The first version of D&D that I played - Basic - instructed players to roll 3d6 for each stat and assign the number to that stat. I believe these are the rules for OD&D as well.

In AD&D the "standard" method was to roll 4d6, ignoring the lowest dice, five times and then pick which stat you assigned each number to. The same system was used in 2nd, 3rd and 3.5 editions.

4th and 5th both use a standard set of scores, assigned to stats as the player wishes.

In all editions there were variations to the "recommended" rules, most notably the plethora of systems introduced in Unearth Arcana. For the purposes of these questions, we'll bypass those and stick with the typical versions.

Are my recollections of the different systems correct? Can anyone explain the drivers that lead to the changes between editions? I am particularly interested in the switch from 3d6 to 4d6, dropping the lowest.


closed as too broad by BESW, Miniman, harlandski, Wibbs, enkryptor Jun 2 '17 at 9:53

Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • \$\begingroup\$ I'm not sure the question is answerable in its current state. The difference between 3d6 and 4d6 methods is obvious - the second one gives higher values. It is being said that PC should be better than average as "heroic" characters. Do you ask about an official developers' comment or what? \$\endgroup\$ – enkryptor Jun 2 '17 at 9:51
  • 4
    \$\begingroup\$ The why part of this question is hugely problematic. For an answer to evidence designer intent for each of the changes in character creation between each of the many editions of D&D makes it extremely broad, and imo unanswerable. If you don't ask for designer intent, it devolves into a purely opinion driven set of answers \$\endgroup\$ – Wibbs Jun 2 '17 at 9:52
  • \$\begingroup\$ This is actually a collection of related questions. I could provide citations about what enkryptor commented but this question, IMO, needs to be divided into a) an history question asking if the edition divide is correct (or directly about which official generation method was used throughout the different editions), one that asks why the change from 3d6 to 4d6 (which I promise is answerable) and possibly a third one for the change from rolling to fixed values. \$\endgroup\$ – Zachiel Jun 2 '17 at 10:44
  • 2
    \$\begingroup\$ You last paragraph is the core question: the change from 3d6 to other choices. That change was from D&D to AD&D. Suggest you re scope this question to cover that change only, as it is the root of subsequent changes. The Point Buy is a separate question Also a minor note, AD&D you rolled six time, not five times ... for stats. \$\endgroup\$ – KorvinStarmast Jun 2 '17 at 12:04
  • 1
    \$\begingroup\$ Lastly, the change to the d20 system (3e to present) should probably be a separate question since the change from TSR to WoTC brought about quite a few conceptual differences. \$\endgroup\$ – KorvinStarmast Jun 2 '17 at 12:15

You can notice that globally the older is the generation system the more random it is. If you roll your dice you can get anything between 3 and 18, so technically you can get a character with 3s everywhere, or at the opposite one with 18s everywhere.

At first it wasn't much of a problem since characters are expected to die soon (see this or that) anyway and were replaced quickly: if he gets bad scores he will just die early, and an other one could be rolled, if he gets very high scores it won't destroy the game balance for long since he will die anyway.

But then people started to actually get attached to their characters, and you can't really get attached to a character you know won't be able to do anything useful because he has such low stats. One easy way to solve this problem was "cheating" with the character gen to make it less likely to get bad scores: on 3D6 you have 1 chance over 216 to score a 3 (~0.46%), but with 4D6/drop lower it falls to 1/1296 (~0,08%)1. The alternative generation methods were just other ways to achieve the same goal: make it more likely to get a playable character.

Then people came to prefer balance to rolling dice. It meant systems like point-buy or a standard set of scores.

1 Here is a graph of the distribution of probabilities for the two distribution. Thanks to xNGTMx from whom I "stoled" this link!


Some of the information contained in this post requires additional references. Please edit to add citations to reliable sources that support the assertions made here. Unsourced material may be disputed or deleted.

  • 7
    \$\begingroup\$ [citation needed] \$\endgroup\$ – the dark wanderer Jun 2 '17 at 8:23
  • 2
    \$\begingroup\$ Hi Anne. Can you add citations for these assertions? We expect subjective answers to back it up with either personal experience or experience of someone else, and answers to this question would either need to cite designer statements or experience of what was happening on a wide scale. We cannot accept speculative responses on matters like this one. \$\endgroup\$ – doppelgreener Jun 2 '17 at 8:55
  • 1
    \$\begingroup\$ This answer explains the diffirence between 3d6 and 4d6. But the OP asked about "the drivers that lead to the changes ". \$\endgroup\$ – enkryptor Jun 2 '17 at 9:56
  • \$\begingroup\$ @doppelgreener: added some references, I can find more if it's needed. \$\endgroup\$ – Anne Aunyme Jun 2 '17 at 11:40

Not the answer you're looking for? Browse other questions tagged or ask your own question.