D&D General here's how to stop jealousy in between lucky players and unlucky ones

[Greenwheat]

• Let them reroll only if a character would be “4-F”, as in they’d fail a draft physical. Any stat below X allows a reroll, but the first set with no terrible stats is it. And of course, they can choose to play a “4-F” set if desired.

What I'm intending for my next 5e(ish) campaign is basically this: more than one ability score of 9 or lower, or none above 13 - after racial ASIs - and you can reroll. Also intending to allow players to increase their key ability score (i.e. the one needed for multiclassing) one point by reducing another by two, and only after choosing race, class and rolling stats.

You might well ask, 'why not play B/X then?' To which I would answer, 'look, a squirrel!'

 

[Scott Christian]

5e does a great job at helping players not roll stats. In fact, it is much better from an encounter standpoint if they aren't rolling.

That said, when we used to roll, we used our fortune/misfortune as a storytelling idea. And we were in our teens. We never cared. Heck, one of my favorite characters of all time, his highest ability was - 13. Yet, the DM did a good job with magic items and we were good to go.
There is way too much debate over abilities. So much so, that WotC eventually had to change the rules in the PHB so everyone could start with a 16 instead of 15.

If I were to run a game now with players rolling for stats, I would make sure everyone wants that, and that everyone is okay with their possible year long commitment being decided on a six ability rolls.

 

[le Redoutable]

Exactly like rolling for HP !

 

[Necrofumbler]

The game is balanced around asome given character power level, and varying stats GREATLY affect this balance.

While the general idea of rolling randomly has some appeal, you have to remember that this balance discrepancy is going to be a CAMPAIGN LONG aspect.

Think if, game #1, you allowed each player to pick up a green d4, or a red d4, then they all roll all at once. Then you say: okay, those that rolled a green die, every weapon you wielld gains that die number as a "non-magical innate divine blessing", which is a bonus to every action based or DC based on your main stat, until end of campaign. But for those that rolled a red die, it is the erxact opposite, a non-magical unremovable permanent demonic curse to all their main stat actions and DC, untill end of campaign.

Then watch the "unfair!" whinings and tantrums blow out from the red die players, and the players dropout. Eventuallly this repeats with new players, until you have EVERYBOPY with a green-die PC remaining.

Over 30 years of DMing, my personal expewrience is that 99% of players WNATING to roll, they don't REALLY want to have the randomness. At allo. what they really want is the opportunity to get a "more powerful" PC. Ideallly, more powerful than the other PCs. the rolled low? They wwhne, complain, ask to roll again, invent some excuses the very next game thhat "they don't oike that class and want to roll a new PC", or even outright SUICIDE their PC to make sure they get to roll a new one. O
In non too obvbious ways of course, such as taking extremely angerous ruisdks, while not fully healed, covering it with a bit of dramatic roleplay and "background personality reasons". Oh my PC is very, very proud, and also protective of X (other PC he "likes"), so he won't accept to get healed only for a few measily scratches (despite being only at 20% HP he just says "Oooh, I'm fiiiiine!" saying wen he inevitably die "
That was just my PC being too proud to admit he was badly hurt, you should have done a medical check to find ouyt how he REALLLY was hurt if you really wanted to heall him up!"), saying he perfers the healers keep their healing magic available to cure X, "in case it's needed".

Then playing the reckless imppatient self-sacrificing hero that tries to solve everything by taking on the most risks. "That corridor is full of traps but hey, we have enemies around, disarming all those traps would take way too long, SO I RUN ALONG THE CORRIDOR TO TRIGGER ALL THE TRAPS!"

Of course they typically die before the end of the 1st session.

Then it's Lather rinse Repeat until they finally get that uber-stats PC.


If prevented to do any of thatm, they tend to simply drop out of the campaign.

Thus clearly proving THEY WERE NOT AT ALL INTERESTED IN THEIR STATS BEING 'RANDOM', BUT IN THEM BEING 'MOAR POWARFULL FOR FREE'.

It's almost always done as a form of desire to get some free extra power creep. NOT as a way to get "variability".

I use Default Array and any new player STRONGLY insisting to roll, is just not reinvited. It always leads to way more problems than it's worth.

 

[FrozenNorth]

Because rolling is fun? And because inequity is not fun? OP’s solution sounds like a win to me.

For what it’s worth, that’s what our group has been doing for years. Roll one series 4d6K3 each. These are the stats arrays available for this campaign, including villains with PC classes.

I’m with @EzekielRaiden on this. I tried a variant of your approach but it ended up unsatisfactory:
(1) a lot of arrays generated were clustered around 10-14, being less interesting to play;
(2) most players ended up choosing the same stat array anyway (or functionally indistinguishable ones) so it didn’t differentiate between characters anyway.

 

[FrozenNorth]

Rolling stat arrays is a way to introduce elements of variance between games without forcing variance between players. I take that you don't like it, but I do. It's not different from varying between 27, 30, or 32 point-buy.

Sure, but my conclusion was that this was unsatisfying: there was no increase in the variance between characters, so it was just like playing a higher point buy game.

What you don’t get, is the one character with like a 6 Cha, who is playing up their low stat to the hilt.

 

[Laurefindel]

I’m with @EzekielRaiden on this. I tried a variant of your approach but it ended up unsatisfactory:
(1) a lot of arrays generated were clustered around 10-14, being less interesting to play;
(2) most players ended up choosing the same stat array anyway (or functionally indistinguishable ones) so it didn’t differentiate between characters anyway.

Interesting, this is very far from my own experience. Maybe less so 1); the 10-14 clustering is inherent of the 4d6-drop-lowest method. It has nothing to do with the sharing of arrays but out of four series, one invariably ends up something like 11,13,12,11,12,13. As for 2) the plethora of classes (and subs), races (and subs), backgrounds, and builds within the same permutations has always ensured high mechanical diversity and differentiation between characters in my experience, not to mention roleplay elements. At any case, it's still better than imposing a single 15,14,13,12,10,8 stat array and is comparable to point-buy which often produces just as functionally indistinguishable scores for characters. Similarly, methods where players each roll three series and get to choose the best end up with functionally comparable arrays (when scores are interchangeable with abilities at any case).

But functional distinction was never a goal of this method; on the contrary, it is devised to eliminate gross power imbalance from the start, which in turn reduces distinctions between characters by pruning the weaker elements. Mostly, the greatest distinction is in between campaigns.

Now I never said this method is the best for everyone and for every campaign, but I believe it has a solid place among (individual) rolling for ability scores, standard/elite arrays, point-buy, and rolled matrices.

 

[Laurefindel]

Sure, but my conclusion was that this was unsatisfying: there was no increase in the variance between characters

Ah, didn't see that post. It would have saved me to write a long one!

Increasing the variance between character is not the intent. On the contrary, the intent is to reduce power discrepancies caused by high variations between characters. If the goal is to increase in the variance between characters, then yeah, don't use this method; it's counter-productive

 

[le Redoutable]

... perhaps to buy exceptional stats you must also buy very low ones ( to the risk of ban lol I want to use material given by autists )

 

[Scott Christian]

Over 30 years of DMing, my personal expewrience is that 99% of players WNATING to roll, they don't REALLY want to have the randomness. At allo. what they really want is the opportunity to get a "more powerful" PC.

This. I might lower it to 80 or 90%. But the vast majority just want stronger characters.