5e (Lack of) Core Character Build Mechanics

There isn't going to be much discussion here, I was going to to try to analyze the basic build mechanics that WotC used in creating its races and classes. But it was proved pretty early on in my analysis that they don't have any sort of benchmark value applied to any of their abilities.

Ability Scores

First, there is the issue with the variable cost of Ability Scores. Buying an Ability Score anywhere from 9 to 12 costs 1 point to raise 1 level.  Buying up for the levels between 13 and 15 cost 2 points to raise 1 level. This is analogous to the 3.X and Pathfinder point buy costs, so we can extrapolate that 16 and 17 cost 3 points to raise 1 level and 18 costs 4 points to raise 1 level. There is no point buy that I know of for 19 and 20 in 3.X or Pathfinder, but I think we can continue the progression to make 19 cost 4 points and 20 cost 5 points.

What this means is that a +2 Racial Ability Score Bonus has a value of anywhere from 2 points (starting value 10 or less) to 6 points (starting value of 15). This same issue applies to Ability Score Improvements you get from level which would have the value of 2 points (starting value 10 or less) to 9 points (starting value 18). For that matter taking the two +1s can have a value ranging anywhere from 2 points (adding to two stats of 11 or less) all the way up to 10 (adding to two stats of 19). This also means that Feats are the equivalent of anywhere from 2 to 10 points of value by this logic.

Saying that the variable cost only applies to 1st level removes the problem of the value of Ability Score Improvements per level but still leaves the trouble of Racial Stat Bonuses have variable values.

So the only possible way to deal with this is to assume that base, starting Ability Scores were not considered when building classes. Given the random determination option, this seems most likely.

Races

Okay, so with the determination to disregard base Ability Scores in analyzing D&D classes and races in the hopes of reverse engineering the basic system of character building they use, we move on to classes. Unfortunately, things immediately breakdown again in the example of dwarven subraces.

Hill Dwarves and Mountain Dwarves both get a +1 to an Ability Score, which we've decided no longer has a variable value after leaving basic score determination. So that seems fine. However, the other facets of their subraces are problematic.

Hill Dwarves get Dwarven Toughness which is essentially the equivalent of taking a half-value version of the Tough feat.

Mountain Dwarves get Light and Medium Armor proficiency, which is essentially the equivalent of taking a half-value version of two different feats: Lightly Armored and Moderately Armored. Both of those feats give a +1 Ability Score and proficiency in one type of armor.

So now, the Hill Dwarf subrace is the equivalent of taking a Feat and the Mountain Dwarf subrace is the equivalent of taking one and a half Feats.

The Dwarf as a whole has +2 Ability Score, 4 Weapon Proficiencies (1 Feat a la Weapon Mastery), 1 Tool Proficiency (1/3 of the Skilled Feat), a highly specialized from of the History Skill (1/3 of the Skilled Feat), Darkvision (possibly 1/3 of a Feat assuming that Darkvision is the equivalent of other perception based abilities acquired through Feats: Alert, Dungeon Delver, Observant, etc) and Dwarven Resilience (approximately 1/2 a Feat if Poison damage is roughly equal to "damage dealt by traps" as per Dungeon Delver).  Making the Dwarf race approximately the same as taking 3 1/6 Feats. With the subraces that would be 4 1/6 feats for Hill Dwarf and 4 4/6 feats for a Mountain Dwarf.

The human is similarly unbalanced. The basic human has +6 Ability Scores while the variant human has +2 Ability Scores, +1 Feat, +1 Skill.  The Feat is the equivalent of +2 Ability Score (going by the Ability Score Improvement equivalency). The Skill is the equivalent of 1/3 of a Feat since the Skilled Feat gives three Skill or Tool proficiencies.

So now the base human is the equivalent of 3 Feats and the variant human is the equivalent of 2 1/3 feats as compared to 4 1/6 and 4 4/6 for Dwarves.

Classes

Just looking at the races I've ruled out the idea that WotC built their races and classes from an underlying system. There is thus no need to analyze the classes. However; just comparing to feats like Tough, Skilled and Weapon Master; I can say that the 1st level of Fighter is basically the equivalent of 12 Feats

32 weapons beyond the basic 5-6 that the least weapon-centric classes get, All Armor and Shields, on average +2 HP per level over d6 Hit Dice, Fighting style (estimated at 1/2 Feat) and Second Wind (estimated at 1/2 Feat).

Assuming all were built on the same benchmarks, this would imply that Full Spellcasting, 3 Cantrips and 6 starting spells is the equivalent of about 10 Feats (Ritual Casting and Arcane Recovery accounting for 2 more).

However, that runs into problems when you add Cleric to the mix given Full Spellcrafting, Divine Domain, Light, Medium and Shield Proficiency, and all Simple Weapon Proficiencies as well as knowing every spell on their list.

Conclusion

The D&D team did not use a core system to build their classes. This is nothing new. Every edition of the game from 1st all the way through the current one, including 4th edition similarly did not design their classes and races from a basic underlying system. This same philosophy of character building can be seen in Palladium games, White Wolf systems and many other games.

Is this any worse or better than games which use a core, underlying system such as HERO, Fate, GURPS or BESM? No. The value of play is still going to be based almost entirely on the GM, the players, their creativity and so on. Presumably, the classes have all been thoroughly playtested to ensure that they are equivalently able to contribute to game play.

The impact of this fact comes into play once we begin trying to tweak the mechanics and do some homebrew stuff.

Take HERO and assume you use one of the superhero templates in the Champions sourcebook. Once you've gone through and taken the pre-created options and put them into place you can then go around tweaking things to your desire. Reduce this attack by 1d6 to get an extra 5 points to increase your Intelligence or add some skills. Drop this ability entirely and make something else which wasn't on the template list of abilities. Take a power from a different template's options which had the same point value.

All of this is easily done with no problem. The point values of each thing you're moving around or tweaking are either listed or easily calculated and you are staying well within the previously playtested rules for character creation. The GM is expected to make a judgment call on any particular character to make sure it stays within the group's desired theme and doesn't surpass any decided maximums, but the tweaks themselves are easy.

On the other hand, in D&D, there is no clear equivalency. As soon as you make any tweak you step outside the range of what the original company playtested and you are now risking unintended consequences. This means you might have to make many changes throughout the game to the home-brewed material in order to keep it from being either too effective or not effective enough.

Basically, the lack of a central benchmark system means that there is a mechanical inflexibility which makes it difficult to adjust the game outside of the narrow set of genres and subgenres it was designed for.

This will also impact the system as a whole since even if they do present on OGL for third party groups to create their own games based on the same set of things, it presents a lot more difficulty and playtesting to rework 5e out of high fantasy than it does to rework Fate or PbtA into any genre.

As such this lack of a core benchmark is mostly an impact on the people that homebrew and the people that would use the system to create their own games (assuming WotC creates an OGL for this system...if they haven't already.)