D&D/PF has presented a lot of different ways to generate your ability scores–some simple, some convoluted. Roll 3d6 each stat, in order. Roll 4d6 and drop the lowest, sort in order. 4d6 drop the lowest, sort as you please. Roll 6 times for each ability, take the highest result. 3d6 each score, in order, for 12 characters, pick the set of 6 you prefer. Roll 2d6 and add 6 for each ability score. Dice pool. Point buy. Standard array. And so on. So many options!
This is some thoughts on ability scores, generating them, and what those ability scores mean in terms of what people can do.
(Update September 23, 2014: If you like this post and where these ideas are going, please check out the kickstarter for my Five Moons RPG, which uses these ideas. Thanks!)
Generating Ability Scores
There are two problems with any system where you roll your ability scores, because of the randomness:
1) Your character might end up worse than another similar PC simply because of a quick of the dice. In other words, “Kreggo and Magu are both 1st-level fighters who wield greatswords, but Kreggo is always going to be better than Magu because Kreggo’s player rolled a lot better.” That’s realistic, but kinda lame for the Magu’s player who rolled worse than Kreggo’s player.
2) You can end up with a character who has ridiculously bad ability scores, which means the game either makes you deal with having a weak character or allows you a “mulligan” in that you can reroll until you get better than a certain minimum (such as “until the sum of your ability score modifiers is at least +1”). Having the possibility of a low-stats character is realistic, but it’s also kinda poopy that you could end up with a weak character, and also poopy that the mulligan option means the game acknowledges its character-generation method can give you a crappy character but won’t replace that method with something better.*
(Character building really only has two points of randomization: ability scores and hit points**. You don’t randomly roll to see how many skill ranks you get at each level, or how many feat, or how many spells you learn, but ability scores and hit points are fair game for doing really well or really poorly… in a way that affects your character for the rest of their adventuring career.)
So for other ability score options there’s point buy, where you get a certain number of points and can “buy” higher ability scores with those points, and usually not at a linear cost (frex, in PF an 11 costs you 1 point, 12 costs you 2 points, and 13 costs you 3 points, but 14 costs you 5 points, 15 costs you 7 points, and 16 costs you 10 points). This is method means you can usually buy a couple of good stats, a couple of decent stats, and a couple of average stats (or perhaps a little better if you drop one of your weak stats to get more points back… although the system is aware this is a problem and doesn’t let you drop it too far for even more points).
And there’s the standard array (15, 14, 13, 12, 10, 8), which is like using a 15-point point-buy system. I personally don’t like the standard array because I don’t necessarily want to play a character who is worse-than-average at something. It’s also weird because the rules for adding class levels to monsters say to modify the base (average) monster’s ability scores by +4, +4, +2, +2, +0, –2, which is in effect the same mods as the standard array, which means a 10th-level fire giant fighter must have one lower ability score than a typical classless fire giant (probably Dex or Cha… so the expert fighter fire giant is either more clumsy or less personable than your typical grunt fire giant). I’d much rather have a system where a monster with class levels had +2 to two or three ability scores, and no penalties (compared to a standard monster of its type).
You roll your ability scores, and use those scores to derive your ability modifiers (the actual plus, zero, or minus you add to a roll). For most things in the game, after you determine what your ability mods are, those original ability scores don’t matter any more.
Those ability scores might make a difference between having a Str 13 and qualifying for Power Attack and only having Str 12 and not qualifying. They might make a difference when taking ability damage (or drain) and determining whether or not your ability score is 0. They don’t matter for stat penalties in PF because the rule is based on even numbers of an ability score damage/drain/penalty (instead of what number it brings your score to).
So if ability scores have so little effect on your character once you’ve derived your ability modifiers, why do we care so much about ability scores instead of focusing on the ability modifiers? Why not just allow people to generate their ability modifiers, and that’s what you add to your rolls? Just skip having the 3–18 ability scores entirely, and use the modifiers from those scores (–4 for a 3 score, +4 for an 18 score, and all the fun between and beyond).
The Meaning of Ability Scores
It’s subtly hidden in the game rules, but did you know that increasing an ability score by +5 means the creature is twice as good at that ability score as an unmodified character?
In other words, a character with 15 Strength is twice as strong as a character with 10 Strength. Check the Carrying Capacity table for proof… a Str 10 has a light load up to 33 lbs., medium for 34–66 lbs, and heavy at 67–100… and a Str 15 character has double those values (light up to 66, medium 67–133, heavy 134–200, if you account for rounding differences). This repeats for every +5 ability score points (Str 20 is twice as strong as Str 15, Str 25 is twice as strong as Str 30, and so on). Let’s think about that in real-world terms.
Let’s look at some bench press stats from ExRx.net for adult males at various body weights. According to this site, an untrained guy about my weight (181 pounds) should be able to bench press 130 pounds (FYI, no, I can’t). The site also says an “advanced” man of that body weight should be able to bench press 275 pounds, which is a little over double what the untrained guy can do. So we could classify the average person as Str 10 (which makes sense), and an “advanced” man at Str 15 (twice as strong as the average).
