Detective games


Murder on the Mississippi . Commodore. Best game ever.


Never heard of either of these! They look glorious (if unlikely to hold up today). And Zinderneuf was co-created by Paul Reiche III, so it’s got that Star Control 2 retroactive pedigree!


Not to mention scrolling graphics. And animation!


I started playing Deadline, due to some German podcast I listen to, and I’m surprised by how much I’m enjoying it so far, considering that it’s older than me. The confined environment and limited time really help to make the unfair design and frequent restarts palpable, at least so far. I’ve made some progress, but I’m quickly running out of ideas and will probably be stuck soon. The parser seems kinda limited by modern IF standards, but still managed to surprise me when sequences like this work:

> McNabb, show me X
“Follow me!”
> follow McNabb

A pity that it doesn’t seem available for purchase anywhere, except the AppStore, making it ineligible for the Classic Game Club.


You can buy it used and every now and then new copies show up.


Sherlock Holmes: Consulting Detective is a cooperative or solitaire board game originally from 1982, and then republished in 2014. Few board games from that era are tolerable today, but SH:CD is an exception. It also feels totally fresh. During those 30 years basically nobody did anything with the system, there were no spiritual or mechnical successors.

Here’s how it works. You start the scenario with a description of the mystery. Usually it’s like a decent pastiche of the opening of a Sherlock Holmes story. But it can be something different too, like just being told “there’s something interesting in todays newspaper, figure out what”. A turn consists of a player selecting a location to visit and reading the text associated with that location out loud. Then the next player chooses a location, reads the text out loud, etc.

There is no further interaction with the scene beyond that.

At any time the group can stop playing, and proceed to answer questions about the case. You don’t know the questions beforehand, so if you didn’t do sufficient investigation, you might not even have any idea of what the question is about let alone have the right answer. Your score (either as a group or as individuals) will be increased by getting questions right, and reduced by visiting more locations.

The game has four main components. Most importantly, for every scenario (10 scenarios in base game) there’s a case book. The case book is indexed by location, with each entry having some amount of text explaining what happens when you visit that location. It might be as little as 10 words, or it might be up to two pages.

How do you find locations to visit? The main tool for that is the Directory, an alphabetical index of people and businesses. The Directory is huge, probably 2k-3k entries, so you’ll never find anything from there by chance. Instead you’d need to pick up the right clues from the text.

Finally, there’s a couple of less important components. The map of London is used mostly to gauge distances between different locations, but can occasionally be useful for other things, like street names or finding major landmarks / locations. There’s also a newspaper (a double sided A3, with normal newspaper sized print) with short articles, ads, classifieds etc. any of which might provide crucial clues.

In some ways the game is absolutely brilliant. It gets the feel of a Holmesian investigation just right. You need to pay attention to tiny and sometimes seemingly irrelevant details, link them up just right, and make bold leaps of deduction. It is not enough to simply think about the mechanics of the crime, but of fuzzier things like character personalities etc. There are red herrings and totally separate subplots that can drag you off course and make you waste valuable time.

The game does however have two major problems. The first one is that the implied goal is to play as little of the game as possible. There’s a “par number” for each scenario, which tends to be absurdly low. I’ve bought a game with 10 scenarios, none of which I’ll ever be able to replay. Why are the incentives of the game set up such that I should ignore 90% of the content?

Second, while most of the scenarios I’ve played through so far have been pretty satisfying (with the exception of case 3, more on that below), a lot of them suffer from the same problem. Usually after about 10 visited locations we have the shape of the mystery right. Often we even have all the details right, but we just don’t have proof, there’s no smoking gun or confession. It is obvious exactly which bit of evidence we’d need, or which person we’d need to talk to. We’ll visit locations that are less and less likely to yield that final bit of insight, until we give up. And it turns out that the proof was never there in the first place; you really were supposed to just proclaim the case solved after coming up with a plausible theory.

It’s obvious why this problem exists. You do not want the players to find a smoking gun by coincidence, but since the contents of each location are completely static, it’s not possible to give the players who’ve deduced thing X different information than those that didn’t. You’d need to come up with ways of hiding the information into a location that only the first group would think to visit. But that’s a hard thing to pull off repeatedly.

