Yearly Archives: 2012

Meaning first

There has recently been another round of discussion about associative and dissociative mechanics. Here is Justin Alexander’s restatement of his original thesis, a post by Zak about diegetic and extradiegetic thinking, Carter Soles on healing surges, and Jeff on whether or not the d30 rule is dissociated. Zak doesn’t use the words “associated” or “dissociated” anywhere, but it’s really the same issue from a different angle. Are players reasoning about cause and effect within the game world or within the structure of the rules?

Consider two examples:

  1. The Labyrinth Lord spell “levitate” (LL page 33):

    For a number of turns equal to the casters level +6 turns, the caster can move up and down as he wishes. The caster mentally directs movement up or down as much as 20 feet each round. The caster cannot move horizontally, but could clamber along the face of a cliff, for example, or push against a ceiling to move laterally (generally at half base land speed).

  2. The Fourth Edition level seven fighter power “come and get it” (4E Player’s Handbook page 80):

    Every enemy (but not ally) within a 15 foot radius is shifted two squares (10 feet) and become adjacent to you do so, and then you get to make a weapon vs. AC attack against them.

In the first example, the meaning is primary. The caster is no longer subject to the limitations of gravity, and can move themselves around at some set speed. In the second example, the effect is primary. Enemies come to the PC to be hit. We don’t know why. Did the PC taunt the enemies? Did the PC lasso the enemies? Either might make sense in different cases, and many people don’t bother at all with narrative explanations. Obviously, it is not an either/or thing, but one of association or dissociation is usually dominant. In the first example, the duration in turns is somewhat dissociated (why would levitation only be available in game-convenient durations that are a multiple of ten minutes?), but unobtrusively so.

The common example of a limited martial resource (such as a daily power) as a dissociated mechanic is really just a special case of this more general “meaning first” or “effect first” principle. It is easier to make a balanced game using effect first design, which is why 4E leans so heavily on dissociated mechanics, as mathematical balance was an important goal for that system.

The examples above also show how this issue is orthogonal to realism. Levitating is obviously not realistic (at least based on my experience), but it is associated. Enemies moving toward you and then you hitting them is realistic, but it is dissociated. Why are they moving toward you? Why do you as a player have the ability to affect the movement of your enemies? You can come up with an after the fact narrative explanation, but the meaning is secondary and the effect is primary.

As has been noted in some of the links above, there are some dissociated mechanics in traditional D&D too. Hit points and experience points are notable. To elaborate, the effects of gaining a level are primary (more staying power, additional spell capability, followers, whatever), and the narrative explanation is after the fact. Was the PC training? Did a demon grant them extra power? We often don’t know. HP and damage are probably the most problematic. How can cure light wounds help with the loss of luck and fatigue? We can come up with an explanation, but it’s certainly not obvious. Weapon and armor restrictions are another common dissociated irritant that has prompted many house rules, my own included (see here for weapon damage by hit die and a system to do away with armor restrictions).

The major difference, as I see it, in how new school games and old school games do things is that the dissociated mechanics of old school games affect encounter-based problem solving less than they do in new school games. And when they do affect problem solving (perhaps equipping every character in OD&D with daggers because they are cheaper and all weapons do 1d6 damage) that is considered pathological, and either fixed with house rules or condemned as against the spirit of the game. From the standpoint of creative problem solving, if meaning is first, the potential effects are limitless, and this is in my opinion why many people are uncomfortable with the extensive use of effect-first mechanics.

Edit: I’m sure this post was influenced by On the Failure of Tactical Combat over at Hack & Slash (though his post focuses, unsurprisingly, more on combat). So go read that too.

First Level Clerics & Spells

There is a really insightful comment over at Grognardling by Mike Monaco (quoted partially) about clerics:

Clerics in B/X don’t get a spell at first level so they spend the first few adventures UNABLE to heal people. This gives them a chance to learn more roles, like throwing holy water, turning undead, backing up the fighters or defending the mages, and so on.

