Spells Through The Ages – Spell Lists in B/X

Additions and Subtractions Over Time; Whence the OD&D Sup-I Spells?

While I personally got into D&D initially with the 1979 Holmes Basic D&D Set, I had a somewhat more well-resourced friend who later got the 1981 Moldvay Basic/Cook Expert Rules sets, and actually handed them off to me, so as to DM him and other friends through higher levels of play. (I was always the DM in our community since day zero, so this just seemed like the thing to do.) We played a lot with those rules for a bunch of years, and every time I go back to them I'm really impressed by the level of reflection and tasteful editing that Moldvay & Cook put into those rules. While Holmes was very connected-at-the-hip to AD&D (with the Holmes text being edited & altered by Gygax himself before printing), the Moldvay-Cook B/X intentionally set its own traditions (like race-as-class), a separately-evolving game through the 1980's. Let's take a comprehensive look at their magic-user spell lists here.

Moldvay Basic Rules

One thing that Moldvay does here is to set a tradition of exactly 12 spells in each level of magic-user spells. That's a little bit nice, since you know you can always roll a d12 for a random spell in any level (as a scroll, spellbook, or maybe an NPC wizard's memory); although it may be a bit of a handcuff in the design space. Recall that the OD&D had no such organizing principle, different levels had different numbers (and in distinction to AD&D, there were actually more at higher levels, not less).

In the LBB's, at there were only 8 spells at 1st level, and 10 at 2nd level. Therefore, to fill out his list, Moldvay had to include some of the canonical D&D spells which first appeared in D&D Supplement-I, Greyhawk -- at 1st level, magic missile, shield, and ventriloquism; at 2nd level, mirror image and web. Now, at first level, that still doesn't fill out his d12 roster, so Moldvay also had to go to AD&D and take the floating disc spell (a.k.a. Tenser's floating disc in those rules). Note that this is the only spell in the entire B/X line that didn't appear in either the LBBs or Sup-I. On the other hand, there was an overflow of OD&D Sup-I spells that didn't make it into Moldvay at second level: namely darkness, strength, magic mouth, and pyrotechnics (several of these being pretty canonical for D&D -- I recently saw fellow players quite surprised that strength and darkness were not available playing by these rules).

Now before I go on, let's compare this to the Holmes Basic D&D list. Holmes' original manuscript had lists that were identical to Sup-I (although alphabetized and unnumbered -- 11 spells at 1st level, 16 at 2nd level). However, Gygax got in after him and added several spells that otherwise only appear in AD&D -- dancing lights, enlargement, Tenser's floating disc (by its full name), audible glamer, and ray of enfeeblement. This then brought the total number in the published Holmes work to 14 at 1st level, 18 at 2nd level (more than in the later Moldvay rules). If you want to see a complete look at that development from Holmes' unpublished manuscript, see the recent Zenopus Archives Blog (spell levels one, two).

Cook Expert Rules

In Expert D&D Rules, Cook starts by repeating the Moldvay list, and then adds the spells you see above for levels 3-6. (Asterisks represent reversible spells, also added as appropriate to the 1-2 level lists). He continues with the design pattern of exactly 12 spells at each level. But here's a wrinkle: since the OD&D LBB's already had that many spells or more at each of these levels (14 at 3rd, 12 at 4th, 14 at 5th, 12 at 6th), he didn't need to copy any Sup-I spells into the Expert rules, and in fact even a number of spells in the original LBB's got cut out (clairaudience, slow at 3rd level; growth of animals, wall of iron at 5th level). All the higher-level Sup-I spells went entirely missing from these rules; and that's probably for the best, because they tend to be wonky, confusing, or under-powered anyway.

Later Editions

The soon-revised 1983 Frank Mentzer Basic/Expert Rules keep almost the same lists as Moldvay-Cook -- except that Mentzer cuts the spells at 5th & 6th levels down to just 8 (other levels have the same 12 as in B/X); maybe for space purposes in the book?

On the other hand,the 1991 Aaron Allston D&D Rules Cyclopedia makes the very strange choice of adding one spell per level to B/X, and thereby having a uniform 13 spells at each level (and thus not immediately rollable on any Platonic die). The spells he adds are, respectively by level 1-6: analyze, entangle, create air, clothform, dissolve, and stoneform. (He also has level 7-9 spells based on the Mentzer Companion/Master Sets, which I won't go into here, but likewise have 13 spells in each list).

Was there perhaps some intermediary product in the late 80's that add these "thirteenth" spells to Basic D&D along the way?

