind accessible if it had used a different symbol on the back instead of the question mark on both.

Colour blind card backs

Distributing these cards though is not actually something that need be done by a colour blind player, and what small issues might be caused by this in unusual circumstances can be alleviated by ordering the cards in rows.

Colour blindness card fronts

We strongly recommend Deception: Murder in Hong Kong in this category.

Visual Accessibility

Secret role cards are dealt out to each player, and this represents one of the key visual inaccessibilities in the design. However, it’s a slightly unusual setup in that you essentially have a moderator in the form of the forensic scientist. It’s possible to house-rule a variation of the game where the investigator physically indicates who the murderer is if necessary. For example when everyone closes their eyes they can flip their cards to reveal their role and the investigator can indicate covertly the state of being a murderer to a visually impaired player. It’s not an ideal setup, but it is a workable solution.

Clue cards

As to game state, the murderer is responsible for indicating murder weapon and clue from the set of cards in front of them and this is likely to be the largest sticking point for this section. When dealing out cards this can easily be verbalised so everyone knows what’s in front of them but selecting an optimal combination is based on ambiguity. If there are lots of people with blunt objects then it’s safer for you to pick a murder weapon that would share physical properties with them. That means examining eight cards for each investigator, and these may be oriented at awkward angles or located at far ends of the table. The cards are also Very Small and that is an accessibility issue across the board. Close inspection of these when the game is up and running is fine – it’s a common thing for all investigators to do and verbal querying would be appropriate and would not leak game information. During the eyes-closed phase though that’s not going to be possible without physically revealing your role as the murderer.

One option would be for the forensic investigator to verbally summarise each player’s cards but that could take quite some time. In any case that would result in a fairly elaborate set of information with implied connections between each that needed to be held in memory. Any attempt to alleviate the difficulties in this (taking notes for example) will reveal the murderer by their physical activity during the closed eyes phase; the time taken for that phase; and any physical traces of those clues when everyone opens their eyes.

The scene elements are likely to be less problematic in terms of player awareness of information but equally troublesome in terms of parsing that into game meaning. After all, if the scientist indicates ‘instant’ for the speed of the death you need to be able to make connections between that and every other card in the game. You’ll be constantly ruling some in, some out, and occasionally from one column into the other and back again. There’s some tractability here in that the discussion around the table will naturally tend to converge on choice bits of evidence. That’s not the same thing as identifying them personally or having an opportunity to make an intuitive leap of logic that has escaped everyone else. You can certainly ask players ‘Do you have anything sharp?’, or ‘Do you have anything that would lead to a neat scene of the crime’ but there is a giant amount of opportunity for lying by omission.

We very much don’t recommend Deception: Murder in Hong Kong in this category.

Cognitive Accessibility

Unsurprisingly, regardless of which role you adopt, this is a game that doesn’t offer a lot of good news in this section. If you’re the scientist you’re not only having to come up with good evidence but situate them in a deductive context. General knowledge is hugely important here for many of the combinations of clues and evidence. For example, consider the sulphuric acid murder weapon. How would that work? Where would you get it? Does it have a smell? How long does it take to kill someone? In what condition would it leave a corpse? Is there a seasonal availability that might come into play? All of that is genuinely relevant to the choices you make. What about kerosene, or liquid drugs? Plague? Scorpions or snakes? How long would it take to kill someone with an electric baton and how long would they make noises for before passing out?

You only have a small number of options to choose from in any category, but that actually makes it more difficult because of how generally applicable they are and how few of them genuinely offer meaningful links between game elements. You need to look at the offering you have in terms of how it all fits together as a whole and that’s not at all easy. It’s especially not easy since at any time there will be at least one player trying to undermine your clues so as to protest their own innocence. You will occasionally have to be reactive to this to steer the conversation, quickly, away from territory you know is going to be a problem for the future. In all of this, you also need to remember details of the crime in terms of the murder weapon and clue and while there are only two things they can occasionally get forgotten in the heat of the discussion. There is no way to get a reminder of these without essentially conceding the round.

