Limiting Hijinks by type in a settlement

I've been working on some rules to enhance hijink play a little (mostly streamlining guild management a little and creating rules for guild wars) and one of the things I've been working on is a limit to the number of hijinks that can be performed in a settlement per month. this is important because criminal guilds wouldn't fight unless there's a limited pool of revenue to fight over.

At first, I just thought of using the "Hideout Size and Cost" table on page 135, but that creates the pervers eincentive of puttign as many of your ruffians as possible doing treasure hunting, which seems silly. So instead, I thought I'd apply a constraint on a per-hijink basis. Each type of hijink can be no more than x% of the total hijinks in the settlement. This also has the advantage of letting rival guilds divide a settlement up by type of hijink (e.g. a settlement might have an assassins guild that does assasination and spying, and a theives guild that does everything else).

I bodged a table of percentages together based off the epxected revenue from each hijink, conditional upon the hijink succeeding:

Hijink exp. value Share
Assassination 1000 5%
Carousing 97.5 55%
Smuggling 359.4 20%
Spying 1300 4%
Stealing 359.4 15%
Treasure Hunting 3500 1%

How does this this look to you? Should I try factoring in odds of success as well? Do any of these categories seem over or under-weighted?

Personally, I'd have used hard caps based on market class rather than percentages. (IE: A Metropolis generates demand for 32 assassinations per month, 400 smugglings, etc.) It works out the same, but the players don't have to figure out what 4% of the hijink-pool is. I'd probably also include some modifiers (Domains with a low morale score will have extra assassinations, domains with high taxes will have more smuggling...)

What hard caps do you use? i really like the idea!

Well, I don't actually set hard caps, because I haven't DMed for a domain-level thief running hideouts (our thief instead chose to set up normal domains for his vassals and be a puppet-king.) I'm just saying that's how I'd do it. But now that I've been asked the question, I have no choice but to think about how to do it. 

I'd say the best method would probably be to take the Starting City Criminal Guilds table, because that tells us the monthly revenue of a guild, and it specifically says that if there are TWO guilds, split the income based on their size. So we know, for example, that there are only about 22,000 crimedollars to be made in a class 3 town. 

So now all we need to do is figure out how to assign those 22,000 illicit coins to crimes, and we're set! Naturally, the most ACKSy way to do it would be to design some kind of vagaries+demand modifiers abomination lovechild to allow you to roll every month to see what kinds of crimes are hot. Admittedly, it would be a lot of work, but it seems like it'd be really cool to have the assassin's guild getting frustrated because they've hit a 'dry' period where nobody seems excited about murder, or to suddenly declare a 'smuggling season' where everybody and their dog is bootlegging. 


haha, i like the idea! if i coume up with some numbers i post them here.

Thanks for reminding me that revenue data exists Susan. That page also includes level demographic data for typical syndicates, so that's nice too.

After a little analysis though, I am left confused. The average revenue per ruffian drops as guild size increases. This is especially clear for Class 1 markets. I hope Alex can shed some light on this because it doesn't make a lot of sense to me. Large cities are more prosperous than small ones, so the pickings should get richer, not poorer as the city expands, shouldn't they?

After susan_brindle's excellent suggestion to look at the revenue per guild information, I've come up with a hijink table by hijink type and market class:

Class Assassination Carousing Smuggling Spying Stealing Treasure Hunting
VI 1 13 6 1 4 0
V 3 25 11 2 8 1
IV 6 50 22 4 17 1
III 21 188 83 17 62 4
II 41 379 165 33 124 8
I 165 1515 660 132 495 33

Note that this table does not perfectly replicate the revenue given on page 237. For one thing the expected revenue per ruffian fluctuates by Market Class in ways that make no sense to me, so I've employed an average. In particular, this table will give massively higher revenues for Class I markets (about double, in fact). Still, the revenue per urban family is still lower for Class I markets, so I think it'll be fine.

I haven't examined the numbers, but there's something about there being 165 murders for hire each month in a class I market that seems high to me. Never mind the regular old "murders for free" that happen in any society. Am I misunderstanding something? Were urban environments during the Late Antiquity that ACKS is loosely modeled on really that violent? I know that in our modern major cities we sometimes have had an annual number of murders in Chicago or New York that could approach the numbers given in the table above, but only a small minority of these were paid hits.

Now that you mention it, it is a lot of murders isn't it?

Admittedly, looking at the demographics of theives guilds, only about 132% of those 165 murder attempts will actually succeed. And according to Stephen Pinker's The Better Angels of our Nature, murder rates in medieval Europe were as high as 100 murders per 100,000 people per year, so it might not be as far off as you would think.

Still, in checking the numbers, I just realised I had underestimated the revenue per hijink, so I need to fiddle with these numbers anyway.

Thanks for the observation.

Ok, based on wmarshal's comments I have a new verison of the table, plus correcting an error I found. There's a whole lot more carousing in this verison.