Wandering Expectations

A number of weeks ago, poster DHBoggs sent me a nice analysis of setting up dungeons in OD&D, including assessment of likely monsters encountered. I've looked at Wandering Monster Levels and Tables in the past, and he and I came up with basically the same numbers for average hit dice from each of the various wandering tables in OD&D Vol-3.

One thing that he and I agree on (I think I'm more of a latecomer to this observation) is that it doesn't quite make sense to equate "monster level tables" (tables 1-6 on Vol-3 p. 10-11) with "level of the monster" for number appearing purposes (which gets compared to dungeon level on p. 11). If you were to blindly do so, then at the deeper dungeon levels (like 13+), since the "monster level tables" only go up to 6, then you'd be committed to a situation where even the top-level monsters like Dragons, Balrogs, and Purple Worms would need numbers appearing in every encounter multiplied (or exponentiated?) by at least a factor of 7 or more, which becomes totally lunatic if you think about it. It makes much more sense in OD&D, for this purpose, to treat the "level of the monster" as equivalent to its hit dice -- just like it says in the example on experience in Vol-1, p. 18 regarding "a troll (which is a 7th level monster, as it has over 6 hit dice)."

Anyway, one thing this discussion brought to mind is that while we'd both computed the average hit dice (level) of monsters at each of the 6 monster level tables, no one had thought of computing the expected hit dice of monsters at each actual dungeon level. Consider the following tables (or see the open document spreadsheet ODS file here):

As you can see in the final column of numbers, the average random encounter on the 1st level of an OD&D dungeon actually features a monster with 2 Hit Dice. In the 2nd dungeon level, the monsters average 3HD. At 3rd level the average is 5HD, at 4th-5th it's 6HD, etc.

Similar to what we've seen before, the interpretation of these results is in OD&D at the lower levels, the average wandering monster is actually higher level than the dungeon level would indicate. (Later, AD&D tables over-correct and have distinctly weaker monsters, while something like Holmes or Moldvay Basic D&D is about in the middle.)

This explains why in the margin of my copy I subtract -1 from the die-roll for monster level table, such that the monsters more generally match the dungeon level on which they're appearing. At the deepest levels you'll still be doing some multiplications of monster numbers for creatures of around 9 or 10HD, although it's rarely necessary for 12HD Hydras or Dragons, and never done at all for 15HD Purple Worms, for example (which is certainly better than treating all of these types as uniformly "6th level").

Spells Through The Ages – Spells In Sup-I

I think I've said this once or twice at this point, but while the 1974 Original D&D LBB spell lists seem really solid (effective, inspired, coherent, etc.), the new spells that were added in 1976 Supplement-I Greyhawk usually seem wonky (confusing, complicated, under-powered, etc.). Less so at the lower levels, and more so at the higher levels.

In fact, even before getting my hands on the LBBs within the last decade, I always had a gut feeling that something was "weird" about the 7th-9th level spells -- while the 6th level list has the stark "powers over life, death, and eternity" (death spell, reincarnation, disintegrate, control weather, anti-magic shell, permanent geas and invisible stalker, etc.) the higher-level spells have often fiddly and complicated powers that are hard to see why they're better. Of course, looking at Sup-I answers the mystery: the LBBs contain just levels 1-6, and it was Sup-I that introduced all of levels 7-9 as tacked-on afterthoughts.

Let me look at some of those "new" Sup-I spells, mostly the ones in levels 1-6 (the ones I potentially play with) -- asking some pointed questions about whether anyone ever found use for them, or if they're basically firing "blanks". I won't look at the entire later history of these spells; generally they didn't change very much, perhaps due to the fact that they were under-powered and no one was breaking the game with them (so, little in the need for errata or fixes). Perhaps I can observe a few categories for these Sup-I spells as follows:

• Low-Level Spells that Are Clearly Powerful. This would include the addition of shield, magic missile, darkness, strength, web, and mirror image (all 1st or 2nd level). Several of these are considered must-haves for wizards of any level, in some cases contending to push the best attack/defense spells from the LBBs out of use (sleep and charm person). These all became canonical identifiers of D&D, of course, and it's really hard to imagine the game without these key additions.
• Low-Level Spells That Aren't So Great. This would be ventriloquism, magic mouth, and pyrotechnics (again 1st-2nd level). Ventriloquism at first doesn't even seem like it should require magic; maybe a thief or entertainer skill. I've never seen magic mouth used by a player; it's more of DM plot device (waits eternally to deliver some information). And pyrotechnics starts a Sup-I'ism in not having a single clear game effect, but instead a series of "maybes" that require interpretation at the table ("A multi-purpose spell... either a great display of flashing, fiery lights and colors which resemble fireworks; or he can cause a great amount of smoke... overall effects of this spell depend on the size of the fire used to cause them", none of which have in-game definitions or effects; p. 22). 
• Mid-Level Spells That Are Questionable. Namely explosive runes, rope trick, suggestion, and fear (3rd-4th level). Of these, fear seems to be the most useful (replicating the Fear Wand from LBB Vol-2, p. 34); the others seem strangely over-specific and limited. Suggestion I wrote about earlier (link); it seems deficient to charm person at 1st level. Explosive runes seems like another DM-protection element (it could just be an added one-liner in the scroll curse table); I've had players use this on a piece of paper and then parley with monsters to trick them into reading it, which seemed to cut against the grain thematically. Rope trick also seems to me an oddly over-specific representation of the (mythic?) "Indian rope trick" stage magic (link); I think I might prefer a more general "step into a small, short-term, extra-dimensional hiding spot" than the whole climb-up-a-levitating-rope business. 
• The Mid-Level Filler Series. These would be monster summoning and extension series. These spells reappear at multiple levels with a Roman-numeral tag on each in the sequence. The monster summoning spells (I-VII, levels 3-9) refer the user to the random "Monster Level Tables", which is useful, and mechanically concise since the element already exists. But I find it to be rather tasteless game design since I can't wrap my head around (a) how the monsters get into the possibly-enclosed space with the PCs, (b) why it totally disregards whether monsters of that type exist in the dungeon, (c) why the monsters just wander off afterward, etc. The extension spells (I-III, levels 4-6) may be flat-out the weakest magic ever; they only extend some other lower-level spell by 50% duration, and if it didn't do the job in the first couple turns, are the next few going to make any big difference? Case studies: (a) I'd rather use a duplicate 1st-spell and double the duration than burn a 4th-6th level spell for only 50%; (b) the only thing that extension II does better than I is to add specifically 4th-level spells to the possible targets; (c) extension III, while in the spell list, doesn't even get any entry in the text descriptions (okay, fixed in the corrections insert at the back of the book). It almost seems like these spells were just filler to round out the mid-level lists to 16 entries each.
• Higher-Level Spells That Don't Get Used. Here I'm thinking legend lore and repulsion (at the 6th level; compare to other "life or death" magic above at the same level). Legend lore "seeks to gain knowledge of some legendary item, place or person", requiring DM interpretations on (a) what counts as legendary, (b) exactly what knowledge is gained, and (c) whether it even works at all (note the word "seeks"). This spell seems like a Rorschach inkblot test for the players and DM, possibly the vaguest spell I know of in D&D; and it also has randomized casting time, from 1-100 days (the only spell in OD&D, of very few in any edition, with multi-day casting time). Repulsion "enables the user to cause objects or entities to move in a course opposite from their intended course towards him", which seems like a restatement of the confusion spell as written in Chainmail (there only 4th level), and seemingly defeated as soon as enemies think to say "I move away from the caster". I've never seen this spell get used in play, ever.
• Spells Whose Power Comes from Disallowing Saves. This kind of burns my chaps, because it seems like a clunky rules hack that totally contradicts Gygax's other rants that "everyone gets a chance" (see AD&D DMG defense of saving throws). It's inelegant in that later rule sets need to start adding a stat-block line dedicated to keeping track of whether each spell is "Yes" or "No" allowing a save. Personally, I don't honor it my games. These spells include (some overlap here): sleep (by errata in Sup-I), magic missile (not explicit in Sup-I, but ruled that way in later sets), explosive runes, ice storm, and everything in the new levels 7-9 (p. 25: "Spells with no saving throw unless otherwise indicated!"). In particular, ice storm (4th level) looks deficient to fireball (less range, only 30 points max and no increase per level, whereas fireball does 42 max damage for a like-level caster and goes up from there), until you note the  parenthetical point that "saving throws are not possible"; 1E added a variant sleety usage that caused no damage, but blindness and a possible slip-and-fall. Hacky!

I guess the only other thing in Sup-I at these spell levels that I'm looking at is the errata for the charm person and charm monster spells, actually the first thing in the list, which introduced the unique recurring save schedule by Intelligence or Hit Dice (and was carried forward through all later editions up to 3E). Even that seems unnecessarily complicated, table-heavy, and unfocused, wasteful design effort; the best version is in Moldvay's Basic D&D where he wisely edits it down to just three categories (day, week, or month). Lastly: detect magic is given a range and duration for the first time, but then why not also do so for read magic and read languages, that suffer the same lack at the start of the Vol-1 list?

