High Roller
The phrase is as old as the hobby. You’ll even hear it outside of pen and paper: “Legion is dropping on the 30th. Who’s rolling up a demon hunter?” Even though there’s no dice rolling or randomization involved in WoW, the meme is still there, locked into the DNA of gaming. You don’t just create a new character. You roll them.
The idea is simple. In the D20 System, every character has a number of ability scores. These represent things like Strength, Intelligence, and Impressing Barmaids, but every character starts out rolling 3d6 to determine the raw numbers. The minimum of 3 is terrible, the maximum of 18 is amazing, and the average of 10.5 is average (despite being impossible to roll on three six-sided dice). Any dang way, chances are that most of my fellow nerds reading this already know all of the above. What’s less likely is that you all like this system.
As long as random stat generation has been a thing, there have been gamers who hated the everloving crap out of it. It’s not hard to understand why. For one thing, older versions of D&D featured ability score minimums to qualify for certain classes. The 1e paladin had requisite scores of Str 12, Int 9, Wis 13, Dex 6, Con 9, Cha 17. If my Goolge-fu is correct, that equates to a 0.10% chance of qualifying. Good luck.
Later editions and later games did away with these requirements, but even then you’re still at the mercy of the dice. Rolling all 8’s across the board makes for a miserable experience, especially if you’ve got some schmuck like Fighter at the other end of the table who somehow managed amazing rolls. (Seriously, Fighter. You’re a jerk.)
The alternatives are many. You’ve got 4d6, drop the lowest. You’ve got 4d6, drop the lowest, re-roll 1’s. You’ve got point buy, where you get assigned a budget to purchase your ability scores. I’ve even got a 5e DM who runs this madcap homebrew where you roll two sets of stats on a straight 3d6, but you choose which one you want for each score, and then you get two re-rolls of a single die (meaning you’ve got to record each individual die of each roll), or you can forgo either or both of the re-rolls to take a free feat. If that sounds complicated, it bloody well is. Here’s the thing though: I understand and sympathize with GMs who invent these schemes.
Rolling randomly allows for diverse and interesting characters. However, you also risk rolling terribly, which can be less than fun. If you go with a solution like point buy you wind up with samey characters. Every fighter has 18 Str and 8 Int. Every wizard has the opposite. You’ll never find a beefy magic user or a particularly charismatic meat tank, and that’s a shame. So GMs attempt to find some happy medium, keeping the thrill of rolling for stats but tempering it with some form of randomness mitigation. The real trick is finding that balance.
That’s why I turn now to you guys with the question of the day. How do you like to generate your characters? Do you roll for them? Do you use some form of point buy? Or does your system of choice offer up a crazy alternative version of character gen that is obviously superior to any of the above? Let’s hear your methods in the comments!
UPDATE: Like what you see? Consider supporting us on The Handbook of Heroes Patreon!
We’ve got a sketch feed full of Laurel’s original concept art. We’ve got early access to comics. There’s physical schwag, personalized art, and a monthly vote to see which class gets featured in the comic next. And perhaps my personal favorite, we’ve been hard at work bringing a monthly NSFW Handbook of Erotic Fantasy comic to the world! So come one come all. Hurry while supplies of hot elf chicks last!
Point of interest re: a common misconception in this piece:
As you say, the 1e Paladin was all-but impossible if you use 3d6 down-the-line, but that wasn’t the method of generating ability scores in 1e. In fact the DMG discourages that method:
“As AD&D is an ongoing game of fantasy adventuring, it is important to allow participants to generate a viable character of the race and profession which he or she desires. While it is possible to generate some fairly playable characters by rolling 3d6, there is often an extended period of attempts at finding a suitable one due to quirks of the dice. Furthermore, these rather marginal characters tend to have short life expectancy – which tends to discourage new players, as does having to make do with some character of a race and/or class which he or she really can’t or won’t identify with.
The DMG gives a few methods as options, first and foremost is 4d6 drop lowest, and place scores as desired.
Paladins are still somewhat rare, but not .10% rare. You usually get to play what you want.
I’m willing to bet that a lot of 1e players used the classic “straight 3D6” method since it was what they were used to. However, I will not quibble. The text has been cited, and I admit my fault. You may push up your glasses with pride.