1974 D&D and the various incarnations of basic (Holmes, Moldvay, Mentzer, and the Rules Cyclopedia) do not give clerics a spell at first level. Strangely, some of the retro-clones (Labyrinth Lord, LotFP) do give first level clerics a spell. The clones that don’t are Swords & Wizardry Core, Swords & Wizardry WhiteBox, and Original Edition Characters for Labyrinth Lord. I believe the AD&D Player’s Handbook was the first place that first-level clerics are given a spell, though more knowledgeable people can correct me if I am wrong.

In any case, I much prefer the no spell at first level version of the cleric. It highlights the martial aspect of the character class, which I think is very important to the cleric as demon hunter and crusader rather than purveyor of cure light wounds.

Maturity & Gaming

Back in the 90s, when I played Second Edition (this was the edition I started with), my friends and I almost always used 4d6, drop the lowest, re-roll 1s, and arrange to taste when rolling up characters. Max HP were generally awarded at first level. I usually played elves, most commonly elven wizards. I often came up with relatively intricate back stories for characters before play, and our characters rarely died (though we did not consciously run a “low lethality” game; such things were not considered). We enjoyed the process of rolling dice, mostly for the variation they introduced, but we didn’t want randomness to rob us of whatever our predetermined vision was.

By that point in TSR’s history, many of the core elements of the original game had become obscured or removed. For example, some of our characters had animal companions or other minions, but we didn’t really play with hirelings. We scorned random encounters, never used reaction rolls, didn’t use hex maps, and ignored all manner of other early techniques (mostly, now that I understand them, to the detriment of our games). This was also certainly affected by the amateur thespian advice littered around the Second Edition books, and the rise of the White Wolf Storyteller games around the same time (which all the older, more sophisticated kids were playing). I flirted a bit with The World of Darkness in high school, but even at that time D&D was always my game of choice.

I wonder, though, if play style is somewhat generational, not in terms of when you were born, but in terms of maturity level. I started playing during my adolescence, and much of one’s life during that period is about defining yourself. Thus, I think there is a strong pull towards wish fulfillment play (“this is what I want to be”). Gary and Dave, when they originally developed D&D, were obviously not in that sort of mental space. They were already mature adults, and were building an intricate game informed by their wargaming experience. There is still a strong strain of childlike wonder present, but that is different than wish fulfillment. To perhaps oversimplify, if adolescence is about self creation, then maturity is about self discovery. The parallels with “character builds” and its alternative “development through play” should be obvious, I think.

Hex Stocking

I’m in the process of trying to come up with a random hex stocking method that works for me. My point of departure is the traditional dungeon stocking method:

  • 1-2 monster (3 in 6 chance of treasure)
  • 3 trap (2 in 6 chance of treasure)
  • 4 special
  • 5-6 empty (1 in 6 chance of treasure)

In the context of wilderness stocking, “monster” would be interpreted as a lair or dungeon, “trap” as a hazard of some sort (or perhaps an abandoned ruin that is uninhabited but still dangerous), and “special” as everything else (including, probably, settlements). Each of those categories could have a subtable or set of subtables to determine the type of lair, etc. I want to keep the system as simple as possible, but I think I need more than this for inspiration, because I don’t find myself actually finishing a stocking process. That’s a sign to me that I need more help from the tables.

One thing that is blocking me is how settlements interact with the stocking. I could just place the settlements, and then stock the areas between them, but I kind of want the generator to do that work for me. It seems that there are really only three or four meaningful settlement sizes for my purposes here. Stronghold, town, village, and isolated settlement (outpost, traveler’s inn, farmstead, etc). Maybe half of the special results would result in some kind of occupied settlement. Ruins would be covered in the monster, trap, and empty (when with treasure) results on the main table.

