Look beyond the box: a looser concept for NPCs
Most game systems are great when it comes to a precise definition of what a character can or can’t do, but there are any number of occasions when the level of precision they impart and entail is overkill. The result is that character generation takes a lot longer than is really justified by the intended role of the NPC within the campaign.
While preparing an NPC for my shared pulp campaign, I had a moment of inspiration. A short-cut to NPC generation that I immediately put into practice and which immediately proved its value. Now, I’m going to share that moment of insight with GMs everywhere. I’ll start by looking at the specific case that led to this idea, and then move on to a more generalized description of the technique.
“Creepy” as a skill
The NPC that I was creating was a butler. While he was to be no particular threat, we wanted his personality, tone of voice, manner of expression, behavior, etc to be, well, “creepy”. Now I could have spent a lot of time specifying skills and abilities and disadvantages and so on in order to define exactly what “creepy” meant in terms of the system mechanics – but time was short and there were better things to expend it on.
The solution was to define “Creepy” as a skill like any other within the Hero System. Suddenly, there was a full set of game mechanics at our disposal – whenever he interacted with a PC, we could simply have him roll his “Creepiness” skill and – if successful – describe the reaction he had engendered in the listening PCs. If they wanted to resist the effect, we could have them oppose the roll.
In effect, we plugged an abstract quality directly from the character description into the game mechanics, hardwiring it to the rules system. It was that simple.
The Limits of the solution
The more involved with the PCs an NPC is going to be, the more closely that NPC should be built to PC standards. An NPC who adventures with the PCs should be as fully-detailed as they are.
A lesser standard is required for characters who are just there to be in a fight. There’s not a lot of need for background skills, but battle-related stats will need to be specified. In D&D terms, for example, the minimum would be Attack, Damage, AC, Initiative, HP, and Saves. In the hero games system, OCV, DCV, Damage, SPD, Body, Stun, and END would be the absolute minimum.
But when the NPC is just there to be a personality or an advisor, this approach offers a way to dramatically reduce the requirements, saving a lot of time and effort.
The General Technique
Using this technique is a two-step process.
Define an abstract quality
The abstract quality that the system is summing up should be defined as simply as possible, and should encompass as much of the uniqueness of the character as possible, preferably all with a single word.
Set the Associated Skill Level
You have two contrasting and possibly contradictory considerations in the second step. The first is deciding on the absolute frequency with which the abstract quality that you have described is going to come into effect; the second is the relative measure of how easily you want PCs to be able to overcome or resist the effect. If necessary, come up with two separate numbers and average them, but most of the time there will be a compromise available.
This requires that you really understand the system mechanics and their basis in probability. If you are talking about rolling 3d6 and aiming to get the target score or less, then you have to know how increasing the target by 1 affects the likelyhood of success.
Of course, this is really easy with a d20 system where each +1 growth in the target equates to a 5% increase in the likelyhood of success.
Assessing the relative level requires some understanding of the attribute scores that your PCs (or NPCs) can bring to bear in opposition or resistance to the desired effect. If the average resistance is d20+3, then d20+5 appears to give a 10% chance of success – but examining the statistics tells a different story. If the goal is to get the bigger total, as it usually is in these opposed rolls, then the actual number of cases out of 400 (20×20) that result in success for the d20+3 is 17+16+15+14+13+12+11+10+9+8+7+6+5+4+3+2+1 = 153 out of 400, or 38.25%.
How did I get that result? Well, the minimum result of d20+5 is 6, so the minimum result to beat that result on d20+3 is going to be 7. Seven minus 3 means that the lower score wins on a roll of 4 or better – which happens 17 times out of 20. With each +1 to the result on the d20+5 roll, the minimum result required for success goes up by 1 – so if the d20+5 roll is a 2, the target is 7, so an 8 is required, which requires a roll of 5, which happens 15 times in 20. The pattern obviously continues until you get to 0 times in 20.
That gives the string of numbers that I added together – the one that runs 17+16+15+… and so on. Multiplying the number of possible results on each die (20 x 20) gave me the 400.