Sure, because, as we all know, players usually turn up their noses at mechanical advantages offered by new editions. 😉
Au contraire! As we all know, grognards immediately latch onto new ways of doing things. There are STILL dudes over in the Reddit thread talking about how “3d6 straight” is the one true way to game:
https://www.reddit.com/r/DnD/comments/4z0z48/who_still_rolls_for_stats/
I don’t know that there were many grognards qua-grognards in 1980 or whatever. These dudes were early adopters at the cutting edge of gaming! They were pumped for new content, not hearkening back to the good old days.
And in any event, 99% of people were seeing the game for the first time. But yeah, I guess there might have been dozens (dozens! http://imgur.com/JqYTmjn) of old dudes griping about how in their day it was done with straight 3d6, uphill both ways in the snow.
What did the PHB say?
To answer your question, though, I like Dungeon Crawl Classics RPG’s method of character generation. Generate 4 “0-level” characters by doing 3d6 down the line, then everyone sends their mooks into a dungeon. The survivors get made into level one characters. Super fun method of “generating” PCs.
You don’t always have great ability scores, but in DCC RPG ability scores’ in-game impact is closer to how it was in 1e, 2e (e.g., Str 17-18 gives you a +1 to hit bonus. No difference below that). It was only in 3rd edition that ability scores became so baked in to virtually every aspect of gameplay that having high scores was absolutely mission critical and having scores that weren’t optimal basically ruined the game.
I recently tried to teach myself 0e D&D by reading the rules. I’m still not sure what ability scores are supposed to do in that edition.
But I have heard of meat grinder character gen, and it sounds like a total blast. Blood and commoner parts everywhere.
4d6, reroll 1’s, drop the lowest, and meet a minimum stat bonus of (insert dm determination based on how much i plan to screw up the party)
What happens if they don’t meet the minimum stat bonus? Re-roll?
In our campaigns, we do the same thing. If the character does not meet the minimum stat bonuses. he is “milk-carted” ie The GM states he walks outside and is run over and killed by a milk cart. Roll a new character.
lol. That’s an oddly specific form of capital punishment. Did you guys just pick “milk cart” out of the air, or did some milk cart related shenanigans happen in a game at some point?
We *used* to do a lot of rolling stats… Until one GM decided that we should reroll until our total was above average (i.e. 64+ points of stats). After we had sat for half an hour waiting for me to roll a set of 64+ points we switched to point buy. :p
Now see, that seems like too much randomness mitigation. A nice idea in theory, but there’s got to be an easier way to set a “high floor.”
2d6+6. Not that I’ve ever tried this method; it literally just occurred to me while reading some of the convoluted methods on this thread to avoid low scores.
I wish I could find the post that inspired this comic. Some dude figured out the ~50 most common rolling methods and converted them all into the point-buy equivalent. Some of the crazier ones averaged out to 40 point-buy or higher!
My preferred method is some kind of rolling, but the GM rolls a set and everyone can choose any of the rolled sets. That way you avoid the classic sadfeels of “Jim’s character with three 18s is clearly better than mine with 1 16”, but you also get some randomness and often there won’t be one statspread to rule them all so you’ll still have a party with different stat setups. And most importantly nobody is sad (unless everyone rolls terribly I guess, in which case at least you’re all sad together).
Sounds like a rock solid solution. Everyone gets to roll (hurrah for the gambler’s thrill), everyone gets the same choices (hurrah for fairness). I’m honestly tempted to try this one in my next campaign.
I’ve always preferred the Standard 4d6d1, but sadly the rest of my group prefers Point Buy. They claim it puts everyone on the same power level, but I think it just makes it impossible to play classes that need more than one or two ability scores. Basically, it just increases what the game already does, which is to say favor casters and screw over interesting martials.
Since this is such a universal issue, and since so many people have such strong opinions, I’m beginning to think that there is no “best solution.” Ramsus’s method up above sounds interesting. So is my 5e GM’s wacky homebrew. Both involve player choice, and that may be the most compelling part of any method.
Your GM insisting on point buy is OK in a vacuum, but he failed to make all of his players happy. Next campaign, I think I’ll take more pains to make sure stat generation tickles everybody’s fancy.