I don’t find these to be fatal flaws, you just need to understand that this is something you play for the experience, not for getting an optimal score. I’m still really looking forward to new game sessions. And I can’t wait for the expansion, so that we can stop rationing scenarios.

The 2014 Ystari edition is kind of rubbish. There’s a lot of small typos / OCR mistakes all over the place. At least one case (case 3; the Mystified Murderess) is totally destroyed by sloppy editing (they changed the plot, but in the English translation they left the original text in place for some locations). The new plot also makes very little sense even if you print replacement components.

Finally, while reading up on Deadline, I stumbled on a 1982 issue of Softline that had an article on detective games. And that had this great bit with one of the designers of SH:CD that really gets at a key distinction between pure puzzles and the kinds of games I was thinking about in the original post:

For Grady, the difference between a logic game and a mystery is that a mystery gives you unexplained events from which you must deduce a larger context; a logic puzzle tells you what happened and leaves you to infer the specific events that make up that context. “In real life, if a policeman walks into a room and finds a dead body, that’s all he knows.”

I suspect that looking at things through that lens makes accepting the game’s problems easier.


A few years ago I played through that Sherlock Holmes board game (we found an old box in the cellar) and thoroughly enjoyed it. We didn’t really play it “properly” though, as we just worked together and tried to follow up every clue and squeeze every piece of info out of it, only once we were sure we’d read everything, did we go for the resolution questionaire.

Good to hear that its been re-released! There were two expansion packs published back in the day, and I’ve only managed to track down one of them. Hopefully they will also get re-released!


There’s still one expansion to the game I know of. It’s a great game.

Oh, and now that we are talking boardgames, Tragedy Looper has a pretty interesting take on procedural deduction. It think it might fit the bill.


There were actually four expansions! Two were very similar in structure to the base game: The Mansion Murders (five cases) and West End Adventures (six cases). These shouldn’t be too hard to get from Ebay.

The other two were huge cases, spanning multiple days and/or multiple cities, with a playtime of somewhere around 15 hours: Adventures by Gaslight and The Queen’s Park Affair. At least Adventures by Gaslight is somewhere between hellishly expensive and totally unavailble.

My understanding is that they can be thought of as being 5 separate cases, one per day of gametime, but with very strong links between the “subcases”. So you don’t need to play through in a single 15 hour sitting, but probably shouldn’t take breaks of more than a week.

So what they’re doing first is Jack the Ripper & West End Adventures. That’s republishing the 6 cases from the original West End Adventures, and a totally new campaign of 4 new cases for Jack the Ripper.


I was confused when I read your post, because I had actually bought the expansion for my sister… It’s been out for a while.

So, after some “detectiving”:

…it seems the Mansion Murders expansion has been reprinted (at least in French and Spanish) as Carlton House. Alas, it seems there’s no English version, nor can I see any planned. It’s 8 cases, with the 5 original ones and 3 new ones to be played with the base game.

With the West end/Jack the Ripper expansion (which I hope we get in Spanish, but I guess I can buy the English edition just fine if we don’t) we will have 2 of the original 4 expansions in print.

PS: this is one of the very few times I see things available locally that are not available on the American market. I make a point of buying most of my boardgames in English (at a slight extra cost) so to avoid expansions not being translated (a sadly common occurrence).


Yeah, it becomes a huge mess once you look at the different language editions. Ystari also did (parts of) the Queen’s Park Affair in French, but I guess not in Spanish?


Afaik all we have here is the base game and the boxed Carlton House expansion. So I don’t think so unless some of the extra cases of the Carlton House expansion are actually part of the Queen’s Park Affair.


I played the mansion murders expansion, so missing the West End one from the original set. Otherwise there seem to be a few fan made expansions listed on board game geek.

Here’s one you can download for free - Murder behind closed Doors.

And another one - The Diavolo Case.


Edit: …and another one - case of the clay potts.


Ystari/Asmodee is reprinting the old expansions in English soon according to an interview. First in Q4 they are doing all new Jack the Ripper cases with West End expansions as additional cases, then reprinting the base game with corrections, then in 2017 doing Queens Park and Carlton House together. All three are supposed to have a consistent design and new maps.

