top of page

Roll For It!

It's session zero with your best group of friends. The table is littered with snacks, dice, pens, papers, books, and maybe even a few miniatures or maps to get your imagination going. You're getting ready to start a brand new campaign, and it's time for the first big question that every D&D group must answer....

"How do we determine our ability scores?"

While it's true that ability scores aren't the end-all, be-all of a character, they do greatly impact a character's ability to mechanically interact with the game world. And the ways you can generate these abilities are as diverse as the people playing the game. So what's the best way to do it?

Disclaimer: the discussion today will focus specifically on D&D 5e ability scores, but the principles and methods could just as easily apply to other systems and versions.

The Goal

Any time you start thinking about "the best", it's important to spend some time figuring out what the relevant metrics are: what makes one way "better" than another; what is the overall goal of the exercise; and are these metrics the same for everyone?

The answer to that last question is probably a resounding "no", but there is a single answer as to why we generate ability scores: to have fun! It's a game and we want to have fun, and not only can the generation be fun on its own, but the scores will form the basis for characters that will provide hours of joy.

So how do we measure fun? Well, that's probably impossible and also very personal, but it's safe to take a stab at a few common metrics for what makes something fun or not based on sports, table top games, video games, and all other manner of hobbies.

Personalization. People want to create. They want to make something their own. That's where any Role Playing Game differs from a other genres - the ability to customize that character to what you want them to be. Even in RPG video games where the character's name, gender, and story are outside your control (like Jedi: Fallen Order), you still get to personalize the skills, abilities, and gear that character obtains.

Uniqueness. How many build guides are out there for your favorite game? For some, it's more important (and fun) to create something they feel is unique and special to them, even if it's not "the best" by someone else's standards. Allowing for someone to make choices "against type" or find alternative play styles is also part of the fun in RPGs.

Another way of thinking of this metric is also unique to the individual player. Even if they choose to play a pretty traditional Human Fighter, maybe they've never played a fighter before? Or a human? That combination might be common to others, but it will be completely new and unique to them.

Power. In games where you are fighting evil, it's fun to feel powerful. It's fun to be able to beat up the bad guys and save the day. Being able to be effective in the game - to have mechanical potency - is fun for folks, and ability scores are a way of obtaining that power. There's nothing quite like starting off at level 1 with a 20 in your primary ability score.

Simplicity. Not everyone writes a 4 page backstory. You might be the only one in your group that spent all night making the perfect mood playlist on Spotify for your new character. Feat trees? Who has time for those. Some people just want to play a game with friends and not get too caught up in all the minutia. For them it's fun to keep things simple. These are the folks that might be just as happy playing a pre-generated character as they would be something they designed themselves.

Fairness. D&D isn't a competition, but that doesn't mean someone won't feel like they lost if things aren't fair. This can be very difficult in a game like D&D and other TTRPGs, where so much of your character designs are intentionally asymmetrical. A fighter can't cast spells, and a wizard can't take a hit - but both of those players are having fun. For some, it really sucks to feel like you are starting "behind" at level 1 because your abilities aren't as high/good/whatever as others. This metric measures how a method interacts not just with one character, but with a party of characters.

The Methods

Okay, we have our goal: to have fun!

And we've got our metrics on how to measure fun: personalization, uniqueness, power, simplicity, and fairness.

Now each person is going to weigh those metrics differently. One person might prefer to be powerful, even if their character isn't particularly unique. All playstyles are valid! And so as we examine the different methods in pursuit of a "best", we'll give each one of them a grade on how they do in each of the four metrics: F, D, C, B, A, and S, with F being the worst and S being the best.

Standard Array

Personalization: C

Uniqueness: F

Power: B

Simplicity: A

Fairness: A

It's the Player's Handbook default: the Standard Array! Take the numbers 15, 14, 13, 12, 10, and 8, and assign them each to an ability. Done!

This is super simple since there's pretty minimal thinking, and it's supremely fair since everyone has the same numbers. The only reason it doesn't get an A in fairness is the same reason it's a B for personalization and power: some builds will struggle to function with this array. At most you'll only have two abilities with a +3 modifier, which means any MAD (multi-ability score dependent) classes/builds will struggle.

This is good, but we can do better.

Point Buy

Personalization: S

Uniqueness: C

Power: A

Simplicity: D

Fairness: B

The classic answer to the boring-ness of Standard Array is Point Buy! For the exact specifics, go check out the Variant: Customizing Ability Scores rules in the Player's Handbook, but the TL;DR version is that you are given a budget of points with which to buy your ability scores. At first the scores cost just a little more the bigger they are (ex: a 10 is only one point more expensive than a 9), but then as you get higher they are priced at a premium (ex: 15 and 14 both cost two more than their predecessor).

The amount of hours we've spent twiddling with numbers and point costs is frankly shameful, but that speaks to how personalizable this method is. You can get exactly the character you want, within the bounds of your budget of course. And since the budgets are the same for each player it ends up being pretty fair. The problem is that while this theoretically opens up more options than Standard Array, it becomes very tempting to fall into the same builds over and over without any randomness coaxing you to explore alternative options.

Still, a great option for those that want to prioritize personalization and power.

4d6 Drop Lowest

Personalization: C

Uniqueness: B

Power: A

Simplicity: C

Fairness: C

The method here is still pretty simple, but is adding the notion of rolling for your scores. You roll 4d6 (four six-sided dice) six times - one time for each ability score. For each of those rolls, remove the lowest die result. So if you roll 3, 2, 4, 6 for the first one, you would drop the 2 and end up with a total of 13. You can then choose which ability score you assign to each of the resulting numbers. It's mostly the same as Standard Array, but with the wrinkle of rolling for things, dropping a number, and adding, so it drops down to a C for simplicity.

