- 5e

A combination of flexibility and efficiency, with a dash of generalization.

What does generalization mean?

If you read any guide (including ours) that tries to explain optimal choices, these discussions are done in a vacuum and only looking at the most reasonable/practical lines of a character build; they also typically have a vague end goal in mind (typically level 20, even though most characters won't get there).

Let's compare what a high elf gets you compared to a variant human.

High Elf

Variant Human

Custom Lineage

What does this mean?

In a sense, an elf is trading a feat for one additional stat, darkvision, advantage on saving throws against being charmed, no need for sleep, the Elvish language, an cantrip, and profiency in four weapons. The problem here is that it's difficult to get all of these extra traits to work on a character without wasting something (especially without variable skills from Tasha's Cauldron).

How often will the Elvish language benefit you (or your party)? How about not having to sleep, especially if the rest of the party has to sleep anyways? Advantage on charms is pretty nice but how often does that come up? Does the combination of all of these extra traits outweigh a feat? Maybe but definitely not when talking about these traits generally speaking.

Okay, so now let's think about a character that can benefit the most from this. Let's make a character that can benefit the most from the cantrip Booming Blade without normally having access to the cantrip -- that eliminates artificer, wizard, sorcerer, warlock, and a few subclasses for clerics, fighters, and rogues. In a vacuum, we probably shouldn't rely too much on multiattack -- so fighter is probably out of the picture at this point since they get 3 or 4 attacks from multiattack. Monks rely on taking an attack action for martial arts bonuses, barbarians rely on rage.

Now let's break down the weapon profiencies... Let's get rid of Longsword right away since that's a strength weapon and we're looking at dexterity instead. Rogues already have profiency with shortswords (and would probably use a rapier anyways) -- I'd still consider this okay for a rogue because this gets them profiency with longbows (already proficient with shortbows because they are simple) but this might be a stretch because I think I'd rather have a hand crossbow in most cases anyways.

I think you can see the problem with these two lists -- the classes that benefit the most from cantrips and the classes that benefit the most from weapon proficiency don't overlap much. I guess that's good because it narrows down our character concept quickly:

I put asterisks next to bards (especially relevant because they don't get extra attack), paladins, and rangers because I think these three classes could still make use of the high elf's free cantrip, especially if Tasha's Cauldron of Everything rules are available and you can trade in weapon proficiencies for other proficiencies.

Can you see how writing a general guide for a class would make it difficult to talk about each race's benefits for it? Guides need to focus on one specific branch of a build and each time a guide goes for generalization, unique or hyper-specific abilities get lost.

What about Multiclassing?

That's kind of the problem here. Taking one level of artificer, wizard, warlock, or sorcerer gets you a lot more than just one cantrip. Multiclassing requirements makes this a little harder to justify since we probably won't have enough charisma to dip into warlock or sorcerer.