And now you’re guaranteed to have three players who insist the thing that would most please them is to determine stats by some method involving DDR, chess, and bird watching.
“Checkmate!”
“Release the pigeons!”
* Valkyrie dimension *
Shit… That’s my dump stat I guess.
I saw a proposed system in a forum once of setting a base of 5 to 8, depending on GM/player preference, each player is allowed to set two stats to 10, then roll 1D6 and add that to get your final stats.
Never player with it, but it seemed interesting.
As in, “My Str, Con, Int, and Wis are all 5-8, but my Cha and Dex are 10 + 1d6?” Not sure I’m reading you right, though it does sound intriguing.
Not quite, the 5-8 would be a universal starter, so if the GM picks 8 then everybody starts with all stats at 8. Players then pick two abilities to set at 10, and then roll 1D6 for each stat. So your final scores for a wizard might look something like this:
Str: 8 + (3) = 11
Int: 10 + (2) = 12
Wis: 8 + (4) = 12
Dex: 8 + (1) = 9
Cha: 10 + (5) = 15
Numbers in parentheses are the results of dice rolls
Ah, right on. I got you now. Feeling bad for that wizard though. He should have tried a little harder to have a dragon ancestor. Sorcery seems more like his jam.
Two questions then. When do you apply racial modifiers, and are you allowed to change class after you roll?
I feel sorry for the party who’s GM picks 5, having most of your stats cap at 11 would be pure suffering.
This seems cool, but I’d raise it to ten base for my players. Gives a chance at an 18, without a whole lot of chances of a nine… If you WANT a chance at a nine still, then you have them start base at 10, and for the two 12’s, you have to lower two other stats to 8. Still gives a chance at that AD&D pally. More I think about this, the more I like it better than my 4d6 drop lowest, reroll ones… Yeah…
2d4 (Re-roll 1’s), roll stats in order, trade stat sets around the table, then do a reduced point buy.
23pts, +1=1, +2=2, +3=4, +4=7, +5=10pts.
Causes confusion, yes.
But it’s interesting, and not too random.
So the max score is a 13 before racial modifiers? That’s low powered, but it is in line with NPC stats, and it’s definitely interesting. How did you arrive at 23 pts though?
23 points is the result of first making the point buy curve grow more steeply than default, then deciding that just under a +3 to each stat felt about right.
My main intention in making the scores come out low was so that I could give out ASI’s as a very unique form of loot.
Very cool. Glad to get a little more insight into your methodology. Thanks for taking the time to explain!
Ability arrays are the way to go, IMO. Randomness during character creation is the bad kind of randomness, IMO, in that it can screw your character over with absolutely nothing you can do about it, AND it can last for a long time (unless you “accidentally” get your character killed and reroll).
Ability arrays negate both the OP/UP problem with random chance, as well as the tendency towards minmaxing you get from point buy. Also, the fact that you’ve either get some extra scores to play with or have to have at least 1 dump stat creates the opportunity for some interesting roleplay, like explaining why your wizard has the second-highest Strength score in the party.
Muscle Wizard, as we all know, casts fist.
https://1d4chan.org/wiki/Muscle_Wizard
You have an enviable knowledge of nerd-culture.
Do not envy me. I would not wish The Training on my worst enemy:
https://www.youtube.com/watch?v=tOUksDJCijw&t=47m28s
There’s a secret cheat baked into Atari’s computer adaptation of Temple of Elemental Evil that allows you to first roll stats normally and then increase the rolled stats using point buy (rather than simply keeping the rolled stats or using the game’s standard point buy baseline of eight)
Atari is a home video game console, right? QED, you’re luckier when you roll at home. 😛
They used to be. Now they mostly sell computer games through Steam and GOG
IIRC, there’s also a bug in the game that allows you to use the counter keeping track of the number of times you’ve rolled as a roll itself and allocate that to an ability score. Roll 100 times, and you have a ridiculously broken character; possibly literally, because it’s possible to get a character’s Int so high that they can’t spend all of their skill points and get stuck in the character creation menu.
I feel like this information will one day save the protagonist of a “Ready Player One” sequel.
