Rolling 2d4 - 4 for modifiers?


Here’s an adaptation of the Knave method: for each stat roll 2d4 and pick the smaller. According to Anydice (1a05d), it gets an average spread of 3, 2, 2, 1, 1, 1.


I like how simple that is @inmatarian. That would be a good way to go for groups wanting to tighten up the spread a little more.


i suppose you could also just roll 1d4 and - 1. it gives you a range of 0 to 3, which is the min and max amount of points i would normally put into a single stat. the average would be a +1 or 2 with a 0 and 3 peppered in there theoretically. i rolled this method a few times to test it out and ended up with an average of an about 8 point spread, which is usually how i start off my campaigns anyhow. this just gives the opportunity to be pleasantly surprised with a couple extra points (the highest i rolled was 11). you can always set up a limit with your group too. “if you roll under a 6 point spread, then you get a mulligan” or whatever. i just like thinking about these alternate stat methods since ICRPG deals with such a reduction in numbers. it is minimalist rpg at its finest, so i dont really have to worry about math or rules, just about making a good story and having fun with some buddies.


Ya, any of the methods mentioned in previous posts as well as yours would be viable. I just depends on the spread and the deviation that works best for your table.


Just wanted to throw this in from my thread about the joy of bad stats which relates to what @Alex said:

This is how that gulf plays out with 1000 stat blocks. My thinking (as i mentioned in the other thread) is that ICRPG is balanced for a +6 sum so i would expect that anything significantly far from that would have notable consequences on play.

just adding this so that folks reading this thread in the future have an example of how this can play out.


So, this is a fu little conversation. I skimmed much of it, but I thought I would throw out some ideas. To help blunt the stat disparity, you could pull an INto the Odd/Electricbastionland, and develop a modified loot chart that would give you better loot based on what your highest stat is - so someone with a +4 gets pretty mundane stuff, and someone with a -2 has something EPIC!

As an alternative, I though a 1d12 roll for stat point pool, then distributing each point randomly with a D6 could work, allowing players to purchase up to a +6 (adding positive numbers only) in stats, by buying them with negative points to other stats. Roll D8 for effort, distributing randomly, if so inclined. This has more swing, but a narrower range of results, and allows for buying up some higher stats at the cost of being bad at other things (which some of us think can be fun). Rock on, everybody!


Like some have already stated, there are six stats and you have six-sided dice. Roll 6d6 and let the number on the rolls determine your stats.

My question is, “how is 2d4-4 different from 1d6-2?”


1d6 - 2 has equal chance of each possible number. 2d4 - 4 the middle numbers are more likely than the extremes. The more dice the closer it gets to a normal distribution, and the narrower width of that distribution.


Got it. I would never use that method. Cool if other people like it, though.


This is my favorite method for staying true to the ICRPG design while getting a bit of that “character discovery” feeling.

That’s my favorite part of character creation. Letting the dice fall and watching the character take shape. Once you have a little framework, your imagination wants to find a way to tie everything together.

I’ve also played around with using the d6 pass/fail roll for stat generation. What I landed on was:

Roll 3d6. 1-2 is +1, 3-4 is 0, 5-6 is+1.

If you wanted to make 3s really really rare, say you only get -3 if you roll three 1s, and +3 only for three 6s.

As a GM I’ve also had people jump into the game with blank character sheets and we assign bonuses as we go. You build your character through actions rather than concept. If you want a character that’s smart, you have to use your INT stat. That’s not ICRPG or OSR, but I think everyone has fun with it.


Here’s one that really mimics old school feel of 3d6 down the line:

Roll 1d6. The result correlate’s to the relevant stat in the str, dex, con, int, wis, cha order. Next roll another 1d6. If it is 1-3 that’s a -1. 4-6 is a +1. Repeat six times. If the total mods add up to a negative amount, or 0 if you’re very kind, reroll them all. Then add racial bonuses if any.