I also cheated on the addition. If we set the “..+2+1” part at the end aside, and add the highest to the lowest, we get 17+3 + 16+4 + 15+5 + … and so on. It only takes some quick counting on fingers to find that the eight result is “10+10” – but there’s only one ten, so that also has to be set aside. So the total is 7×20 + 10 + 2 + 1 = 153. I only pulled the calculator out for the last part – converting 153/400 to a percentage.
Some more examples
There are all sorts of ways this technique can be applied. Here are just a few of them:
- “Emotionless” – on a fail, the character exhibits an emotion.
- “Rigid Self-control” – on a fail, the character exhibits a reaction when surprised, given bad news, given good news, etc.
- “Foolish” – on a success, the character does something stupid or silly.
- “Deceptive” – on a success, the character projects the reaction or impression that he desires to express and not his true feelings or intent.
- “Politician” – on a success, the character says or does whatever he thinks is going to be the most popular/beneficial regardless of his true intent or belief.
- “Charming” – on a success, the character makes the person he’s speaking to feel comfortable, safe, secure, at home.
- “Wealthy” – on a success, the character attempts to use money to solve whatever problem he currently has.
- “Pious” – on a success, the character exhibits his faith, resists doubt, disbelieves evidence to the contrary, etc.
- “Redneck” – on a success, the character acts like a redneck. (duh!)
- “Italian” – on a success, the character acts like a stereotypical Italian.
Any adjective can be treated in this fashion. Just choose your description and let it be your character!
Going Even Further
It doesn’t have to be just NPCs, either. This approach also works – perhaps better than anything else I’ve ever seen – from defining organizations…
- Formal
- Casual
- Greedy
- Public-spirited
- Arrogant
- Militaristic
- Progressive
- Conservative
- Radical
- Violent
- Pro-Farmer
- Bookish
…to governments…
- Stuffy
- Sanctimonious
- Defensive
- Paranoid
- Martial
- Subversive
…to laws…
- Protective
- Permissive
- Conspiratorial
- Vengeful
- Fiery
- Myopic
- Effervescent
…to anything else that can be described with an adjective! Weapons, animals, cars, wagons, pets, spices, novels, music, poetry, clothing, artwork, recipes… these can all exhibit a personality, as I discussed in With An Evil Gleam.
An organization example: “Florid 14/-“
“Florid” means ornate, flowery, showy, ruddy, or high-colored. So, an organization that is “florid” has:
- A lot of bureaucracy, especially detailed forms that need to be filled out and about which they are very fussy;
- A very strong and showy public relations department;
- A penchant for dramatic gestures and big, showy projects;
- A tendency to make big, even boastful claims;
- A liking for making speeches;
- The habit of using flowery and long-winded statements, filled with grandiosities and showy excess, and probably short of a lot of practical detail.
So whenever a PC interacts with this organization, or reads about them in the paper, or whatever, you simply have to roll the die and interpret the results.
A government example: “Repressive 5/- Paranoid 5/- Blunt 5/- Militant 10/-“
There are 25 combinations when you have 4 attributes. Any given action by the government can be:
- Repressive but not Paranoid, Blunt, or Militant
- Paranoid but not Repressive, Blunt, or Militant
…
- Repressive and Paranoid but not Blunt or Militant
…and so on, through combinations of three at a time, culminating in a final:
- Repressive, Paranoid, Blunt, and Militant
There are so many combinations that it is really quite inconvenient thinking about them, even counter-productive. Instead, think about each attribute, and it’s opposite.
- Actions can be Repressive, ie designed to pick on one particular sub-population or practice that would normally be tolerated or even accepted – or encouraging, designed to make a particular practice or policy more attractive to the general population.
- Decisions can be Paranoid, targeting enemies real, theoretical, or imagined, or they can be optimistic, assuming that people will support the government.
- Phrasing and applicability can be blunt, or they can be subtle and discrete.
- Decisions can be militant, emphasizing enforcement, or passive, expecting cooperation.