There are about a thousand variations of "Xd6 phrase", but 4d6 Drop Lowest is easily the most common and well known. As such, both personalization and uniqueness are B's here. You will likely get a lot more variance than standard array, but dice still like to follow averages, and most people that do this will still create re-roll rules for lower and upper bounds (example: scores need to add up to a minimum of 74). Still, rolling 4 and dropping the lowest will give you pretty solid numbers, hence the A for power.

It takes a hit in fairness because it's extremely unlikely that two players will roll the exact same scores, let alone the same totals. This can leave your wizard with god-like stats while your fighter is struggling with 13's across the board. Any amount of fairness requires the group to start mucking with the method, and thus changing it entirely.

This is a good one for thrill seekers that like a bit of gambling with their statistical averages, but there's plenty of room for improvement.

3d6 In Order

Personalization: F

Uniqueness: S

Power: A or F

Simplicity: S

Fairness: F

Every ranking needs a wooden spoon - a clear loser to help find that lower bound on the metrics. Well look no further than 3d6 In Order, which is where you roll 3d6, sum them up, and put what you roll into an ability score in the order you rolled it. For example, if your first roll is a 13, that goes into Strength. If your second roll is a 7, that goes into Dexterity, and so on until you have all six scores.

It's pretty easy to see why personalization and fairness are F's here. Sure its simple - you literally make no decisions the entire time - but you could end up with a demi-god, something akin to a slug, or something great that doesn't fit the character that you were hoping to play.

Aside from folks looking for extremely experimental one-shots or for a good laugh, this is nigh unplayable.

Multiple Choice Arrays

Personalization: A

Uniqueness: D

Power: A

Simplicity: A

Fairness: S

This is the same as the Standard Array, but with a multiple choice twist: instead of only one array that you then assign to your ability scores, you get to pick from a selection of arrays. Here's a sample selection that we've used for our home games and seems to work pretty well:

A - 18, 16, 13, 13, 9, 7

B - 16, 16, 14, 12, 10, 8

C - 14, 13, 13, 12, 12, 12

This gives you more choices about what kind of character you are playing than Standard Array (better personalization), and also makes it less likely you are exactly the same as someone else (better uniqueness). It's still pretty simple - no rolling or math - and every array adds up to the same total, making it completely fair.

In fact, the reason fairness is an S is because it not only applies to raw number values, but also to build enablement. Someone that just wants two really good stats can go for array A, giving them a pretty powerful start with some fun negative stats to exemplify character flaws. Meanwhile someone that needs more balance in their build can go for array B, giving them some things to excel at while still enabling builds that need 3 high scores (paladins, some fighters, monks, etc.). Then there's array C for someone that doesn't like to have any negatives and wants to be able to really see their power grow as they gain levels.

This solution was specifically designed for games with large player pools that want a bit more uniqueness than Standard Array without having to math themselves into a coma finding a perfect rolling method. Definitely a keeper.

Roll and Subtract

Personalization: S

Uniqueness: A

Power: A

Simplicity: D

Fairness: A

Rolling is fun, so a good ability score generation should probably involve rolling, right? But randomness is inherently unfair. Well, not with MATH!

This Reddit post goes into greater detail, but the quick explanation is that you roll 2d6+6 three times, then subtract those three numbers from another set of three, static numbers (26, 24, 22). The rolled numbers and the three differences (results of the subtraction) make up your six ability scores! It's totally fair because the sum of all those scores will be exactly the same as everyone (the sum of the three static numbers). You might roll poorly or too average and whiff on a particular style of build, but most things should be covered. And you're going to get solid power since your adding 6 to 2d6, keeping your average roll at a 13 instead of the average 10 for 3d6, or 12.24 for 4d6 drop lowest.

Allowing for multiple decision points, such as how to subtract the numbers and then how to assign them, gives this method a ton of personalization. The only place this method struggles is in simplicity. With this much math, and three numbers that you use and then throw away, it can feel very confusing for folks that don't see the magic behind the math.

If your group is up for a bit more number crunching, this is a great way to produce fair, unique, customized characters.

D&D Bingo

Personalization: B

Uniqueness: A

Power: B

Simplicity: A

Fairness: B

A picture speaks a thousand words, so here ya go!

If you've never done it before, we highly recommend giving this a try. It's a lot of fun, and it makes great use of the d12 - a criminally underrepresented die! #barbarianlife

There's not a lot of personalization here since you aren't making a lot of active choices. But you do get more arrays to choose from than Standard Array or Multiple Choice Array, so as long as the dice aren't too mean you should have some good options. Considering the sheer number of dice you are rolling and the random +1s and +2s scattered throughout the board, you're likely to end up with solid numbers. And it's incredibly simple to use, even if you might go cross-eyed with the numerical word search at the end.

This is great for folks looking for a fun, new way to generate ability scores that still lands you where you want to be mechanically.

Endless Variants

This was just a handful of all the ways that exist. While it is not meant to be an exhaustive list of all the methods, it should be fairly representative of the myriad options out there for players.

Maybe you like to strike a balance between standard arrays and rolling by rolling a die and subtracting it from one score while adding it to another? Or maybe you delight in the true thrill of rolling 6d20?

Another personal favorite is to use whatever rolling method you want, but to have each player roll one of the numbers. The group then uses the array that everyone made. This combines the uniqueness and power of dice-based methods with the fairness of everyone having the same scores.

Is there a method you prefer that we didn't cover? Leave a comment with your own rating of it, or send us a message on social media!

Recent Posts

See All


bottom of page