Point buy is my preferred method of generating characters. I will also accept a good ability score array. I will not, however, join a campaign where die rolling is used to generate ability scores.
In my first D&D 3.0 campaign way back when, I rolled up a sorcerer. The DM was using a custom die rolling system where you rolled 6d6 for one ability score and lost one die each time as you went down the line until you hit 4d6, then dropped down to 3 dice per score. Even with this, I managed to roll an unplayable character (which, considering I was rolling a Sorcerer, is really saying something). The DM let me try again; the same thing happened. After the third or fourth re-roll, he gave up and just let me roll 6d6, drop lowest 3 for all my stats.
That was how the story of the Sorcerer with nothing less than a 16 in his ability scores started. This is not a unique occurrence for me; I have made multiple DMs throw up their hands by rolling unplayable characters repeatedly.
My younger brother ends up rolling nothing less than a 16 on normal 4d6 drop lowest regularly. Hence why we always play with point buy nowadays.
You poor bastard. I gather that Thief is your spirit animal.
https://www.handbookofheroes.com/archives/comic/trapsmith
In every group, it always seems like there’s an unluck lodestone. I’m just glad it isn’t me.
One more person on a archive binge!
The method my group has always used was roll up 2 sets of 4d6 for each stat, keep the set you like better.
It was this method that resulted in one of the most interesting characters I’ve ever had the pleasure of playing with, affectionately nicknamed “Paliderp”
Paliderp’s player decided to go with a “race” called a hellborn. Basically this is someone who was evil in life who repented so hard once in hell that some deity of good took pity on them. She made this character a palidin, but part of becoming hellborn is take -2 to a stat. She picked intelligence, resulting in a score of 8.
Essentially she became a red oni in plate armor, without the booze, kinda dumb, and terrified of fire. The best part of it was, she played to the intelligence score in all roleplay interactions with the rest of the party.
Welcome to the comic, Hlatbern! Binge well. 🙂
Gotta love a good dump stat. If played well, it can define a character just as much or more than maxed ability scores.
I like the feel of 3d6 in order, but i know it just tanks any sort of character plans people have going into it, so i like 3d6 in order, get to replace one stat with an 18. You still can have some kooky stuff like a 17 int barbarian, and you will very likely have something that you arent very good at. I also recently offered two 16s being replaced instead because one of my players pointed out it wasnt fair to gishes and other MAD dudes, but it kinda also takes away a large part of the appeal of 3d6 in order so im kinda torn on it. My players seem to really enjoy the system
Seems like a solid way to balance the thrill of chance and the fun of character planning. What system are you using?
I’m quite fond of the tic-tac-toe method from here:
http://eldritchfields.blogspot.nl/2016/10/ability-score-generation.html
I’ve never seen that one. I’d heard of something similar with a 6 x 6 grid where you can choose any column or row as your array, but this one looks like a nice gritty “hard choices” method. Neat!
What I do is roll 2d6 and then add 6 to the roll for each ability score. Then I assign the numbers to the ability scores at will. This typically leads to characters with decent stats most of the time.
archive binger! honestly the next time I start a campaign I am going to get everyone together for a session zero and tell them to just choose their stats….I really don’t care if they have three 18’s at level one….if that’s what makes em happy then go for it. I can work around it as a DM
Welcome! Please enjoy your binge. Exits are located to the front and rear of the comic. Make sure to take breaks for sleep and a mid-binge meal.
For me, the concern is less for power level than for player disparity. I’m less concerned that things are fair than that players feel like they’re fair. That’s tough when one dude has a 15 point buy and another dude is rocking maxed stats.
I just want to say that I fucking adore what the Pathfinder Playtest is doing to fix this age-old problem. It’s basically just what 5e did with skills, except now your entire stat array is justified by your character’s history.
For those reading this that aren’t aware or need a refresher, Pathfinder 2 is going to do away with point buy and is relegating 4d6 drop lowest to mere optional rule with a heavy disclaimer about its problems. Instead, all characters start with a 10 in all stats. Those 10’s are then modified by ability boosts, which is a +2 to a stat unless it’s already 18 or higher, in which case it’s only a +1. You then get four opportunities to get ability boosts at level 1.