So there's my personal rant about the Johnny-come-lately, frou-frou, fiddly magic-user spells and alterations that appeared in D&D Sup-I. While many of the spells at 1st-2nd level are must-haves, the ones at mid-levels are either vague or questionable or clearly worse than pre-existing options. And the new higher-level spells generally cheat and complicate the system by prohibiting any saving throws, so I don't like those, either.

What are your thoughts on those later-added spells; do you agree? Is there anything here that you found absolutely indispensable, to which I've been blind? Have any unusual interpretations that clearly "fixed" one of these spells for you?

3d6 Variable Parameters

Nothing new here, but I find myself continually re-computing this stuff so I'm documenting it here. Consider rolling a 3d6 probability variable, as for ability scores in OD&D, or starting money if we multiply by 10.

• Curve is bell-shaped.
• Average is 10.5 (mean or median).
• Quartiles are 8, 10.5, and 13 (i.e., about one-quarter are 8 or below, etc.)
• Standard deviation is almost exactly 3 (more specifically: 2.958)

Table & histogram follow:

Edit: My old website also has the 4d6-drop lowest curve.

Spells Through The Ages – Passwall

Here's another light offering for the "Spells Through the Ages" series, the 5th-level wizard spell passwall: you make a temporary hole in a wall to walk through -- simple, right? Well, yeah, pretty much. (Although one of the Endless Quest books interpreted it as phasing through the wall in a magic cloud -- maybe Pillars of Pentegarn?)

Original D&D

Pass-Wall: A spell which opens a hole in a solid rock wall, man-sized and up to 10' in length. Duration: 3 turns. Range: 3".

Opens a hole in a wall, 10' deep, "man-sized". Note the very short duration (this will steadily get longer over time). One material is affected: Rock.

D&D Expert Rules

Pass-Wall 
Range: 30'
Duration: 3 turns

This spell opens up a 5' diameter hole up to 10' deep in solid rock or stone. When the spell ends, the hole closes.

This version is effectively the same as the preceding. Two materials are affected: Rock or stone (synonyms?). Still 10' deep, as usual.

But consider the specific interpretation for width & height (formerly just "man-sized"); Cook makes it circular, 5' in diameter. And sitting above is the Jeff Dee art piece (see top of this post), further emphasizing the idea. It does give it nice organic look, although I'm not sure how comfortable it is to walk through a curved cylindrical tube; and stooped over to just 5' height. Yet this is my normal conceptualization of the spell, even though it's not shared by any other edition -- another example of the "Power of Pictures".

AD&D 1st Edition

Passwall (Alteration)
Level: 5
Range: 3"
Duration: 6 turns + 1 turn/level
Area of Effect: special
Components: V, S, M
Casting Time: 5 segments
Saving Throw: None

Explanation/Description: A passwall enables the spell caster to open a passage through wooden, plaster, or stone walls; thus he or she and any associates can simply walk through. The spell causes a 5' wide by 8' high by 10' deep opening. Note several of these spells will form a continuing passage so that very thick walls can be pierced. The material component of this spell is a pinch of sesame seeds.

Here, the base duration has doubled as compared to OD&D, and a caster-level addition included on top (so, for a minimum caster level of 9, the duration is 15 turns, quintuple the OD&D rule). Three materials are cited: Wood, plaster, and stone.

And for shape, we see that Gygax specifies a hole size of 5' × 8'. That's clearly not circular (as in Cook Expert). Maybe that's intended to be a sharp rectangle? Or alternatively: Note that it's close to exactly half the stated dimensions of the "Standard Corridors" in the Temple of Elemental Evil, which we might take as the default Gygaxian corridor size (p. 43: "The 10' wide corridors have gothic arches, peaking at about 17' height"). So perhaps it's intended as the same gothic arch shape, which seems more aesthetically pleasing?

AD&D 2nd Edition

Passwall
(Alteration)
Range: 30 yds.
Duration: 1 hr. + 1 turn/level
Area of Effect: 5 x 8 x 10 ft.  

A passwall spell enables the spellcaster to open a passage through wooden, plaster, or stone walls, but not other materials. The spellcaster and any associates can simply walk through. The spell causes a 5-foot wide x 8-foot high x 10-foot deep opening. Several of these spells can form a continuing passage so that very thick walls can be pierced. If dispelled, the passwall closes away from the dispelling caster, ejecting those in the passage. The material component of this spell is a pinch of sesame seeds.

Same thing as 1E, although "safety bumper" language has been included in case of a dispel while someone is walking through. Identical wood, plaster, and stone target materials. Still a 5' × 8' (gothic?) doorway 10' deep.