I don’t care much about things like logical food supplies (I can come up with after the fact explanations), but I do sort of like the idea of graduated civilization and wilderness. Here is another place where three or four categories seem appropriate: civilization (town, fortress, etc), threatened ground (the border between civilization and wilderness), and wilderness. There is a mathematical choice to be made regarding how this works: should the stocking roll be independent or related to results in adjacent hexes? If the process is independent, then we can infer the level of civilization (and thus danger) from the resulting distribution, which will end up being regular.

If it is dependent, then the process would be more like an organic outgrowth from some seed hex (probably the starting town), which would have some chance of going down in civilization level and some chance of going up. The chance of civilization level decreasing as you expand outwards would probably be greater than the chance of civilization level increasing, resulting in a setting that is dominated by wilderness (and thus adventure opportunities). Victor Raymond uses a system like this to generate terrain type in his Wilderness Architect series of articles in Fight On! (issues #2 and #3). He places settlements by determining random direction and distance from the starting settlement.

So, to expand the the “4 special” hex result:

  1. Trick (magic statue, etc)
  2. Settlement
    1. Stronghold (50% chance includes another settlement)
    2. Town
    3. Village
    4. Outpost

The meaning of this table (based on expected values): 1 in 12 hexes will contain a settlement, and 1 in 48 hexes will contain a stronghold. Following the DCC recommendation of 100 miles square, I am considering approximately 16 x 16 six mile hexes, which is 256 hexes (and also compatible with my ideas on hex zooming). Overall, such a wilderness would have (approximately) 86 lairs, 86 empty hexes, 43 specials (21 of which would be settlements) and 43 hazards. How does that distribution look? One thing that does not seem quite right is that an outpost is just as likely as a stronghold using this scheme, but on the other hand this will lead to around 5 strongholds on the map, which seems to be about right (especially if they are of varying levels of power and influence). Also, the “monster” result would include things like bandit forts and the towers of evil magicians.

Any ideas welcome.

Magic Disciplines

I read this post over at The Mule Abides about Starting Magic-User Spells, and for some reason that got me thinking about how higher level spells often seem to be improved versions of lower level spells. With a bit of mental flexibility, many spells almost feel like they belong in a progression. For example, consider a hypothetical discipline “gravity” with the following effects: floating disc, levitation, fly, telekinesis. Or a discipline “interposition” with the following effects: shield, protection from normal missiles, globe of invulnerability.

Extending this idea further, what if magic-users just had access to the discipline as a whole and never needed to learn (or prepare) individual spells? Higher level effects would just be harder to cast. Following the conventions of the 3 LBBs and B/X, there are six levels of effects (or spells) within each discipline. Spells are not learned; disciplines are determined at first level.

A magic-user gets one discipline, plus one additional discipline per point of intelligence bonus. All magic-users would get the metamagic discipline for free. Yes, this means that magic-users with a higher intelligence are more flexible, but we’re all playing 3d6 in order, no? A magic-user with an 18 intelligence is assumed to be a rare an wondrous occurrence. For example, a magic-user with an intelligence of 14 would have access to two spells, determined randomly (or selected, you cheater). It is not possible to switch disciplines after character creation.

The system I am envisioning for casting is similar to these variations of Vancian magic, but magic-users don’t need to prepare any spells beforehand. They may cast any spell which is of level less than or equal to half caster level (rounded up). For example, a 3rd level wizard may cast up to second level spells. When casting a spell, a magic-user must make a saving throw versus spells. Upon success, they may use spells from the discipline again in the same day. Upon failure (but not a roll of 1) the spell still goes off, but the magic-user may use no spells from that discipline again until they have had a good night’s sleep and studied magic books. If a 1 is rolled, the spell fails or backfires in some inconvenient manner (use the spell fumble system of your choice).

Higher level effects may be attempted, but at greater risk. The same procedure is used as above, but the saving throw takes a penalty equal to the spell level, and the save must succeed for the spell to go off. A roll of 20 is always considered a success. Also, the fumble range is extended by the level of the spell. So, if a 4th level magic-user (max spell level: 2) is attempting to cast a 5th level spell, they roll their saving throw with a -5 penalty and the spell backfires on rolls of 1 through 6. This same procedure will obtain until the caster reaches 9th level, when the save penalty disappears and the fumble range drops to 1. In other words, the progression is not linear (though the base save versus spells does improve at 6th level and 11th level); this is intended. You don’t get it, and don’t get it, and then it finally clicks.