Edit: sorry, Jsnell already linked the first one.


Other possibilities:

Gumshoe, a Chandler/Hammett style mystery game in the vein of SHCD. Never played it, hard to come by on Ebay.

Mythos Tales a Kickstarter Cthulhu SHCD based on a print and play but rewritten and with 7 new cases. It’s been printed and should come out in a month or two.

Circuit’s Edge is a late 80s cyberpunk adventure game where you play a detective. It’s been 25 years since I played it but I remember it being quite linear, while definitely having a gritty hard boiled atmosphere translated into cyber future.


Her Story is an odd one.

The framing device is that you’re accessing a mid '90s police interrogation database, which contains video clips from the interviews done with a woman over multiple sessions. The database has been corrupt though, and you only have access to her answers, not to the questions asked by the police. The unstated goal is to view the clips and find out what happened.

There is no actual interaction, so the basic structure is very similar to e.g. Sherlock Holmes: Consulting Detective. The main twist in how the set of videos available to view is determined.

The old archive computer has a search engine, which when given one or more search terms produces the 5 chronologically earliest clips that contain all the terms. You view those clips, and then make a new search. The restriction to show only the first 5 clips is the core driver that makes sure that the game doesn’t just degenerate to just linearly seeing the story from start to finish. To access the later clips, you’ll need to find terms that are present late in the story, but either absent or very rare early on.

I felt Her Story suffered from some structural problems:

First, it’s great that the narrative is non-linear. But as implemented, it’s way too easy to stumble to clips telling too much too early. I think I accidentally found the core of the mystery somewhere around word 7 or 8. It wasn’t that I was probing for that thing specifically. I searched for a fairly innocent word, and that clip happened to have some major information and a bunch of very juicy keywords to continue on with. I then had almost full coverage of the most revealing sequence while still having very spotty coverage of the bulk of the story.

That’s not what I want from a mystery investigation. It doesn’t leave me feeling clever about having interpreted the clues right, and pursuing the right lines of inquiry. Instead of having the drive to slowly uncover the mystery, I was trying to fill in the blanks just to make sure there wasn’t something important there that I’d missed.

This seems like a fundamental issue with the use of a full text index as the way of finding the next bit of data. It’s just going to be too hard to make reaching the later clips challenging but possible.

Second, Her Story is very aggressive about giving the player no closure. How do you know that you’re done with the story? Well, you just decide that you probably know what happened and stop play. How do you know whether you got it right, that you got the full story? You don’t. You can’t even find it out online, since there are two competing theories both with rabid supporters. (With the outcome that the inevitable plot holes have been revealed, and the only way any theory works is if you just arbitrarily reject some information as untrue while accepting other bits). I find this deeply unsatisfying in a game like this.

But despite these issues, Her Story was a decent enough way to spend a couple of hours. The mechanics and storytelling of the game aren’t really as innovative as the’d been made out to be, but the FMV really carries the day. I liked the acting. And I really liked having to interpret the videos for visual and behavioral clues, not just relying on text. It really adds a lot to the experience. (And must also be really hard to execute well, especially for an indie game).

Incidentally, the “old computer archives” framing device is brilliant. The usual problem with this kind of non-interactive structure is the dissonance between supposedly running an investigation, but not actually getting to ask the right questions. It’s infuriating. You know exactly which bit of information is needed, and there are characters that could provide that information. The bumbling detective refuses to ask the question, and you have no control over it. But when the investigation happened a long time ago, and you’re just looking into the case now? Suddenly it’s the most natural thing in the world that you can’t ask the obvious question.


Anyone come across Knee Deep?

You play three different characters (blogger, journalist, private investigator) all looking into the same case from different angles. I believe it’s relatively short (3 hours or so?), which makes the full price look a little tricky unless they are going to add further stories. Some of the dialogue is very painful but the voice attacking is pretty decent and the storyline intriguing. The style is particularly unique - it’s set up like it’s a theatre production, with occasional audience reaction. You don’t have to do any walking about, it just highlights things to look at, and when you are in conversation you pick a response. Several scenes end with you filing your take on things to your employer, and your decisions here will affect what conversation options and NPC responses you encounter.

Definitely worth a look if you are into this sort of game.


Played LA Noire, totally loved it.


I played Consulting Detective and all the expansions back in the day. Still have the 80’s versions. The original and Mansion Mysteries were probably the best.

They’re really good, but the comment that you keep searching when there’s nothing left to find is true. It is also closed to a shared reading experience than a proper board game… I think it’s best with 2 or 3 players as a result

Also have Gumshoe, it’s sequel set in noire San Francisco. 6 cases. Similar, not as good, and it does some choose a/b stuff that makes it feel even more like a Choose Your Path to Adventure book. Oddly, it has a thread through all of the cases that almost requires the same set of gamers for each case.

Finally, there will be a new board game next year in this vein. Different mechanics entirely, but realistic mysteries. 8)


Watson & Holmes: From the Diaries of 221B (2015) is a board game that’s clearly inspired by Sherlock Holmes: Consulting detective, but goes in a very different direction. The big difference is that it’s a purely competitive game.

A short introduction gives you some starter clues, and a question to be answered up front (e.g. “How did the victim die”, “Who was responsible for it”, and “What was in the safe?”). After that a set of about 15 location cards is placed on the table, with the only thing that the players know about the contents being the name of the location. All players will select one to read, using what is effectively a penny auction. Everyone reads their card (anything from a few sentences to five paragraphs) and takes notes. After the round is over, you’re not allowed to refer to the location again. So you’d best make sure the notes include everything you really want to know.

Instead of visiting a clue location, you can go to 221B to answer the questions. There’s also a couple of special power that can be used to lock other people from visiting a specific location, unlock a locked one temporarily or permanently, or force someone to read their location aloud.

Later scenarios will introduce small tweaks to the game mechanics. In one you don’t get the questions up front. In another the locations are split to two sets, and moving between the sets costs some of the bidding currency. More locations will start to give special power tokens, etc. Especially this last thing I don’t like a lot; it kind of distorts the investigation. when players right away visit locations they shouldn’t care about just since there’s a special power token available there.

There’s a lovely tension during the bidding phase as you’re trying to figure out what other people know, and why they are going to the place they’re going to. Did they find out something in the previous location? Do they really expect to find out something useful in the location they are going to, or is it just a shot in the dark? Are they chasing the same theory you are or something else? If you’re low on the bidding currency, should you go to 221B even with a half baked theory just to do it unchallenged? (If two people want to solve on the same round, the player with more currency gets first dibs).

One interesting bit is how much the order in which you visit the locations matters. If you find out when visiting the body that the body was hit from the back with the left hand, you’re probably going to keep an eye out for even slight hints about whether someone is left- or right-handed. Likewise there are some locations that are going to be totally useless unless you have a very specific bit of information from elsewhere.

The writing and the design of the cases has been mostly good, with a couple of slightly unfair (but still solvable) bits. The one thing that is completely infuriating that Holmes refuses to ask the most obvious questions. Somebody will mention that once everyone else left a dinner party, four people still remained at the table. But will they tell who those four people were? No. Will Holmes ask? Hell, no. You need to find out the information the hard way. There are also some cases where a story beat is split in two locations, but it’s extremely unclear which location actually has the second half.

There are two things the players need to accept for the game to actually be enjoyable. First, people should be good sports and take notes sensibly rather than copy every single word. If people can’t do that, you might need to introduce a note taking timer. Second, the cases don’t often provide absolute certainty. You need to read through the lines, but worse yet you’ll occasionally need to make guesses just based on what appears in the text and what doesn’t. “Well, the murder weapon was a heavy object and the only heavy object I’ve seen was this bust. Guess it’s probably that”.

Watson & Holmes scales pretty well to large player counts. We played one game with 7 (all others with 4), and it worked just fine. The actual gameplay of reading + taking notes is done really well in parallel. And it also adds much more contention on the board, making that part more tactically interesting.

We’ve played 8 out of 13 cases so far, and Watson & Holmes has continued working well. A case takes about 60-70 minutes, so we can fit these in to our weekly lunch hour game session, where something like SH:CD would be impossible. (Of course it’s also a much slighter experience than SH:CD). I’ll almost certainly pick up the teased expansion if it actually comes out.