According to the ExRex.net chart, an “elite” weight trainer should be able to bench press about 345 pounds, which is only 1.25x the “advanced” value, so we can classify that person at about Str 17 (the max weight capacity of a Str 17 character is 1.25x that of a Str 15 character).
Keep in mind that the bench press world record was 363 pounds from 1898 (when the International Powerlifting Federation was founded) to until about 1950. Bench pressing 363 pounds is 1.3x the advanced man’s 275 pounds, and 1.3x times the (Str 15) advanced man’s heavy load of 200 pounds is 260 pounds, which is the max heavy load of Str 17.
(That also doesn’t take into account that Doug Hepburn, the 1950 record-breaker, weighed 300 pounds and should be able to press proportionately more than a 180 pound man, so the above stats are actually being very generous to an “average” person’s strength estimates.)
To be fair, we’re just looking at one aspect of human strength, but my point is this: for a very long time, the strongest human was less 3 times as strong as the average human. In D&D terms, the strongest human up until 1950 had Strength 17. Using point-buy for your ability scores, you could buy a 15 Strength, and then use your “+2 to one stat” for being human, half-elf, or half-orc, and get a 1st-level character who is as strong as the strongest human for half of the 20th century. And that’s without boosts from magic or +1-every-four-levels.
Isn’t that… weird?
If you instead did that to your character’s Intelligence score, you’d get a 1st-level character who is as smart as the smartest human for half of the 20th century. “Hi, I’m 1st-level, I’m as smart as Albert Einstien.”
Think about that the next time you build a new 1st-level character and decide that a 15 in their primary stat “isn’t good enough”… you’re saying that a 1st-level character who’s at the peak of human ability isn’t good enough… because the +2 bonus from that stat isn’t as good as the +3 you’d get for a 16 or 17.
What to Do?
I don’t want to go back to 1E/2E AD&D’s system where the difference between an 8, 10, and 14 was essentially “nothing” for most characters; one really great thing about 3E as a change from 2E is it made most of the mid-range ability scores actually worth something instead of being a dead zone before you got to the worthwhile ability scores. So here’s my plan for Five Moons RPG:
1) Creatures don’t have ability scores, they just have ability modifiers (in effect, their modifiers are the scores). An average stat is +0, not 10–11.
2) Use a point-buy system so players don’t have to worry about rolling lower stats compared to their friends, which also means the game doesn’t have to include a “mulligan” rule or expect GMs to house-rule it.
3) This point-buy system is simple, and doesn’t allow you to build a 1st-level character to superhuman levels. A 1st-level character gets a +1 in two stats and +0 in all other stats. This system creates characters who are above average, but not herculean (no more “strongest/smartest/swiftest character in the world at level 1”). Optionally, a character can take a –1 in one stat to get a +1 in a third stat (or use that +1 to increase one of their existing +1 stat to a +2); some people like having a stat penalty for roleplaying (and it’s not a huge advantage given other things in the system, which doesn’t promote stacking bonuses really high).
4) Having such a low starting range for bonuses leaves more room to “grow” as your character levels up, and it means that “growing” doesn’t require proportionately bigger numbers to feel significant. Frex, if your Strength bonus is +3 (from Str 17) at 1st level, increasing it to +4 at 4th level (by bumping Str to 18) doesn’t feel like a big deal because it’s only increasing by +25%; but if your Strength bonus is only +1, increasing it to +2 at 4th level is a +100% boost, which feels a lot more significant.
5) This probably means the every-four-levels ability score bump probably has to change its frequency, as it becomes twice as powerful in this system (although in Five Moons the level spread is expanded, so you’d only get them at level 8, 16, and 24).
(It seems weird to not have characters with a lot of stat bonuses, or a big bonus in one stat, but that’s how it was in Basic/1E/2E… and because everyone has a couple of +1s, everyone gets to be better than average at some things… instead of the possibility of having a Basic/1E/2E character with 14s in every score that give them a sad +0 bonus in every score.)
* I really wanted to use point-buy or standard array for ability score generation in the Beginner Box, but I was overruled. Which meant we had to write a mulligan rule specifically for the Beginner Box so new players wouldn’t get stuck with a loser character and have a bad experience.
** And as with ability scores, most GMs have a mulligan house rule for players who roll too low on hit points (usually “if you roll a 1,” or “1 or 2,” or even “less than the average result)… which is another sign that the official rule is a problem.
Reminder: If you like reading this blog, I strongly suggest you subscribe to this WordPress blog (by clicking the “Follow” button at the top or bottom of the window), especially if you usually hear about these blogs from my Facebook fan page. Facebook is changing its algorithm for whether or not you get all updates from a page you “like,” which means that you might never be notified of new posts to my Facebook fan page. Subscribe here, follow me on Twitter, subscribe to my YouTube channel, whatever you’re interested in, but you can’t rely on Facebook actually pushing these notifications to you. This goes for all pages you “like” on Facebook, not just mine.