The question seems straight-forward, but I'm having trouble answering it to my own satisfaction.

Is it multiple red herrings? Strong characters? Atmosphere, particularly if you're going for noir? Should it be confusing or simple?

I enjoyed Hammett and Chandler and Doyle when I was younger, and I just read the Strike series (and obviously, I liked the early Dresden books). And now, I'm thinking about writing one myself.

So, I'm asking: In your opinion, what makes a good murder mystery book?

 

So, I think murder mysteries can sometimes get a little to hung up on spreading clues throughout the story. To me, it is not so important to figure out who the murderer is, I'm much more driven by the characters and the general atmosphere of the text.

Are you going for any particular type of murder mystery? The market is quite saturated, so many authors these days seem to be trying new twists by having the main character perhaps not be a police officer at Scotland Yard etc.
Don't know if you've read the Millenium Trilogy by Stig Larsson? I think he managed to take the genre to some new places and he emphasized two things: very strong characters and a deep knowledge of what he was writing about, sprinkling interesting details here and there.

Really, the Harry Potter series is at it's heart mystery stories and I think you can take some inspiration from them. Craft interesting characters, place them in an interesting world and half your job is done.

 

There are mysterious aspects to stories that do not fall squarely within the mystery genre. To write a true mystery story, the author must write a novel that is meant for the readers to have fun trying to puzzle out. This introduces two basic constraints.
The mystery can't be impossible. There are books like this and they're not that fun. It doesn't really feel like a mystery if it's like one of those Scooby-Doo episodes where the monster was a character to whom we had not been introduced.
Finding the correct answer with any degree of certainty must be improbable. If you know who is dressed up as the monster from halfway through, you might as well stop reading/watching. An important thing to note is that if a reader guesses correctly, but isn't remotely sure of it, you haven't necessarily lost points. If you have a large audience, and you've correctly addressed the first constraint, someone will probably have the right guess due to the infinite monkey theory if nothing else. This is why random reddit users have already figured out much of the long cons that GRRM was pulling in ASOIAF; they've had years and years to discuss the books, and sooner or later, if for no other reason than just by chance, someone will have the correct guess, and it will make perfect sense and get a million upvotes.
Constraint 3- Do your best to get the mystery done in one book.

One of the most reliable approaches mystery writers have been using over the years is basically just hindsight bias. In an experiment, subjects were shown like 30 pictures of random men, one of whom was a serial killer and asked to point him out. They never got it right, but when the answer was revealed, they were like 'oh, of course, he looks just like a serial killer, how did I not see it before?' and surprisingly enough, even when the researchers pointed to the wrong man, the subjects had much the same reaction. Authors will use this is by revealing vastly insufficient evidence to reach the appropriate conclusion, have the detective point out the correct conclusion the moment he receives the last piece of evidence, which to best effect is not shown before the reveal. A frequent mistake is not having any additional evidence, meaning the detective makes an unfounded accusation, but hindsight bias might just pull your ass out of the fire anyway.

A method I'll admit to having employed is keeping multiple conclusions basically tied in the amount of evidence they have, then revealing the answer as one of them, with the final piece of evidence being something that simultaneously undermines what was more likely and supports the less likely prior. This is something I would classify as a dumb trick more so than a method for writing cleverly, but in that particular story I couldn't find any other way of doing it.

The final thing that comes to mind is something that works spectacularly well with fanfiction. In this method, the missing piece is either evidence from the source material, or something else that for some reason the author does not have to explicitly state, like Bernoulli's principle. Don't ask me how that's the missing piece, it's probably an episode of Monk.

Try to have other people who are reasonably smart figure out the mystery of your story as you're writing it. This can be hard, but it's probably the best indication of how easy or hard your mystery is to solve.

 

For me, the dynamics between the characters is the most important. I think one of the best Thrillers out there is Dexter (only seen the show though so far), because you've got a main character with a compelling dark secret that will ruin him if anyone finds out. You have a foil character, a detective who is on the verge of finding the truth about him. It all keeps you on the edge. It's one thing to be able to pull the veil over the audience's eyes and have X be the unexpected murderer instead of the butler, but a good mystery's reveal always makes me feel like I got gut-punched in the best way. It makes me feel almost happy that it's them, because the author has built up such an interesting character, that I'm glad they were given a spotlight to that degree in the story.

 

I appreciate the feedback. What I'm hearing is, keep the focus on characger, and keep the pace up. Everything else is secondary. That lines up with what I figured.

 

Drop enough clues for the reader to be a participant. It doesn't have to be designed so that they're meant to solve the case before the main character or anything, and misdirection is obviously good, too. But you want to keep the reader guessing and jumping at every noise just as much as the protagonist. I think that might help with immersion and investment.

 

What makes a good mystery?
I'll give you the answer later.

 

Video that I have found useful in planning year 2 of Victoria Potter: