Premature speculation

I say, comrades!

Sherlock Holmes once said that it is a capital error to theorise without sufficient data. All but one of the roleplayers I have GMed for have demonstrated that it is also an irresistible temptation.

I run — or used to run when I GMed — a great deal of material that is or was essentially investigative. Mysteries and investigations, obviously, but also scams and capers and covert ops adventures in which the PCs had to scout the opposition and gather intelligence before planning their operation, intervention, or coup de main. In running this stuff I have run again and again into the same problem: that the players fall into interminable speculation and planning before they have collected enough information for theorising to be a useful activity. For example, police detectives will take the initial report and examine the scene of the crime, and then collapse into out-of-character speculation without either (1) canvassing for witnesses, interviewing family and contacts of the victim etc. or (2) testing any of the speculations by an empirical test. It’s sometimes called “analysis paralysis”.

I have explained again and again that a role-playing adventure is not a puzzle¹, that for an adventure to occur the players have to get into character, and that the characters have to (a) go somewhere and (b) do something. It has been to no avail.

Have you fellows encountered this in your exploits? Do you have any suggestions of how to deal with it other than by luckily having a forceful player who wasn’t susceptible to it and casting him as the boss?


¹ One of the players² in one of my longest-running and most successful investigative campaigns³, having reflected on the party’s long career of triumphant successes summed up thus: “We never solve any of these mysteries. What happens is that the villains over-estimate how smart we are and severely under-estimate how deadly we are.” That’s because a good RPG mystery scenario is not a puzzle: it is an adventure of which solving a mystery acts as the objective.

² It was Phred Smith.

³ It was the first Flat Black campaign, in 1987–88. The PCs were a team of criminal investigators in what was then called the “Justice Department”.

I haven’t seen this particular problem, but my usual advice from having run and played quite a lot of investigative games is “read GURPS Mysteries”.

I’m reading the Kate Daniels urban fantasy series at the moment, and there’s enough self-awareness for Kate to point out in the early books that she’s not actually all that much of an investigator: what she does is find some plausible candidates for bad guy, and annoy them all until the actual bad guy attacks her. That’s an entirely viable solution to a mystery scenario in an RPG.

When the players get to talking rather than doing… that’s a player management thing, I think, and sometimes if everyone’s enjoying it the best thing can be to let it go. (Similarly with @MichaelCule’s players and their shopping expeditions.) If it drags on, throw in another corpse…

My favourite players developed a playbook of “methods to deal with one of Brett’s investigative adventures”: the James Bond gambit, the Mike Hammer algorithm, the Quiller gambit, Kirth Gersen’s algorithm, Kobayashi’s algorithm, Sarah Sakata’s rule, the Columbo method….

1 Like

Okay, I can guess at the Mike Hammer algorithm and the Columbo method, but the others are opaque to me.

1 Like

The James Bond gambit was to get captured in the hope that the villain would explain his scheme to you or take you to his otherwise-undiscoverable lair. @frank.hampshire particularly disliked this one.

The Mike Hammer algorithm was to grab someone connected to the case and deal harshly with them until they gave you the name of someone connected to the case, and loop recursively until you reached some sort of terminating condiction.

The Quiller gambit was to behave blatantly but mysteriously to as to bluff the opposition into thinking that you were doing something they had to stop.

Kirth Gersen’s algorithm was to grab or stake out the McGuffin (or anything else that the opposition wanted badly) and use it as bait in a trap.

Kobayashi’s algorithm was a database dragnet. It involved starting with a sort of Fermi estimate of the number of people on the planet who matched the data known about the perpetrator, and when that got down to a manageable number started checking them all for connections to victims, places, and organisations involved in the crime. (“The population is 930 million. Half of them are men. Half of those are about average height. Half of those have medium builds. Half of those are “I dunno. Neither young nor old.” 85% of those are right-handed. Half have black hair. One in five hundred own Zegnatti shoes. That’s fewer than 50 thousand suspects. Get me a list.”)

Sarah Sakata’s rule was “always check the body” (and “attend the autopsy” if possible).

Finally, the Columbo method was to annoy and pester a suspect until he of she did something you could have him or her mindwiped for.

1 Like

It might be of some minor interest that this method has at least one much earlier precursor. In Crime and Punishment, Porfiry Petrovich, the detective investigating the murder, uses a very similar method to interrogate Rodya Raskolnikov, questioning him about an article he wrote some months ago, “On Crime.”

1 Like

Perhaps I can sum it up by saying that the players take the stance of the readers/audience of a whodunnit instead of taking the stance of the characters in an investigation. They neglect to exercise their characters’ agency in gathering and especially eliciting information. The dual result is that they can’t solve the puzzle because they don’t have the pieces, and they don’t have an adventure because all the time is taken up by out-of-character discussion and the characters never do anything.

Ah, the Doctor Who strategy!

That’s just good sense in most games, really.

I find that many of mine are rather too good at eliciting information from one particular source: the game master. They read me rather than my NPCs. I guess that’s mostly a flaw in my GMing style.

