Target Numbers VS Roll Under
When I started designing my rules- this was a big hurdle for me. First, I had to decide which way I wanted my mechanics to go: Target Number VS Roll Under. I needed to figure this out before I moved onto Ability Scores and Modifiers.
I pondered (what I felt were) the strengths and weaknesses of both, gauged my players thoughts and reactions to the different games and mechanics we played, and considered my desire to create something that is immediately easy to grok, but is engaging and has longevity (hopefully).
When running Barbarians of the Ruined Earth (which uses Black Hack as a base) and Death is the New Pink (which uses Into the Odd) games for my players, I got to see how easily they understood the rules.
ME: “See your Strength score is a 8?”
Player: “Yeah.”
ME: Ok- so you need to roll a 1d20 and get a 7 or below.
Player: “Man- my Strength sucks.”
ME: “Yeah it does.”
Player rolls 1d20 and attempts to get a 7 or less.
And that’s it. Super fucking simple.
While the game is simple- I did notice a few issues that seem to be a common issue/complaint with very rules-lite mechanics.
First– Roll Under goes against how we are programmed. 1 is less and is shit and 20 is more and is exciting. Even when I run a game with people who have never played a RPG ever get excited when they roll a 20. “Oh! A 20! That’s the highest on this dice, right? Does something cool happen.”
Second– Without target numbers and modifiers, I feel the system loses an amount of robustness, be it through skill points, modifiers for attributes, or modifiers gained though magical items, spells, drugs, etc. It usually boils down to gaining Advantage or Disadvantage for the situation (or canceling one another out) and rolling the dice. This is a super simple approach, easy to grok, and do- but all situations end up feeling the same mechanically.
On the other side of the coin I was worried about the escalation shit storm that usually accompanies target numbers and modifiers. This is why I find any level past 5th in 3.x, Pathfinder, and 5e so fucking annoying. All the math… Negative modifiers, positive modifiers, situational modifiers, Feat modifiers, spell modifiers, magic item modifiers. This escalation creep was created to keep the game challenging. You look at the target numbers of high level play (like 38 or some shit) and I’m all like, “Why? What purpose does this serve to anybody? What fun does such high and ridiculous amounts of modifiers really add to the game?”
To me- it doesn’t. It’s annoying, makes character creation a chore (which as I’ve said I don’t want), and bogs down the game more than it enhances it. The only thing that high modifier does is tickle the child part of our brain that “more is better.”
After much pondering about what excites me as a GM and what my players seem to enjoy and react to I decided to go with a TN/Roll equal to or over mechanic. Here’s notes from a previous post:
In the end, I decided to move towards a TN/Roll equal to or over mechanic. 1- I enjoy the fuck out of rolling a 20 and the thrill/dread or rolling a 1.
2- I feel (and this is totally my opinion here) that a game with small range of modifiers has a wider berth for long term play, and as my goal was to attempt to create something I’d want to run for one shots, con play, and long term- this sat right with me.
3- Those modifiers help to facilitate an easy mechanic for skills that help flush out characters to feel interesting/varied.
So What are the Target Numbers?
With my desire to avoid escalation I knew I wanted to keep the modifiers and target numbers down to a reasonable level.
The target numbers for my rules are: Easy: 5, Moderate: 10; Hard: 15; Extreme: 19; Impossible: 22. Mechanically, this means roll 1d20, add appropriate modifiers (ranging from 0 to 4 at character creation, up to +6 through leveling up, and up to +8 through magic items/spells, drugs, etc. and finally a +2 proficiency skill (more on this later).
These rolls are for tests like resisting poisons, charms, traps, explosions, etc. Combat is slightly different. That is roll 1d20 + STR/DEX modifier and +2 for Weapon Specialization (if applicable) and compare to the armor’s threshold (more on this when I talk about combat).
There will be Traits (think very simple feats), potions, situations, etc. that will let a character roll with Advantage or Disadvantage (more on this later too when I talk about modifiers).
On Rolling In General
I tend not to make the player roll the dice unless there is a reason or interesting situation that will come from it. I find that the thrill of giving the character’s an option is more enjoyable. “Ok you can open the lock on the treasure chest. You can take your time and will automatically succeed and you’ll disarm the trap (if there is one), but I’m going to roll a d6 and if it comes up a 1-2 something’s gonna happen… I’m gonna roll on one of my charts you dread so much. Or you can attempt to unlock the chest by succeeding on a roll. If not, there may be consequences.”
The player’s evaluate their situations and make the call. It gives them agency over the current situation in a way that normally would boil down to a roll.
I believe this is what 3.x was trying to do with taking 10 (or taking 20), but it just felt clunkily bolted on and mechanically driven more than an organic choice. Chris McDowall (author of Into the Odd) was the first person I noticed preaching this kind of philosophy and it’s definitely one I really enjoy, so I gotta give that man props there!
Alright- enough for today.
DIY RPG Productions 
My concern about target numbers is that they’re not always predictable to the player and up to the whim of the GM, which kind of prevents informed decisions.
I love the way Torchbearer handles this, by having a list of “factors” for each skill in a list that increase the TN. For instance if I said “climbing factors: lots of handholds, loose rocks, few handholds, slippery, sheer surface” or something, you’d could up for the ultimate TN (maybe in your case, multiplying by 5).
A system like this makes difficulty mostly predictable to anyone who checks the rule book.
I get that. Personally I’m not a fan of list or defined terms- gets to gamey for me. I prefer a quick conversation based on the situation. “You want to climb that slippery wall? That’s really difficult, so that’s a TN of 19. Oh, you have climbing spikes? Then let’s make that a TN of 15.”
This might just boil down to differences in GM philosophy, but I tend not to set a target number below 10. If a given task is easy to do, why bother with a roll? Adverse conditions will bump up the difficulty anyway.
I usually don’t ask for easy rolls unless I am giving the chance to roll a critical fail.
However, I included the TN for those who do prefer to roll for easy TNs.
I think your logic here is sound. I was a pretty big advocate for roll under until I started trying to write my own d20-based mechanics and felt like the simplicity of the roll seemed to steal depth from the action being performed. And just like you mentioned, it just doesn’t ever feel quite intuitive making a natural 20 a poor roll. Where I always get stuck is on the attribute stat itself. We have a range for STR which could be 3-18 (or 1-20) and that stat points at a modifier. So then I would ask myself, “Well the 14 means nothing, it’s the +2 that really matters here. So why am I rolling 3d6 down the line instead of just a d6 or something that would just give me my modifier without referencing a table?” Roll-Under makes the natural attribute score really mean something and that’s why it always drew my attention, but at the end of the day, it just makes for a less interesting experience. I’m still working out ways to make that 3d6 attribute score really feel like it means something and isn’t just a number pointing at another number.
I’m enjoying the read and interested in seeing how all this works out for you. Thanks!