If you’re an investigator it’s a different but equally cognitively complex task. You need to deflect attention away from you and on to someone else. That’s going to involve being able to refute allegations made about your cards while also recasting them to fit the cards of another player. You need to do that in relation to what information the scientist gives you when some of that information may not even be helpful. Again, you’re doing this while at least one other player is trying to get everyone else at the table to disbelieve your case.

Scene tiles

If you’re the murderer then your job is to do exactly the same but to throw accusations everywhere by looking for the widest possible interpretation of what the scientist has indicated and then making compelling arguments as to why that should be the key line of inquiry.

Regardless of the role there’s a lot you need to remember, such as which evidence was most strongly expressed by the scientist, in what order, and which cards have been ruled in or out of contention by the collaborative ‘wisdom’ of the table. It’s just an awful lot to do and under circumstances where at least one person is trying to undermine whatever narrative is emerging.

Coupled to all of this is an expectation of literacy; a complex and evolving game state; and a deep requirement for contextual knowledge for interpreting and giving clues. We don’t recommend Deception in either of our categories of cognitive accessibility.

Emotional Accessibility

I mentioned in the review that failure states in Deception are hilarious, and that’s true… for the right group. The thing is that while it’s certainly possible to make informed deductions based on clues and cards and evidence it’s always going to be done in the context of a scientist that has to select the least bad option on the tiles they have. As such, the whole thing is set up to fail in a way that cascades into farce. The last time I was in that position I was demonstrating to people how to play by taking the role of the forensic scientist . For a good chunk my time in the role I had my face in my hands laughing to myself. Not because everyone else was playing badly, but because I was. Literally the first words out of my mouth when the murderer got away was ‘In my defence…’

And that was great because I love it when games make me and other people laugh. I passed the forensic scientist card on to someone else, warning them it was harder than it looked. And it was, although she did so much better than I did. That was also funny because she went through many of the same emotional reactions as me but had more success at the end. When it’s played in the right mindset this is a joyful game even if it’s about a grim subject matter.

However, that right mindset doesn’t always exist and in a group where people are interested in games as systems rather than as visceral experiences there is a real chance it’s going to be more frustrating than anything else. We didn’t even track wins when we were playing – if you absolutely need a score and need that score to reflect achievement this game is going to be a problem in this category. It’s semi co-operative in that there is a team that wins together and a murderer who wins if they get away. As with the Resistance membership of the teams is not transparent and you might well find your own immaculate argument undermined by someone on your side to the detriment of the investigation as a whole. Someone is always going to be trying to make your argument sound stupid.

In this there’s a need for the murderer to bluff and lie and effectively deflect unwelcome attention on to other people. That can require some quick thinking under semi-stressful circumstances. The lies though don’t need to be on an emotional level – you don’t need to convince people to trust you, you need to convince them to believe your logic. As such, those uncomfortable with lying can focus instead on alternative valid hypotheses. ‘All the evidence we have also suggests this as a likely possibility…’

Deception doesn’t have a player elimination system, but you only get one guess during the game. When it’s used up, it’s gone but you still get to contribute to the conversation. That’s actually one of the nicest systems that I’ve seen for this – it adds weight to making guesses while retaining the involvement of everyone at the table.

For those playing the role of the scientist there is a lot of attention on you, up to and including the body language that goes into placing a clue. The manual says ‘The order in which the forensic scientist chooses to place the markers, as well as any signs of decisiveness or insistency on the part of the forensic scientist, can be clues’. Given how misleading some of the clues can be otherwise it can occasionally result in circumstances where your every action is scrutinised under a magnifying glass. Some people won’t like being the centre of all that. Coupled to this is an inability to undo mistakes – when a bullet is placed, it’s there for the duration. You do get one tile you can replace towards the end but that can often do more harm than good.

