Dec 11, 2016

BCG Retrospective XXXIII: NPC Tiers.

I already wrote a lot about the three NPC tiers in BCG and BCZ both. I wrote about how to build them, how to combine them, how to make NPC combatants relevant in Intermissions, how to improvise the Attributes and Skills of combatants, etc. So what I'll do is focus on what worked about them, what didn't, and how they could be different.

So why are there three NPC tiers? The idea was to have one tier equal to PCs, one lower powered than PCs, and another higher powered than PCs. Nicely enough, one Grunt is half a PC while one Boss is two PCs, keeping things simple should you want to mix and match the tiers for NPC Squads. From a design standpoint, I was trying to keep the number of rules to a minimum here (as I usually do) and it seemed like three tiers of NPCs should be all a game could need. If a GM needs very weak trash mobs or very strong superbosses, the rules do support them, even if they don't have their own tier.

Grunts are PCs with less total XP, no Genre Points or Powers, and the Paired Attack Action. They didn't always have Paired Attack though, and they had major balance problems back then. Without Paired Attack, low PL Grunts were weaker than half a PC. You know how PL 0 PCs are terribly, terribly weak? Well, Grunts have stats lower than that and lack the Genre to compensate for their low rolls. They get a lot better at PL 3 and above, being very close with the XP totals of PCs and having their numerical superiority compensate for the lack of GP. With the Paired Attack Action, they got a lot better at low PLs, especially for the glass cannons. Paired Attack can only be used when Grunts are bunched up, this keeps them more vulnerable to area attacks and makes the game flow faster by reducing the number of turns that the NPCs take. Overall, I would say that the Paired Attack Action is my favorite thing about them as a designer, it does a lot to make them play better.

Grunts naturally lean more towards elite mooks than trash mobs, but they can represent hordes of weak enemies with the Squadron Feature and a glass cannon build, preferably with a Threshold of 0. They're still powerful though, they can down PCs easily with Support debuffs, Paired Attack, and autodamage from Bombardment. Generally speaking, PCs want to destroy the Grunt forces as fast as possible and leave the Rivals/Bosses for later, because the glass cannon nature of most Grunt builds makes them extremely dangerous. This makes me think that perhaps there is room for a tier of NPC below Grunts for dedicated swarm/mob/minion Enemies. They could have less attack power and just hand out buffs for the higher tier NPCs while debuffing PCs, then they would still contribute to combat but wouldn't be a larger threat than the Rivals and Bosses.

What about their Intermission performance? Grunts provide standardized Attributes to use as DNs when dealing with NPCs, and that's really all they need to do. If you use them in combat, the superfast and extremely lethal nature of Character-scale battles means that numbers give two Grunts a large advantage over a single PC. While this is good in that it gives Intermission combat PC builds a much needed time in the spotlight, it's perhaps a bit too extreme.

Rivals are the middle ground between Grunts and Bosses, but they take the most effort to build and play because PCs are more complex than non-Rival NPCs. Thus, the number of Rivals per Operation should be kept to a minimum, they just have too many Powers and secondary abilities to manage. Rivals have so much variety in Upgrades and Weapons that they're the most fun type of Enemy to fight. Throwing Transformers, Combiners, Guardians of Steel and Technicians to the PCs does a great job of keeping things fresh.

Rivals built around exploiting the weaknesses of PCs can be comparable to a Boss in difficulty, especially if given abilities like Oldtype or Unstable Reactor to cheat the Power Ratings in their favor even more.

Rivals during Intermissions are basically doubles of the PCs and set adequate DNs for challenges as generalists or stonewall PCs with great challenges as specialists. If you have to beat a specialist in, say, Diplomacy through a Contested Test then all the advantage in numbers in the world isn't going to save you. Likewise, a single combat specialist Rival can take out an entire PC Squad if none of them are prepared for the occasion.. Or they could roll a low Initiative Test and fall to a single turn's worth of 10's. It's unlikely, but... Well, Intermissions aren't as balanced as the Operations. They're still better balanced than Grunts here, though.

Bosses have very high Attributes, repetitive Powers/Weapons and extremely powerful Upgrades. They're not very versatile, but they're strong and set the rules of the battle, forcing the PCs to come up with a counterplan or suffer the consequences. Having two Weapons/Powers keeps them from doing the same thing turn after turn, but it is the Upgrades and Capstones that make Bosses truly shine. At PLs 1, 4 and 5 they are at their strongest, and at PLs 0, 2 and 3 they are slightly underpowered for their Power Rating.

There's a considerable amount of advice in both books to help you make your Bosses into Superbosses meant to match more than 2 PCs in power. The essentials for this are either Component Rivals orthe Oldtype Feature... Probably both, if we're being honest. You can make them go beyond PL 5 with a bonus 30 XP and a Capstone per PL, but that doesn't give them the increased survivability they need to weather 3+ attacks per Round when they're only supposed to endure 2 of them. Combine all three techniques and you end up with some very, very scary baddies. It does take a bit of system mastery to figure out how to build and use them, though, so I figure maybe the game could have had a proper Superboss NPC Tier that incorporates bonus Utility Actions and extra passive defenses to their core rules.

Intermission Bosses are pretty much always enemies to be fought on foot. Sure, some of their Traits have utility value that is useful for Bosses not meant to be Character-Scale combat powerhouses. But for the most part, most Intermission NPCs designated as Bosses have that title because they're going to fight the PCs... And hooo boy they're kind of super good at doing that. We'll talk about that when we discuss Boss Traits though. Other than that, they're basically Rivals but with crazier abilities, so there's not much else to say.