One of my players used to engage my major NPCs in discussions of history and moral philosophy to work out whether they were good guys, bad guys, or what.


That’s actually perfectly in keeping with some detective fiction.

Oh yes! I’m not complaining, I had fun with it. (And it wasn’t just in detective investigations.)

1 Like

I’m not sure I’ve ever had this exact problem with my players, though I’ve had groups where the mystery has stalled due to other behaviours. These can be summarised as:

  1. Something that looks rather like your analysis paralysis, but is caused by the group naysayer.
    “We should use the James Bond strategy!”
    “No the bad guys will just execute us without telling us anything.”
    “We should use the Quiller gambit.”
    “No, we won’t be able to think of anything mysterious enough.”
    etc etc

  2. The risk averse group/s. One lot wouldn’t do anything which would put them within stabbing or sidearm range of a major bad guy, in case they got hurt. This was annoying because they were playing SAS soldiers, hard-bitten James Bond types, etc, who should be risk takers. Another group wouldn’t talk to cops or journalists or pathologists, because… er… um… frak knows. This was annoying because (a) they were werewolves and thus immune to anything a normal human could throw at them, and (b) part of their mission was to find out if any inklings of supernatural stuff had leaked out.

  3. The group with a Bossy Boots and a reluctance to split the party. They should be interviewing the butcher, the baker and the candlestick maker. Bossy Boots insists the butcher is the only witness of importance, so they dedicate vast quantities of time to him, and ignore the other witnesses.

  4. Ooh! Squirrel! They need to talk to people or do forensics or whatever. Unfortunately the player who has the [delete as applicable] talking/forensics/whatever skills is - yet again - playing a character who is more interested in watercolour painting or ballet or dress-making than following the plot. So instead they send the PC who has the social skills of the Incredible Hulk, and it doesn’t go well.

  5. The 3 Clue group - who expect to be able to solve everything with three or fewer clues.
    “Right we know the murderer is a dark-haired man with an Aberdeen accent. We know he drives a blue transit van. We know he used a magic knife to cut out the victim’s heart. Dammit, why can’t we find him? It’s not fair!!!”


Your “3 Clue Group” sound like a version of the problem I have so often had.

Mysteries are tricky from a player’s position. A common mystery genre trope is “solving the mystery with the least clues possible shows skill and expertise” and another is “if you take too long looking for clues the villain will have time to win before you can stop them.” This combines poorly with rpg “rubbery” time where a lot of stuff that takes endless table time takes virtually no time “in game” and other stuff that takes a lot of “in game” time is fast at the table. Dr Bob’s observation about reluctance to split the party is related - many many RPG tropes inflict terrible punishment on any group that dares to split up.

So the incentives and expecations are misaligned for many players.

You might get a lot of traction from some variant of:

“The actual time limit on this mystery is screen time - we’re emulating a movie, not reality. “In game” time, within reasonable limits, does not matter. However, if you don’t solve this in three sessions the villain’s plan will come to fruition at the beginning of the fourth.”

Never done that, but it might work.

I think that the adversarial GMing style of the early dungeon-bashing days (“if you didn’t say you were lighting your lantern, I won’t tell you you’re in the dark, I’ll just assess penalties for doing stuff”) casts a long shadow even over groups that didn’t necessarily experience it. When I’m GMing a mystery adventure, my primary objective is for the players to have fun finding the solution just in the nick of time.

1 Like

Splitting the party can suck from the player’s perspective. The GM can only really interact with one group at a time, so if your character is in the group not being serviced, you have to sit there with your thumb up your butt. There are ways to improve that, but it’s a hard problem, especially in small groups, where there might just be one character not involved in the action.

This is one of the few things that is easier to do right in play-by-mail and other asynchronous play than it is in face-to-face games.

You can also run into amusing problems when the party gets split up. I was in a long running D&D campaign where the party was split in two because we failed to stop an earthquake. One group of characters went one way, the other another. The groups were large enough (and we were youngsters with free time, sigh) that we met for separate sessions. Later, when we met up after several months of real time, and a year or so of game time, we realized that both groups had been in a small city, for the same religious holiday, as honored guests of the same temple. It was a month or so time difference in the real world, so we couldn’t really meet back up.

Speaking of splitting the party up:

In first edition Werewolf werewolves gained Rage the first time that they saw the Moon each night. Whenever a PC gained Rage they had to make a roll to maintain self-control. If they failed that roll they had to either flee from or charge and attack whatever had caused them to gain Rage.

I was in a party of four PCs, one of which was low-Rage enough to always make her rolls to control herself. The rest or us almost always failed. One always fled the Moon. Another always charged it. The last had a Gift that made him move twice as fast as the others.

Fair, though some campaign styles do this anyway - in @JGD’s Infinite Cabal campaign, while we may all be in the same place, we’re often interacting one-on-one with the GM for a few minutes and then stepping back while someone else does it.

“I feel a bizarre urge to go to Cape Canaveral.”

1 Like

This does rely on the players in the spotlight being reasonably entertaining, which is usually accomplished.

1 Like