Your race gives you boosts to two specific stats, a free boost that you can apply to a third different stat, and then a penalty (-2, even if it’s 18 or above), while humans, half-elves, and half-orcs simply get two free boosts to two different stats. Your background then has you pick one of two specified stats to boost, then you get a free boost to a different stat of your choice. You then get four free boosts that all have to be applied to different stats, just for being level 1. And then finally you get a boost to your class’s key stat, with fighters and monks being able to pick between STR or DEX. No stat can be above 18 at chargen.
It seems like such a wonderful fix, one that I wish would have made its way to 5e. The abundance of free boosts along with the restriction that they have to be spent on different stats than what was already boosted by that same source means that everyone can and should start with a maxed-out 18 in their class’s primary stat, but the rest of their stats are going to naturally be fairly well-rounded. And because you can stack three free boosts from different sources on the same stat, that means your race and background aren’t pigeonholed by your class anymore (at least so long they’re not fighting a penalty). An elf barbarian is just as capable of getting an 18 in STR as a dwarf barbarian, and so there’s so much more room for making varied and interesting characters that aren’t simply picking whatever gives you a +2 to the stats you need. After playing so many RPG’s where race isn’t really a choice and where half-orc wizards simply don’t exist, it’s an exciting change.
And it’s simple and easy, your stats just fall into place while you’re making your character. You don’t have to conciously decide that you have to take the sailor background because there’s no other way you’ll be able to hit your target otherwise. No house rules necessary, no specialized calculator is required like with point buy. It just works.
Hopefully this gets accepted enough that it becomes a common homebrew for 5e, because dam near every table seems to insist on rolling for stats and then acts surprised when the people who rolled low and don’t ever seem to succeed stop showing up.
Goddamn I need to find a play test group. I just moved cross-country, start this frigging’ PhD next week, and am sorely game-deprived. SOMEBODY TELL ME WHAT ROOMS ARE NORTH, SOUTH, AND WEST OF ME!
18/4d6/4d6/4d6/4d6/8.
Coming at this binge reading in 2019, I submit to the comments section my group’s general preferred method of character generation:
DICE POOL!
In dice pool, the DM takes 18d6 and rolls them until satisfied with the result. These dice are then put into the dice pool. Players may now select any three dice for each stat.
PROS – PCs get to pick how their stats manifest, nobody is any more powerful than anybody else could have been, there can still be a rather heavy amount of randomness between characters depending on who is willing to sacrifice what to get it. Extremely easy to allow people to “roll” at home by sending a picture to all players.
CONS – Not truly random, OP/UP dice pool means everyone is OP/UP, not as thrilling.
Works good for my group and I suggest people give it a shot!
Welcome to the comic, Beefki!
In you experience, do DMs usually roll in the open? Do they skew high, skew low, or vary it up by campaign?
I like to whip up characters. Sometimes I just want to make a certain race or class. Sometimes I’m making a concept. When I make them, I use a 20 point buy because I find that easier. Especially since I make more characters than I actually play and I don’t have to prove dice rolls when I do use them. If I join a game that wants me to roll, I roll.
I guess I don’t mind rolling. But if I am, I definitely want to roll before I pick my character concept. No fun discovering that I’ve got a 9 Int when I had my heart set on playing an alchemist!
Hm, a 9 in Int? Guess I’ll be playing my Cavalier.
Start with twelve everywhere. You can remove two points from an ability (or one point each from two abilities) to give one additional point to another ability, with a maximum of 14. Then roll 7d6, drop the lowest. Add each die to an ability. Abilities are capped at 18, as normal, so 14+6 will not give you 20.
What does an average array look like? 15.5 across the board? High likelihood for multiple 18s? Seems pretty super-heroic.
The aim was to avoid the dump stat phenomenon, get something were you can choose one or two good stats but that doesn’t mean you’ll necessarily be lame elsewhere.
Another idea that was had but not actually tried for this was: choose either
– 18 in one stat, or
– 16 in two stats, or
– 14 in three stats
For all others, roll 1d6+7.
It is probably a bit overpowered, but then again, D&D characters are super-heroic after a few level-ups anyway.