Thus, magic-users may attempt any effect at any level, though doing something like conjuring an elemental or attempting telekinesis at first level will almost certainly result in disaster. This could also be done with an ability check and some DC math, but I prefer the simplicity of the traditional saving throw. If possible, I wanted to build this system entirely with traditional D&D spells, using the more iconic ones where possible. Here is an (incomplete, preliminary) example of how the disciplines might look:

Discipline 1st 2nd 3rd 4th 5th 6th
Transmutation enlargements
polymorph self, polymorph others transmute rock to mud stone-flesh
Illusion phantasmal forces hallucinatory terrain projected image
Divination detect magic locate object clairvoyance wizard eye contact higher plane
Necromancy VAMPIRIC TOUCH animate dead, magic jar death spell
Destruction magic missile fireball disintegrate
Ward protection from evil prot. from evil 10′ radius
Shield shield protection from normal missiles anti-magic shell
Domination charm person forget confusion feeblemind geas
Thuraturgy knock, wizard lock pass-wall
Gravity floating disc levitate fly telekinesis
Stasis hold portal web hold person hold monster
Optics light invisibility,  invisibility 10′ radius massmorph
Energy shocking grasp lightning bolt
Summon conjure elemental invisible stalker
Space rope trick dimension door teleport
Metamagic read magic dispel magic remove curse

Spells that look like this are originally from OD&D.
Spells that look like this are originally from Holmes.
Spells that look like this are originally from B/X.
Spells that look like this are originally from AD&D.
SPELLS THAT LOOK LIKE THIS ARE ORIGINALLY FROM 2E.

Note that by “originally” I mean showed up in a major ruleset; they may have also appeared earlier in a periodical like The Dragon or Strategic Review; this is not intended as an historical treatise. That being said, any corrections are still welcome. Shield, magic missile, and ventriloquism were from Supplement I: Greyhawk, not the 3 LBBs. Merciful Shiva, but the spell list exploded in Second Edition!

Ideally, I would have one spell per discipline per level. Sorry about the overloading of terminology; I hope the meaning is clear. Though maybe it’s okay if a few disciplines are just inherently more dangerous, like necromancy or summoning. In a final system, the level of a few effects would probably be adjusted (for example, animate dead could be level four).

This system seems to have several benefits:

  • Familiar D&D spell effects
  • Simple character generation
  • No time spent selecting prepared spells
  • Risk/reward trade-offs
  • Genre flavor (apprentices overreaching their power, etc)
  • Each discipline could be given to a player as a one-page handout
There are a few downsides too:
  • No spells as treasure (at least not in the traditional manner)
  • Scroll rules would need to be revised
  • Potentially more powerful magic-users?

This almost ends up looking like a White Wolf power system (three circles in celerity allows you to do X, Y, and Z) but with traditional D&D effects. Such a system would work well for cleric spells too (for example: purify food and drink, create water, create food), though I’m not sure the flavor works unless you are merging the spell lists like Akrasia’s colors of magic system.

Okay, the data part of this post is not quite complete, but it is taking way too long to cross reference all the rule books, so I’m just going to go ahead and hit publish, and if I still care later I’ll come back and finish that part. In any case, the idea should be clear.

OD&D ability scores

Ability scores are explained on pages 10, 11, and 12 of Men & Magic. Here is an alternate presentation.