The subject matter in Deception isn’t represented graphically but the discussions will trend towards dark and unpleasant areas. You’ll find people debating exactly in what state you’d find a body after someone had sulphuric acid thrown in their face. People will be discussing whether a wrench or a hammer is most likely to leave a corpse disfigured, or whether a bystander would hear someone crying for help if they’d been buried alive. The art and thematic style of the game isn’t gratuitous but exploring these topics is an inevitable part of making an argument for a murderer. If there are clues that are broadly applicable to what the scientist indicates then you need to work out which is more indicative. The crime happened on a sunny weekend on a building site. Was the smell noticed by a bystander from the corpse itself or from some feature of the crime? How long did it take to be discovered? Assuming nobody was working at the weekend, in what state would you find the corpse?

We’ll tentatively recommend Deception in this category, with caveats. If you’re concerned about emotional accessibility and have someone that asks ‘so, how does scoring work’ it’s probably worth spending time considering if it’s likely to be appropriate. The subject matter probably going to be distressing for some people, but you’ll be a better judge of that for your group than I could be. Just bear in mind this isn’t CSI: Disney.

Physical Accessibility

There isn’t a lot of physical activity in Deception that can’t be handled by the table as a whole. The scientist has to indicate clues, but there’s nothing to say that couldn’t be done verbally provided you maintain the rest of the rules regarding what you can fairly disclose.

The identify of the murderer and forensic scientist is properly handled by a card deal, but I’ve found it’s more fun to have the latter of these rotate around the table. Checking the role card is the first physical activity that can’t be conducted by the table but there are workarounds. The forensic scientist, as discussed in the section on visual accessibility, could flip over each of the cards and physically identify the murderer if necessary. The existence of the scientist solves a number of the problems that we have discussed in One Night Ultimate Werewolf and the Resistance since no information is leaked by having them act as a kind of human accessibility aid.

The murderer must indicate two clues, and that’s usually done by pointing. The key thing is that has to be done silently so as to not give away the identity of the murderer. If this isn’t appropriate some other system would be possible – for example, the investigator could hold up fingers to indicate which clue and which murder weapon, with the murderer giving some quiet but observable indication of agreement.

The rest of the game is conducted verbally, and aside from ‘handing in your badge’ there’s nothing that requires any physical activity. Or at least, nothing that absolutely mandates it. It’s hugely helpful, given the small size of the cards, to be able to get up close and examine what cards another player has. That can be done verbally without too much difficulty but it removes some scope for being tricky with your accusations.

We’ll recommend Deception in this category.


This is obviously a massive problem category. The discussion is intense, complex, and will always be done with at least one person attempting to undermine any given part of it. Literacy is needed and while my copy is multi-lingual that’s only going to work for the specific languages supported. Otherwise a relatively wide-ranging object vocabulary is needed to relate clues to cards and evidence. Most of the problems we saw in The Resistance manifest here – fluency of communication is important for credibility and agitation in gameplay may result in compensation regimes breaking down. There’s also an adversarial approach to evidence gathering and presentation that means at least one person is incentivised to make other players think you’re lying even when you’re not.

Deception though does have something in its rule system that I absolutely love – there is a formal part of the game that is called ‘presentation’ and it permits an uninterrupted opportunity for a player to express their theory on the crime. This is a moment of calm in what might be a maelstrom of discussion and argument. It doesn’t solve the other problems regarding the more free-form discussions that are the backbone of the game but it does alleviate some of the worst issues that come from games like this. While someone may find themselves ignored or spoken over they can’t be entirely shut out of the discussion. An uninterrupted chance to put forward your case can completely circumvent all the previous attempts people made to make you look unreliable. The rules say ‘Each player may take about 30 seconds’ but then immediately adds ‘This is just a guideline. You may adjust the time limit as you see fit’. The effect of this is that you have, built into the structure of the game, permission to allow those with accessibility needs to take what time they need to fully contribute.