Personally, I’m all about the point buy. I’ve had many years playing HERO system and that’s all points buy. I wouldn’t run a game any other way.
But rolling can be fun. Some years back a friend of mine was running the Age of Worms adventure path. He went for: roll 4d6 drop 1 for all stats BUT if our stats came to a lower point value than the equivalent of 30 points worth of stat buy we got to buy up to that minimum. Very generous I thought. I then rolled outrageously well (in front of the GM!) and got the equivalent of something like 44 points worth of stat buy. THEN I went and power-maimed it by playing a monk (in 3rd ed. monks rate somewhere about fighters for effectiveness.) Good times.
Well hey, if you’re ever going to play a 3.X monk, the 44 point buy game is the time to do it.
Archiver/New Reader here, I’ve recently started running Temple of Elemental Evil for a 2e group so we can experience the classic module, I’ve found I really like Method VI.
Every score starts as an 8
Roll 7d6 and record the individual results
You can add those rolls to your scores, you cannot exceed 18 and you cannot split dice, it often means players can have an 18 if they want it, and classes like paladins maintain their shtick of being rarer because you still nee good rolls, but you’ll almost never be truly screwed.
Welcome to the comic! I hope you enjoy your stay.
I like the idea of allowing paladins to feel rare. That’s their history after all, and seems appropriate if you’re going for that 2e feel.
What was the spread like in your Elemental Evil game? Seems to me you might have a high variance with that method.
The worst character I’ve ever rolled on 4d6 Drop 1, was Xzar the Quad(Dual)-Cursed Oracle. Generated at level 5 and survived until campaign end at level 14. Str 3, Dex 7, Con 11, int 5, Wis 9, Cha 13. I picked Merfolk, and skipped the Strongtail racial, so movespeed was 5. Oracle curses were Clouded Eyes and Haunted.
This was the character that completely turned me off of pretty much any rolling in the last decade. The other five characters in the party had an average of two 18s, one player had FOUR!
Ever since Xzar was born, I’ve made it a point to encourage point buy or arrays, although I’ve still played several games with rolls because some people actually like the risk.
Super late to the comments, but I’ll at least add what I prefer.
I just offer a choice: the player can pick if they want to point buy, or if they want to roll. PB of course lets someone comfortably have a character, but if they roll they have the old risk v reward of 4d6d1.
To me, this makes them acknowledge the choice and possibility, and lends itself towards they feel satisfied with their character coming from their own decisions on that end.
Some interesting math on that topic: https://www.reddit.com/r/mattcolville/comments/73jcs8/ability_score_point_buy_balanced_w_4d6_drop_low/
One idea I’ve thought of but never tried: In a game where stats are determined purely by random rolls, perhaps give bonuses to players who roll badly and penalties to those who roll well. Determine some set average; for each point below that average, provide quality armor, magic items, artifacts, or whatever; make them loved by merchants. For each point above, start them with less stuff, give them a punchable face that makes monsters prioritize them in combat.
It’s a neat idea. But at that point I feel like you’re just re-engineering points-based character building.
A stat rolling method I’ve toyed with but never used (because even if you use a digital dice roller, it can be a major pain to do) is roll a 6×6 grid of stats (up to you how to roll, but I’d say 3d6 is probably closer to the expected stat arrays of the standard 4d6h3 method when all’s said and done, but you could maybe do something like 6d6 drop highest 1 and lowest 2 or other options). Then you label each row and column by the stats in order, and choose one roll from each stat’s row or column for that stat, but you can’t choose something from a row or column already chosen (say, for example, you roll an 18 in the wisdom row, dex column, and you decide to use it for your Dex; you couldn’t the use any more rolls in the Dex column or Wisdom row)
It has the advantage that you’re rolling so many dice, it’s hard to have a completely unviable character, but because your choices will usually force you to make tradeoffs down the line, you can still wind up with some unique stat arrays. The big problem is that it’s a lot of work, and if you’re not careful you can accidentally wind up at the end with an incomplete stat array (if you take, say Str from the Wis row and Dex from the Wis column, you can’t take Wis from anywhere), or you might wind up that the last stat you have to place would be like a 5 or lower, and if you’re not okay with that, it can require some mental effort to untangle or avoid in the first place.