Abilities
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Str Ftr XP -20% -10% +5% +10%
Int M-U XP -20% -10% +5% +10%
Lang. Common + Alignment +1 +2 +3 +4 +5 +6 +7 +8
Wis Clr XP -20% -10% +5% +10%
Con HP -1 per HD +1 per HD
Survive 40-50% 60-90% 100%
Dex Missile -1 +1
Cha Hirelings 1 2 3 4 5 6 12
Loyalty -2 -1 +1 +2 +4

I think this table makes the effects of ability scores much clearer without needing to shoehorn them into a common pattern. This table could be printed on character sheets, and then the character attributes could be circled, or highlighted like a bar graph, as shown below.

5E Playtest Reactions

I ran the playtest module for my group yesterday evening. There were five players, so all the pregen characters were in play. I didn’t bother with any backstory or justification, since we expected to play this as a one-shot. Reaction was almost entirely positive. The player who (I thought) was the biggest 4E fan said he didn’t miss anything from Fourth Edition, though he thought it played differently. Everybody enjoyed the advantage and disadvantage mechanic. We got about 2.5 hours of solid game play in (this is about average for us, since we game after work).

Here is some player feedback:

  • Uncertain about the healer’s kit mechanic. It seems like a strange requirement if HP are supposed to be abstract. Also, it might lead to PCs carrying too many healer’s kits.
  • They would have appreciated a bit more healing. My players are used to 4E healing surges though. They thought the clerics needed to prepare specific spells also, so they probably should have had a few extra healing spells following the rules as written exactly.
  • They liked constitution + hit die for starting HP, though it seemed high to me.
  • Simplified action economy was appreciated, though some sort of limited “parting shot” type of opportunity attack was suggested (I believe 1E and 2E had simplified disengagement rules that might work well).
  • Everyone liked split movement, myself included.
  • The cleric of Moradin (dubbed “clericadin” by the player) was played aggressively, as I would expect a traditional plate + mace cleric to be played.
  • The wizard player liked the balance of at-will cantrips and prepared spells.
  • One player said she missed the battle mat. She felt like she had to continually ask me to redescribe the physical relationships during combat. I do run some 4E combats without minis, so this was not entirely new, but when there are lots of combatants we usually use minis.

The PCs went in cave D first, one of the goblin lairs, not trying to be especially stealthy. They wandered into a dead end and were ambushed by a patrol of 6 goblins. They routed those goblins, killing half of them and letting the remaining survivors escape to warn the others. They went deeper, following the fleeing goblins anyways, and got caught in a heavy melee with more goblins who had overturned a table to use as a barrier.

The goblins had already taken the opportunity to pay the ogre, so the party ended up trapped between a group of goblins and an ogre from the rear. The wizard cut off any chance of parlay with the ogre by a shocking grasp to the nether regions, resulting in one very angry ogre. They ultimately prevailed, looted the ogre’s cave (giving the remaining goblins time to build barricades from storeroom junk and prepare). They ended up defeating those goblins too (though the leader escaped).

Some other highlights:

  • A goblin was frozen to the ground with a ray of frost and two of the PCs held a dexterity contest to determine who could get to it first.
  • The cleric of Moradin sampled the invisibility potion and his lips became invisible. This led to many “why so serious” references. He also got drunk off the ogre’s brandy, so we got to test the intoxicated rules.

My feedback as referee:

  • The ogre had too many HP.
  • The frontal assault strategy they chose certainly would have gotten everyone killed if we had been playing B/X. One PC was briefly in negative territory.
  • The rules seemed to promote creative tactical play without much overhead. For example, the halfling jumped on the ogre’s and stabbed him in the back repeatedly while the ogre flailed around trying to get the halfling off (attacks at disadvantage, I have him an intelligence check chance to think about slamming the halfling against the wall, but he failed that check).
  • There was little resource management, but it would have been hard for that to come through in a one-shot anyways. The rules currently seem to be of two minds on this. On the one hand, at-will light cantrip. On the other hand, healer’s kit, antitoxin, and healing potion.

5E preliminary notes

I’m running the play test Caves of Chaos tomorrow for my group, so I thought I’d put down my pre-play thoughts beforehand and then compare them to how it goes. My group is pretty diverse (a few very casual players, some who are primarily video gamers, some 4E fans, and one player with mostly 3E experience; none of them are very active in the online tabletop RPG community), so it will be interesting to see what they think.

