Investigation in the Big City

A city on a hill, where even the lowliest peasant
has the calf muscles of Adonis

I've been running games in a massive city for a while now. The city is called Garton, capital of the country of Garlancia. It's large and sprawling, on the level of Baldur's Gate or Waterdeep in the Forgotten Realms. It's a walled city, but farmers and beggars live outside the walls and are (mostly) protected by city guards. There is a lot to do, shops on every corner, guild houses, and a noble district complete with royal palace and city parliament.

I think most people would like to run a huge city game like this, but it can be pretty daunting. The sheer amount of available options can cause creative paralysis. Fortunately, I've been experimenting with different ways to handle such an influx of information, and I've landed on a good solution.

Investigation in the Big City



In fact Arabian Nights are not at all like Arabian Days

First things first, you need a reason for the characters to search the city. I find that using the Three Clue Rule is a good place to start. Players need lots of guidance to get on the right path, because what is obvious to you is shrouded to a player. So make sure whatever they are searching for has plenty of paths leading to it. I usually go for double the three clue rule and put six paths that lead to where I want the players to go.

Second, you need some alternate things to find. Use this as a chance to show off some cool areas of the city. Is there a statue, palace, or eccentric citizen that can be showcased? Also, check out your player's skills. Is one of them far better at a skill than others? Stealth and performance lend themselves to certain classes, and it's unlikely you'll have more than one person with a high value in each of those skills. If this is the case, you can put character-specific plot threads into the list as well.

Finally, (and this is going against the Three Clue Rule and conventional wisdom) you can choose to throw in one Red Herring. Just make sure that it clearly contradicts some of the actual evidence, so if the players choose to follow it they would have to ignore other evidence. That will make their choice matter.




On this search you find... a dead body holding a tin can full of worms

Now, write up a short scenario where a player encounters each of these clues, landmarks, plots, and diversions. The scenes should be generic enough that any player could find them (except the character-specific ones), but also contain information that won't hinder the game if it isn't found. We want there to be a consequence for failing to find something in the search.

Here's an example from one of my games for a landmark:

You don’t find anything of importance today, but you do manage to get a good view of the royal palace. It looks like the tower they were building is finally finished. You head back to your base and get drinks with your companions at the nearby tavern.

Note that this particular search happens over the course of a day, This doesn't have to be the case, you could easily do an hour per search. But remember that each item on the list should take about the same amount of time to find.

If you live here, you can compete in illegal underground sewer-surfing races

Next, print out your scenarios and cut them up into individual notes to be handed out. Once you have that, assign a skill and DC to each search. Easy searches (landmarks, simple clues) should be low DCs, difficult searches (red herring, complex clues) and character plot items should be high DCs. If you have a search that leads to your Bard's long-lost stage manager, you'd better make that performance check high to deter other players away from attempting the search.

When a player performs a search, they will roll a skill check and try to meet the DC associated with the search. If they pass, give them the slip of paper. They can then read the information and corroborate it with their allies. Depending on the amount of information on each search, you could allow them to discuss after each round, or you could say they are out searching until the end of the scene, and have them discuss their clues all at once.

Session Implementation

During the session, you have some options. You could:

1. Lay out all your clues and searches face-down on the table, marking on the back what the skill and DC required is
2. Assign certain clues to certain city areas, or certain days of searching, and slowly release the clues over several rounds.
3. Make several stacks of clues, with the easy searches on top and the difficult searches on bottom, to reflect the players becoming more familiar with the area over several rounds.

If you choose to use rounds, I would give the players one extra round to search than the number of clues in each pile/area. People are going to fail, and you want them to have a fighting chance to try again.




If the city has floating areas, make sure they are home to the prissiest of nobility

On the subject of failure, I would also prepare a list of common tasks that you can run with players that failed their search while others are reading their clues. Simple things, that require a single roll that the player can choose to make. If they pass, they gain resources (gold, ale, items, whatever). If they fail, they lose resources. Gambling, drinking, shopping, etc could all be reduced to a single roll and used while other players read over their clues. I would use a random tavern name generator for this.

Whichever method you choose, make the scene more interesting by limiting the amount of searches the players have available. This is part of why I prefer to expand the three clue rule to six, at least. If players feel like they are making a choice between investigations, they will be more compelled towards the scene.

Also, putting an in-game time limit on the searches can also help increase the scene tension. Tell the players they hear a tip about an assassination happening this evening, and they only have a few hours of daylight left to figure out where it will be taking place. Have them search for a plot-important NPC who is known to leave town quickly. Or have them try to find the location of a dark ritual before the evil cult completes it.

Results of the Investigation




See how far you can take this before your players realize you're just running National Treasure

In the end, your players will hopefully gather enough clues and make the right connections to lead them to the subject of the search. If they don't, the time limit is a good way to force them to make a choice rather than keep searching. Also, if they do choose the wrong path, follow the red herring, etc. you will want to make their failure apparent. Don't rub it in their faces, but make it clear that they chose wrong and the search failed. They hear the city guard talking about the assassination or the person leaving town, they see a beam of light or feel the ground shake that indicates the ritual is starting.

However, remember to give your players time to discuss and review their information. Each one of them is going to have an incomplete picture, and they will need a chance to figure out how the clues fit together.

Also, feel free to link several searches together: do a couple rounds, play out a scene at the search location, then do a couple more searches. This can help break up the monotony of searching and give the game a more hectic, lead-following feel.

And you will also have shown off your city, advanced character plot, and allowed them to "wander the big city" without having to map out every building and street. If they don't find certain clues, plots, or landmarks, then it's easy to set those scenes aside and put them into later games.




Also works in Sci Fi, just make sure you include teleporters and laser carriages

So, that's how I prefer to do investigations. It's a little extra work typing up the scenes (my upcoming investigation game has 30 of them...) but I've received excellent feedback from my players in regards to capturing the feel of searching for information and deducing the solution.

Thanks for reading!