In last week's Legends & Lore column Mearls talked about his favorite solution of solving a problem by removing it entirely, and fixing many, smaller issues with a few major changes. Of course he demonstrates the former by barely even changing the problem, and in the latter he cites advantage/disadvantage, which not only has its share of problems but was apparently inspired by a mostly impossible and unlikely majority use of 4th Edition's action points, so...yeah.
This week wraps things up with the last two precepts of what he considers to be design finesse: rules that are invisible to players that do not need them, and going with the flow of the game.
Mearls believes that one way to achieve elegance is by keeping rules localized, by which he means making it so that a rule is only really applicable to those who "need" it or "putting the burden on that player". To a point I agree with this: if I am playing a street samurai in Shadowrun, then I probably do not need to know how magic works. Unfortunately as with many things I have seen come out of Legends & Lore, his example using opportunity attacks leaves something to be desired.
The problem with 3rd Edition's attacks of opportunity was that they were unnecessarily complicated: numerous actions may or may not have triggered one, and there were other factors like feats, reach, class features, magic items, spells, etc that could negate some of them. For example Combat Reflexes changed the number of attacks of opportunity you could make in a turn, and Disarm let you ignore the attack of opportunity that triggered when you tried to disarm someone. Hell, page 141 was mostly tables and footnotes that told you which actions provoked them.
4th Edition not only shortened the term to opportunity attacks, but boiled it down to two very easy to remember cases: moving out of a threatened area and making a ranged attack (which included area bursts). That is it. The only thing that I could say routinely shook this up was if a monster had the Threatening Reach trait, which was explained in each monster's stat block so you did not have to look elsewhere.
5th Edition takes things a step further by getting rid of the ranged attack trigger, meaning that now you only have to worry about them when you move away from an enemy. While this is technically simpler I am not sure if it is necessary given that everyone still needs to know how the base rule works anyway; does it really complicate things that much by having a second, easily remembered and oft-used trigger?
He claims that spellcasters "have never been eager to engage in melee", which is true if you ignore—for starters—clerics, druids, rangers/scouts, paladins, psychic warriors/battleminds, valor bards, swordmages, and depending on edition various wizard and sorcerer builds, prestige classes/paragon paths (like spellswords and Spiral Tower wizards respectively), and so on and so forth. Since players can apparently not be bothered to remember a second trigger to a rule that they must already know, the solution to deter spellcasters from melee—based on his assumptions—is to have them remember another rule entirely.
That does not sound very elegant.
Do not get me wrong, I actually like the idea of concentration spells. I think 4th Edition did them better with the sustain mechanic, but at least this rule has precedent in fiction where a wizard must concentrate to maintain a spell with distractions potentially disrupting it (as opposed to, say, spell memorization, slots, and levels). I do not even mind that it requires its own rule. Where I take issue with it is when Mearls tries to claim that this is a more elegant solution. Well that, and that only characters with access to concentration spells will need to know how it works.
While technically fighters and rogues do not need to know this rule, they are still going to want to know about it because it will help inform their decisions when they are fighting monsters that use concentration spells, since they can potentially disrupt them. Also, spellcasters like the cleric and paladin are not only notorious for getting stuck in melee, but they also have full access to every spell of a level that they can cast: do you honestly think that they are not going to end up knowing what concentration means? I highly doubt that a cleric or paladin player is going to see all these references to concentration and not bother to look it up.
In other words it is still handy for non-spellcasters to know the rule, and pretty much every spellcaster with a concentration spell somewhere on their list is going to know it.
The second part of the article is about creating rules that stem from the player's understanding of how the game is supposed to work, somehow without any knowledge of the rules. In a game like Dungeon World I think this works, partially because the players know generally what to expect from the results (10+ is great, 7-9 is good, perhaps with a cost, and 6- is bad), and partially because so much of the results are dependent on the narrative. In a game like Dungeons & Dragons? Not so much, because not only are most outcomes solidly defined, but they can have unexpected factors and results.
An example of this is is grapple. You would think that at least you would end up spending your turn mostly preventing the other person from being able to do much except try to pry you off, but in reality all it does is reduce the target's speed to 0; they can still hit you—or anyone nearby—just fine. If you manage to keep it going you can spend another action on your next turn to try restraining the target, which not only gives your allies advantage when attacking it, but imposes disadvantage on its attacks...except against you. I have had players consider grappling before seeing how much effort it takes, then just going with an attack because it was both easier and more effective in the long run.
Cover works a bit better (assuming that all things are equal), but in either case I think a more realistic outcome is that players will consider a course of action and then want to know the rules before committing to anything: they are not going to just try to do what seems logical because they have no idea what the results, odds or other factors are. I would not say that this is exactly inelegant, probably not as elegant as Mearls thinks it is given that it interrupts the flow of the game and expectations vary, but is more or less par for the course when it comes to Dungeons & Dragons.