It's the third of my ongoing series of RNG rants (see rant one or rant two if you're interested), and I still can’t find a concise way to talk about expected bonus values. I'm starting to wonder if I ever will be able to do it in a blog. So in the meantime I’m going to rant about the system differences between single and multiple RNGs, which is really me just using short hand to talk about the deep effects of whether your system allows multiple growth tracks or not. Allowing multiple growth tracks means that you can have a potentially difficult DC for one group be a potentially easy DC for another group when the groups are the same level before counting situational adjustments, though this is generally more obvious at higher levels where their level based bonuses diverge more sharply. Using a single growth track means that a DC is potentially the same difficulty for everyone involved, though again this is before counting situational adjustments. And situational adjustments aren’t worth trying to keep track of right now, so for the rest of this blog let’s pretend that attribute bonuses (I will probably spend the next rant on attribute bonuses and how they can screw things up actually) and training bonuses and feat bonuses and whatever else bonuses don’t exist. We’re looking at base number growth rates only.
We even have really convenient examples of these setups. 3E and 3.5E are both multiple track setups. Different classes gain BAB, saves, and even max skill ranks for individual skills at different rates. An AC that is a moderate DC for a high level fighter is likely an extremely hard DC for a high level sorcerer. A high spellcraft DC is probably not a big deal for a high level sorcerer but likely impossible for a high level fighter (assuming he’d been investing in that for some reason). 4E goes the other way with this. Every class gains bab, saves, and skill ranks at exactly the same rate. If a DC is difficult for one character and easy for another, it is because of really minor static class differences or those situational things we’re not worried about.
Getting back to RNGs, in a system with a single growth track you can assign a DC to a task, and know that you have set a difficult or easy task for every character at a particular level. Which is kind of a big deal, because it means that you can decide you want a hard task and set a highish DC for the current level and actually get a hard task for the entire party that the entire party can also participate in. If you’re the designer, you can just write in DC guidelines that are simple and meaningful. And if you want to know if a character has a set of bonuses that push them off the RNG, you can just compare their bonuses to these DC guidelines. It is much harder to go off the RNG in this setup (assuming designers did their jobs) because base bonus values don’t diverge at all. This is actually a pretty cool thing if you want to make designing and running adventures easier (we’ll see the more complicated case in a minute), it just comes with a cost I’m not willing to pay. If everyone grows at the same rate you don’t get characters who get better at things relative to their party members as they level. They certainly get better at things as they level, but that’s not quite the same thing as getting better than your fellow party member. These are things that I want more than I want single RNG, but YMMV.
Anyway, in a multiple growth track system, when you assign a DC to a task you set the difficulty for 1 or more groups of characters at a particular level, and you set it to different difficulties. If I decide that I want a difficult AC for a medium bab class, that’s a moderate AC for a full bab class and a hard AC for a poor bab class. That doesn’t really start to set in until around level 10 though, but once it’s there it just gets more noticeable with more levels as the bonuses diverge. You can’t set a difficult AC for a high level ranger that a wizard could hit without rolling a 20. The 7+ point difference in their babs means that you’re already stretching any shared RNG pretty close to as far as it can go. You can sort of accommodate this by picking the middle track and restricting your DC recommendations to rolls relative to that track, and if you care about everyone being able to participate in a roll you absolutely should do that (*cough* saves *cough). That actually lets people with the good track get better at that thing as they level and people with the poor track get worse as they level, and as long as the divergence isn’t too great it’s manageable. But if you want hard challenges for the guys on the good track and easy challenges for the guys on the poor track, you have to let go of a shared RNG, because you can’t fit one in with any sort of significant class based progression bonus divergence. At that point, it’s better to look at things in terms of multiple RNGs, and make sure that each character is where they are supposed to be on their own RNGs rather than making sure that they stay on some shared RNG. If you don’t care that some characters go over the top of the RNG or fall off the bottom of it, then you also don’t care if some RNGs are lower or higher than others. You might care if someone goes off the top of the good bab RNG, but you probably don’t care if people fall off the bottom of it since those people just have the poor bab RNG and are sitting in a perfectly reasonable place on it They can’t complete in all of the checks that that good bab RNG people do, but as long as they don’t need to do that to keep playing the game it’s not a concern.
So there we go, that’s how we get single RNGs and multiple RNGs out of advancement setups and decisions to care if we go off the RNG with some guys or not. If you want a shared RNG, you can’t have significant bonus divergence at all. 4E doesn’t have any bonus divergence, and so it could go on for another hundredy bajillion levels without pushing people off of a shared RNG. The 6 point differences in 3.5E high level saves, on the other hand, is about as far as I would go since you already can’t really have difficult challenges for the guys with the good saves. If you want RNG overlap, like we have in 3.5 bab, you can have a little more divergence between higher and lower layers, but not really between adjacent layers. The 5 point bab differences between good, moderate, and poor are basically fine, even though it’s a 10 point difference between the bottom and the top as long as you don’t care about those RNGs overlapping.
TL;DR: If you’re playing 4E you have a single RNG to worry about, and you can set DCs pretty straightforwardly. If you set a hard DC it will be hard for everyone. It’s actually really good about that, attribute and miscellaneous bonuses not withstanding. If you’re playing 3.5E you have multiple RNGs to worry about, and some characters are better than others in substantial ways at higher levels. If you do something silly like make difficult DCs against the good progressions, your game will have lots of failures in it (which I generally think translates to problems) as players who aren’t on the good track can’t keep up. You should probably set DCs based on the middle bonus track instead, and just accept that some characters will overperform at those checks and some will underperform. They’re supposed to as a consequence of different bonus growth rates after all.