Class Assassination Carousing Smuggling Spying Stealing Treasure Hunting
VI 1 21 1 1 1 0
V 1 44 2 1 2 0
IV 3 88 3 3 3 0
III 10 331 12 8 12 2
II 19 661 25 16 25 4
I 77 2647 100 62 100 14

"What is there to do in this town"
"Well, every night we get drunk."
"Is there.. anything else?"
"First saturday of the month we kill a guy. Second saturday we steal something. Third we sell something stolen. Last Saturday of the month is the Lord's bathing day so we all go watch."

That town is going into my upcoming game. It’s just too perfect to pass up. I shall call it Brindleton in your… honor.

Whoo! May it fare better than the other towns that carried its name. 

[quote="James K"] Admittedly, looking at the demographics of theives guilds, only about 132% of those 165 murder attempts will actually succeed. [/quote]

Only 132%? How much collateral damage did you expect?



Admittedly, looking at the demographics of theives guilds, only about 132% of those 165 murder attempts will actually succeed.

-James K


Only 132%? How much collateral damage did you expect?



Yeah, that was supposed to be 13%

Coming at this one a little late, but:

Assassination, in particular, has always caught my eye as it has an extra note that the assassination itself is "carried out on a target within 1d2 levels of the perpetrator's level" ; and an implication that an outside entity has ordered the hit. (to a lesser extent, something like Spying could also be an outside order more often).

Ignoring the determination of the desire to have someone killed; there's four factors possible involved: the level of the person ordering the hit, their disposable income; the level of an available assassin, the level of the target. And we know via the demographics tables what we've got where.

It'd be an interesting exercise to correlate the assassination attempts ordered versus the demographics of the payers, the killers, and the targets.

One would assume that most of the assassinations would be ordered by the upper classes because of the high cost. Still some of the targets could be without levels, so could the price of a hit on a normal man be extrapolated as 100 gp?

Manuel Eisner ("Long-Term Historical Trends in Violent Crime", 2003) estimates the murder rates and also gives the data from Ruggiero (1980) on the social status of all murder offenders in early reneissance Venice.

The homicide rate in Medieval Europe varies from 3 to 200 per 100 000, with an average of 32 in 13th to 14th centuries, 41 in the 15th and 19 in 16th.

The social status of murderers (1324-1406 in Venice) are following:

  • Nobles 4%
  • Important people 9%
  • Workers 70%
  • Marginal people 16%

Nobles were around 4% of the population. Important people include merchants, professionals and civil officials (ca. 10% of the city's population). Workers include laborers, craftsmen and mariners (75% of population). Marginal people include vagabonds and beggars (8% of population) and are where we would put the thieves and assassins. NB! The data is probably skewed as not all crimes of the lower classes were noticed or handled by officials.

What is lacking from that data are the victims' class. With the bias on reporting, we could say that some of the murders by marginal people were noticed only because the target was of a higher class. These would include the actual hired hits. If we had any numbers on the prevalence of murder for hire (compared to all homicide) and the social class of the victims, we could match those with the demographic and economic data in ACKS.

Some estimates put the amount of assassinations (contract killings) at around 3% in contemporary US and 4% in Australia. This is close to the magic ratio 1/33 of all murders. If the ratio was the same in pre-modern times (which I doubt) and the medieval European murder rate was 30/100 000, that would result in only around one assassination per year per 100 000 urbanites (20 000 families), and less than 1 per million in a month. 

I've just posted an update to our Patreon called "Of Coins and Coinage" that provides some detail on how much income an urban family makes. To summarize what is explained there, while an average peasant family produces 16.25gp and pays 12gp to the lord per month, an average urban family is producing somewhere between 25gp and 75gp per month. Median income is about 6 to 12gp per month, but the cities are where the wealthy live and they drive the average way up.

Hopefully that data helps you assess whether or not the percentage of income earned by the syndicate makes sense.

I found some more contemporary data on contract killings. The highest modern rate was 5/100 000 (Mexico 2013), which was 28% of all homicides. With this really high percentage the rate of assassinations would still be less than 10 per annum and less than 1 per month (per 20 000 families). Maybe some of the assassins get hired for leg breaking and other assaults, maybe even arson and destruction of property. The 1st level assassins are thugs and could do thuggish stuff for living. Unless something like this is added to the hijinks for them, there won't be many opportunities for income and xp for any assassins. The competition for the rare assignment is fierce unless there's only one assassin per 3000 other syndicate members. If so adding a level 6 "real" assassin to the ruffian list would fit the need.

What should be the fee for these other violent hijinks? 500 gp like the level 0 targets?

Some more digging of crime statistics (including income structure of organized crime) could help to set the percentages for different hijinks. Also a closer look at the urban demography and distribution of wealth is needed to create a random table for the assassins' assignments.