Balancing the NPC Tiers.
While the Power Ratings were always 2 Grunts = 1 PC and 1 Boss = 2 PCs, the numbers themselves changed a couple of times. I remember originally thinking that Power Rating should be equal to Level for Grunts, PR x2 for Rivals and PR x4 for Bosses. The numbers were very easy to remember and work with, but before the first playtest I realized that this made Level 0 enemies have no PR at all. I really liked how intuitive and elegant the mechanic was and tried to work around the Level 0 problem, but it just wasn't working. Thus, all the Power Ratings were bumped to start with 1, 2 or 4 points for Grunts, Rivals and Bosses, then add the Power Level multiplier from there. But those numbers still had balance issues, in particular when it came to Grunt swarms, and so the we ended up with the current PR table that has NPCs start at a Power Rating of 2, 4 and 8.

In the end, the PR mechanic wasn't as elegant or intuitive as I wanted it to be, but it works better than the old version. I suppose the moral of this story is to not underestimate the power of weak enemies in high numbers. That's probably the most important lesson I've learned in NPC design as a whole. Bosses need optimizing in order to keep up with multiple PCs, but optimized Grunts are brutal and almost unfair. Actions are the most important resource in a turn-based RPG, don't underestimate the power of extra Actions, no matter how weak the users might be.

The other big lesson here is that trying to work five NPC archetypes into three worked out quite decently, but only for GMs that get systems and know how to bend the rules to their will. For everyone else, it is difficult enough that having two additional tiers of NPCs (for Minigrunts and Superbosses) is something I should have at least tried out.

I care a lot about encounter and NPC design, it is one of the most fun parts of being a GM for me. This section may not be as long as some others in the book, but these are rules that I care for and think about how to improve a lot. Whatever I do next for an RPG, the NPCs are going to take a lot from the lessons learned with BCG.

Next: Enemy Features.

Gimmick Out.


  1. If I ran a BCG campaign, or were to give a suggestion for one, I would insert this clause: "MP amounts for NPCS are merely a guideline. Experienced GMs feel free to assign extra MP or deduct MP from your NPCS as you feel is suitable" ;)

    As you noted, optimization plays a MAJOR part in BCG... an optimized L1 Rival with 130 MP can be a more trouble than a sub-optimzed L2 Rival with 160 MP. ... not to mention when the GM gets to do stuff like design the map terrain etc to favor the enemies, or stuff like Oldtype, which THEORATICALLY is +0MP so the boss isn't any stronger than it should be, but in actual fact, you've probably boosted the boss more than an extra +10 MP worth of stats would have.

  2. The clause is implied, that optimization for NPCs needs to be comparable to the optimization of PCs. But yeah, making it explicit somewhere couldn't have hurt.

    1. Can take several sessions of playtesting to figure out what has *actually* been optimized, and what was meant to be optimized but turned out to be self-crippling. lolz.

      ... yeah, point buy systems. ... I feel pretty sorry for GMs who have 1 player who is a munchkin at heart (even not "cheating" his build ends up near top optimization), and another player who says "Macross 7 was my favourite, I am building Nikki Basara, EXACTLY!" ... "but, that won't work too well in battl..." "I will build him and his mecha, EXACTLY!!!" ... poor GM, figuring out how to balance the NPC mecha for that...

      ... maybe our group was an outlier, but it made me feel maybe there could be guide on "what to do when you have over/underestimated the difficulty of the operation" ... haha...

    2. To be fair, trial and error is part of learning any new game, tabletop or otherwise. Doesn't mean I wouldn't like it more if the process was faster or more invisible in BCG, just that having the process at all is inevitable.

      The rules lean on making things harder for the PCs by default more often than not. This is because it's easier to underplay NPCs tactically and have them make a mistake or two to balance things down, rather than balancing things up should they be accidentally too weak. I don't remember having written that down anywhere in either manual, so I probably didn't, when I really should have.

    3. nah... It's sort of understood that Indie games work better for experienced GM/DMs... There's quite a number of RPGs that I wouldn't really recommend to a group who have all never touched an RPG before, but which are really fun if everyone can get used to things quickly. ;)

      I like BCG a lot, as you might have guessed by now! ;)

  3. You have no idea how powerful a component rival attached to a boss could be. Combining the incredibly powerful boss weapons and high stats with an great number of normal passive upgrades and extra utility genre powers is crazy. And the free maneuver every round makes the already tanky boss almost immune to damage.

    1. Here's what I had to say about them when I wrote the expansion's GM advice section:

      All Combiners are extremely powerful and you could accidentally create an invincible foe when you slap a Rival with Universal Component on top of a Boss of Power Level 4 or above. If you use this strategy, do it with extreme care and try to stay away from the Boss Upgrades with the most raw power like It Keeps Coming Back and Caustic Touch.

      I look at it as another way to make a 'single' unit that can stand up to 3+ PCs. I know it is extremely powerful, but it's the one bulletproof strategy to make a challenging superboss. The stats are just too high and it has too much GP for it to be easy, so the challenge is in not making it too hard.

    2. I forgot to add quotation marks. Only the first paragraph is from the expansion, the second one is new commentary.

    3. It's just that in this article you make it sound like not that big of a deal if you want to give the boss both oldtype and a component booster.

    4. Rereading it again, "very scary baddies" is probably underselling the kind of enemies created with this, yes.