That said, while this is a tremendous feature in a game of this nature we still have a pile of problems that mean we still can’t really offer a recommendation. For one thing, while having the stage to yourself means you can construct a clear, coherent argument it doesn’t eliminate the fact other players can make you look dodgy when the game is at its most energetic. Players are incentivised to take advantage of confusion, poor phrasing and lapses in the conversation. Sometimes the argument constructed can fail just because you personally were made to look like the murderer when the argument was at its loudest.

We don’t recommend Deception in this category.

Socioeconomic Accessibility

The cover is both good in that it shows a prominent woman investigator, and bad in that it shows that she turned up to a murder crime scene in what are effectively a pair of hot-pants. It clearly can’t be because it’s too hot because the male investigator is wearing a full suit and trenchcoat. She herself is wearing a coat.

Deception box

The investigator cards are silhouetted but you can still make out it’s a man on each of them – it’s the same one for each card. The murderer and accomplice are a little more ambiguous, as is the forensic scientist. Most of the rest of the cards in the game don’t have any gendered representation, with a few exceptions – mostly those are focused on men. Men are the subject of paintings, on the cover of magazines, represented in photographs, and so on. There’s also this weirdly specific objectification in the cards you might get dealt:

Weird stockings card

Notice something? All of the clothes are devoid of occupants except for the stockings – it’s so specific that it’s borderline fetishist. Just a sexy woman’s leg in a stocking. It’s the purest example of objectification I’ve seen in a long time. The panties you find elsewhere in the deck are not being worn, which is consistent with the rest of the garments. The stockings though?

That’s weird, right? It’s not just me?

Deception: Murder in Hong Kong has an RRP of around ~£30 and given the player counts it supports and the sustainability of the experience I think it’s a bargain at that price. We’ll recommend Deception in this category.

Intersectional Accessibility

There aren’t many categories where specific intersections are likely to adjust any individual recommendations much. Emotional accessibility issues that intersect with socioeconomic concerns regarding the subject matter might be an issue but they’re likely to be more philosophical than actively exclusionary. Otherwise, I’m not sure there’s any specific issue that I would suggest requires consideration.

The game is quite intense for its playtime, but it’s going to be one of those games that are made up of a number of distinct sessions rather than one session with many rounds. As such, while the game doesn’t cleanly support dropping in and out of play it’s easy to reset for a smaller play-count at the end of any given round. Since rounds of this last maybe twenty minutes under normal circumstances there are plenty of opportunities for those experiencing discomfort to remove themselves for a time and then come back in later.

Given the nature of the game, there’s an intersectional issue in that nobody at the table is incentivised to offer accessibility support beyond that which is actively requested. The murderer is usually acting alone (unless you’re using an accomplice) and the investigators don’t know who is who. Our normal advice stands – play games with those as interested in the collective fun as they are in their own.


Deception: Murder in Hong Kong is an absolutely lovely game that gets deeper and more interesting with familiarity. It has all the fascinatingly misleading interpretation that you find in Mysterium while retaining the vibrant metagame of something like One Night Ultimate Werewolf.

Deception: Murder in Hong Kong, Meeple Like Us, [CC-BY 4.0]
Colour BlindnessA-
Visual AccessibilityE
Fluid IntelligenceD
Physical AccessibilityC+
Emotional AccessibilityC+
Socioeconomic AccessibilityB

Unfortunately these games involve hidden roles, intense communication, and often a degree of covert activity and that results in low grades in a number of these categories. Deception has a few features that dull some of the sharper edges (the forensic scientist as a kind of moderator, explicit conversational space for uninterrupted deduction and so on). Those improve the situation but they don’t fix it.

We gave Deception four and a half stars in our review and as a game we’d recommend it to anyone. We played Mysterium and felt the extra setup it layers onto Dixit doesn’t give a proportional amount of extra fun. We love Dixit but admit it doesn’t have much of a game surrounding its charmingly whimsical aesthetics. Deception is both more interesting a game than Dixit and less cumbersome to play than Mysterium. It’s just a shame it’s not likely to be an accessible experience for a lot of the people that would love it.

A word 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.