For example, consider a change in tax policy. A repressive policy might be to triple the number of annual audits while targeting one particular group for no specific reason. An encouraging one might be a tax rebate equal to half the additional tax recovered to anyone who (in confidence) informs on a tax cheat. A Paranoid policy might add imprisonment and/or steep fines or even seizure of property to be paid on suspicion of tax evasion and pending the investigation, while an optimistic one would assume that everyone is trying to do the right thing to the best of their understanding and hence levy no punishments whatsoever short of a bill for the additional tax owed. Blunt tax laws might start by describing tax evasion as a crime and then referring to people throughout the tax code as “suspected criminals” or “criminal tax evaders”. A subtle and discrete tax law would use four paragraphs of dense legalese to define the word “tax” – “a debt incurred through the accumulation of profits and revenues consequent to employment, labor, ownership of business or property (whether encumbered or otherwise) and on and on and on… . Finally, we have militant (automatic mandatory audits and if you don’t lodge a return, nosy people come around asking “why not?”) or passive, which assume that people who don’t lodge a return don’t have to pay taxes at all.
In practice, I think that four such traits are too many. Pick one that’s overwhelming and one secondary choice at most. “Militant 10/- and Paranoid 5/-” is enough to characterize a government.
I could continue with more examples, but I think the point has been made…
Skill Applications
One final note, added at the very last minute: you can use the same technique to synopsise tricky skill sets. “Lensman 15/-” works perfectly well. So does “Green Lantern 10/-“. Or “Mad scientist 16/-“…
Discover more from Campaign Mastery
Subscribe to get the latest posts sent to your email.
June 25th, 2012 at 9:32 am
Wow, that’s a lot of work for making a simple solution! Grand idea though and a good read.
shortymonster recently posted..The Witcher 2, as reviewed by a table top gamer.
June 25th, 2012 at 10:34 am
Thanks, Shortymonster!
June 28th, 2012 at 2:31 pm
Hi Mike,
This is exactly what HeroQuest 2 does; any trait can be an Ability, and the mechanics determine an outcome regardless of the nature of the Ability. HQ2 even has a sliding scale of success, Minor Victory, Major Victory, Complete Victory, etc, to aid the GM with interpretation.
Happy Gaming
Phil
June 29th, 2012 at 4:27 am
Thanks for that, Phil. I actually have an original game system built along similar lines that I’ll present here one of these days.
Mike
June 29th, 2012 at 5:01 am
[…] the “dang, why didn’t I think of that” category this week is Mike Bourke’s article @ Campaign Mastery about a more general concept for NPCs. Creating numeric levels of “creepiness” (or some other abstract quality) and then […]
July 7th, 2012 at 7:22 pm
This is a great solution for those of us who like to include a mechanic to help with NPC portrayal. I know nerves can lead me to freeze up whilst GMing an NPC, but having a mechanic to inspire me will help.
Robert McLean recently posted..Zen in GMing: NPCs are Characters too
July 7th, 2012 at 10:46 pm
Use it in good health, Robert! I’ve also ‘frozen’ from time to time, especially early in my GMing life, and have always found that all I need to get moving into character was a starting point.
March 28th, 2014 at 11:55 am
This reminds me very much of Risus, in that characters are defined by their key characteristics.
March 28th, 2014 at 11:48 pm
Interesting – I’ve never heard of Risus, but it’s sometimes said that there are no really original ideas :)
October 22nd, 2014 at 7:43 pm
[…] INT and a little bit WIS and a lot neither of these. This is where a technique that I proposed in Look beyond the box: a looser concept for NPCs shows its power. Instead of fuffing around trying to decide how much of an INT score represents […]
May 26th, 2015 at 12:42 am
[…] Look beyond the box: a looser concept for NPCs […]
December 5th, 2015 at 5:07 am
[…] the “dang, why didn’t I think of that” category this week is Mike Bourke’s article @ Campaign Mastery about a more general concept for NPCs. Creating numeric levels of “creepiness” (or some other abstract quality) and then […]
November 4th, 2016 at 8:17 pm
[…] Look beyond the box: a looser concept for NPCs, […]
November 11th, 2016 at 1:09 am
[…] Look beyond the box: a looser concept for NPCs, […]