I’ve mentioned previously that I like the idea of backgrounds and themes, and I still like the idea after seeing a few actual examples on the pregen character sheets. I’m not crazy about all of the abilities, but the underlying concept is solid, and a huge improvement over the raw skills and feats available in Third and Fourth. Backgrounds as presented are a slightly more elaborated version of the original “secondary skill” systems present in 1E and 2E. The backgrounds are a nice way to add some detail to a character without wasting time on extensive backstory that (let’s face it) rarely gets read by anyone other than the writer. Put the list of backgrounds on a random table and that would make a pretty good character creation tool (compare to the recent Before First Level posts over at The Dungeon Dozen). The backgrounds as given by the pregens are knight, priest, soldier, commoner, and sage.

First Edition Secondary Skills (AKA “backgrounds”)

Themes I see as select class features factored out of classes. For example, the backstab/sneak attack ability is not associated with the rogue class, but rather the lurker theme. So you can create a fighter with the lurker theme and that is a pretty good assassin. The magic-user theme grants the use of several cantrips, so a rogue character with the magic-user theme is a pretty good approximation of The Gray Mouser, who knew a few spells from his time as a wizard’s apprentice. Conan? A fighter with some sort of barbaric background and a theme like slayer or lurker. This is better than multi-classing, in my opinion. I would much rather see a limited number of classes with more themes than a huge number of classes. Unfortunately, many of the non-core classes are sacred cows (ranger, paladin, barbarian, etc) and so are likely to be full classes despite just being slight variations on the fighter. The themes as given are magic-user, lurker, slayer, healer, and guardian.

One of the problems I noticed is the lack of support for games that focus on resource management. The at-will light cantrip is the biggest offender, but there are several other problematic abilities. Detect magic is also an at-will cantrip, and dwarves have a racial ability called stonecunning which allows them to unerringly retrace their steps while underground; this makes mapping less critical. These features are fine, but should be available modularly, not as defaults within the core races and classes. To paraphrase Jeff Rients from a G+ conversation, I respect the right of anyone to play Omega Level asskickers, but I don’t see why there can’t be room for low rent bastardry at first level, and such play relies on resource management. Also, if you want the game to be about anything other than killing monsters, combat can’t be the first resort (in other words, the game has to be more deadly, at least optionally).

From a high level design point of view, a shift to a lower power curve is by far the most promising aspect of the new system. This might even have advantages for old school play over some early editions of D&D. Consider the escalating power levels of classes in AD&D, for example. A high-level AD&D fighter can pretty much hit anything that does not have an absurd AC. It is hard to get a full picture of how this might look from the play test materials, since they only go up to third level, but it is notable that not a single class gets any increase in base attack bonus over that period of time. There are a couple bonuses to damage here and there (presumably to compensate), but I think that will have much less of an effect on the way the game plays. Even the advantage and disadvantage mechanics are designed around avoiding bonus inflation by using a d20 version of 2DTH (roll two dice, take highest or lowest). For an example of how this is used, consider penalties for wearing armor. Armor causes disadvantage to several kinds of rolls (stealth, saving throws when not proficient, etc). I love this mechanic, so this makes me happy.

Gone will hopefully be the nonsense of 4E scaled defenses. Such scaling results in absurdities like a first level pixie having a 15 reflex and a 20th level ogre having a reflex of 30 (I made those numbers up, but they are representative). Some people have described these play test materials as “3E lite,” but I just don’t see it. Pretty much the only point of comparison is the listing of task DCs, and as far as I can tell the whole point of flattening the power curve is diametrically opposed to the reality of Third Edition, which is by far the edition with the steepest power curve (4E starts out more powerful, but advances less steeply).

I love the idea of ability scores as saving throws, despite the fact that it might fight against 3d6 in order (because ability scores will be so potent that players will always want them to be high). That said, the benefits are many, and include avoiding some of the skill tax problems of perception and allowing the simplicity of 3E saving throws without needing separate numbers on the character sheet. This also means, however, that saving throws will not get better as levels increase (one of the few aspects of level-based inflation that I like, because it allows players to earn survivability through smart play rather than optimization). Does this mean that ability scores themselves will inflate as levels progress? Or maybe there will be separate saving throw bonuses? We will have to wait for more details before that will be clear. Also, this means no evocative saving throw types like the save versus death ray.

The Caves of Chaos adventure is almost a primer to old school gaming. Specifically, it gives the following advice:

  • No single storyline
  • PC motivations are not assumed
  • There are no plot points and encounters are not ordered
  • Dynamic dungeons; PCs not just expected to clear the site
  • No balanced encounters
  • PCs may go where they want and pick their battles
  • Monsters are supposed to be played strategically
  • Make up monster ability scores if desired
In other words, this is pretty much the Caves of Chaos as imagined by Gary Gygax. There are several suggestions about how to avoid killing PCs if they are in over their heads, but even Gary had suggestions in the original about ransoming PCs. The only significant modification from the original is that DCs are given for avoiding hazards and finding things like secret doors. However, there is also language about how a player must indicate, for example, the correct place to search in order to have any chance at all, and that the referee can always award success based on good description with dice rolls. In other words, player skill. This might seem like schizophrenia, but one might also read it as a license to make the game your own.

There are still lots of important things missing. XP award guidelines, for example. From the materials, it seems like you only get XP for defeating monsters, and that will not encourage the kind of careful play that the original module expected. There are no mentions of retainers. There is not variable monster reaction system, which will probably lead referees to assume hostility. Given that this is not supposed to be a complete game, I am willing to withhold judgment about those omissions for now. Also, the solutions they propose to fill those gaps need not necessarily be clones of OD&D or B/X to be successful. But the function of those original systems needs to be understood.

Varying mortality and lethality

The current D&D Next play test rules for lethality and dying are way too soft for games that I would like to run. I want there to be a greater threat of death, for a number of reasons, but not least because I don’t want combat to be the first resort of PCs. Mike Mearls has already said that HP is likely to come down, but I don’t think that is enough. The dying rules also have to be addressed.

The thing that frustrates me about this discourse is that it is not an either/or proposition. It is easy to build several possible play styles into the core. First, you can always start at higher level. Some people object to this because higher level is also higher complexity, so you are really affecting more than one aspect of character design (though I’ll also note that a first level 4E character feels about as complicated to me as a 5th level traditional D&D character). Including options for different play styles does not take anything away from anyone else.

The various editions have already given us a plethora of death rules. All that is needed for 5E in this regard is for them to pick several possibilities that address different play styles and present them, along with pros and cons, in whatever becomes the referee guide. For lethality, three natural options jump out at me. For a lethal game, dead at 0 HP (perhaps with a constitution saving throw to be incapacitated rather than killed). Other options would be dead at negative 10 (or negative constitution) HP, and the current playtest version that is reminiscent of 4E (with all those fiddly death saving throws) could also be retained as an option.

Here is a proposal for supporting different levels of lethality. Vary starting hit dice. This has the advantage of not increasing complexity for other character aspects. Also, First Edition play falls out as a natural corollary (the 1 hit die variation). One die of self healing is also very close to common binding wounds and liquid courage old school house rules. Bump the starting hit dice up to, say, 4 or 5 and you will have a game that feels much closer to fourth edition; starting hit points will be greater, and PCs will have more hit dice available for spontaneous healing.

The other major aspect of lethality that is potentially problematic in Fifth Edition is the recovery provided by a long rest. In the play test rules, a long rest restores all HP and hit dice. This should be another part of the game with a menu of options. I would suggest recovery of one hit die for an old school feel (which would then need to be spent for any healing to occur), recovery of all hit dice for a less deadly but still random feel, and recovery of all hit dice and HP for a super-hero feel.