D&D 3rd Edition

Passwall
Transmutation
Level: Sor/Wiz 5
Components: V, S, M
Casting Time: 1 action
Range: Close (25 ft. + 5 ft./2 levels)
Effect: 5 ft. x 8 ft. opening, 1 ft./level deep
Duration: 1 hour/level (D)
Saving Throw: None
Spell Resistance: No

The character creates a passage through wooden, plaster, or stone walls, but not through metal or other harder materials. If the wall’s thickness is more than 1 foot per caster level, then a single passwall simply makes a niche or short tunnel. Several passwall spells can then form a continuing passage to breach very thick walls. When passwall ends, creatures within the passage are ejected out the nearest exit. If someone dispels the passwall or the character dismisses it, creatures in the passage are ejected out the far exit if there is one or out the sole exit if there is only one.

Mostly the same as the 1E-2E lineage; the hole is again the 5 by 8 ft. dimension. The affected materials are still the "wooden, plaster, or stone" stipulated in 1E (although the fact that other "harder" materials are explicitly prohibited, means there may now be some gray area between the two classes). Duration has been increased and normalized in these rules to a lengthy 1 hour/level. But...

The depth has changed from a simple 10' to a variable 1 ft./level. WHAT THE HELL 3E? Is that plus-or-minus a foot ever going to be so important as to justify that change? Are you really going to deny players getting through a wall because the wizard is only 9th level when they first pick up this spell? A dumb complication.

D&D Next Battlesystem

Just saw this post, from the end of April, from Mike Mearls about the planned Battlesystem for the D&D Next game, title "The Art of War". It's interesting that while D&D 3.5/4E definitely went in a direction downplaying tactical miniature battles, it appears that this version is more like the original wargame experience. In fact, in some ways it seems a bit eerie how much parts sound like my own Book of War game (see sidebar).

It uses the stats for monsters and characters mostly unchanged. If you know the combat rules for D&D, you're already 90 percent familiar with how Battlesystem works.

As I wrote in 2011 for section on conversions (p. 10), this was also one of the guiding principles for Book of War (prior games from TSR all had quite different mechanics & statistics; see all of  Chainmail, Swords & Spells, and Battlesystem; contrast with the Book of War core rules here).

The big changes in the rules focus on scale. Large battles naturally take more time than single combat, so a round of combat in a Battlesystem mass battle takes 1 minute. Battlesystem uses a combat grid divided into squares measuring 20 feet on each side, and scales up the number of creatures a single miniature represents. 

Note that Book of War is also at 1" = 20 feet, whereas all the earlier games from TSR were at 1" = 30 feet (as above, see assessment that led to the Book of War scale here).

If a stand attacks a solo creature, the stand takes one attack for each creature in it. A solo can avoid such potentially devastating attacks by joining up with an adjacent friendly stand, relying on the creatures in that stand to protect it. The enemy stand can still attack the solo, but it makes only one attack.

This "devastating attacks" statistical observation re: solo vs. mass was incorporated into Book of War, and in some circles was the most controversial part of the game; the prior Battlesystem from TSR had explicitly inflated solo stats to make them super-durable on the battlefield (while fantasy Chainmail was man-to-man only, and the implications in Swords & Spells were obscured; more discussion here).


Of course, I agree with Mearls that this kind of approach makes for a much more satisfying add-on game to D&D, and I'm glad to read that Wizards have pivoted away from the 3.5/4E approach in the last several years. Coincidentally, I've had an unusually large number of friends staying over at our place for the past month, and we've actually been playing a lot of Book of War lately. It's great to see how excited folks are by the game when first introduced to it, and then perhaps to the closely associated Original D&D game as well. I'll probably have some battle reports coming up from those games -- and if you want to check it out, of course, Book of War is available right now in the sidebar and here at Lulu.

Spells Through The Ages – Swords & Spells

You may have seen me thank Zenopus Archives a few times in recent weeks, after he helpfully reminded how useful it can be to look at Gary Gygax's Swords & Spells miniature rules for late-era OD&D, which included a complete Spell Chart for almost every spell in the game (a comprehensive list of Range, Area Effect, Turn Duration for each spell; p. 12-15). In many cases this filled-in previous blanks or made changes to the characteristics of certain spells. For reference, commentary, and scholarship, here is that list for the magic-user spells (my copy is well-worn and a bit yellowed, so you get the full effect below):

Here are some of the things we can observe:

• First, the listing is nearly identical to the spell list format found in D&D Supplement-I, Greyhawk. Sup-I additions are included, in the same order (not yet alphabetized even within a spell level), etc.
• The list includes every magic-user spell in the game except the following, presumably not useful in a mass battle: read magic, read languages, telekinesis, and legend lore (plus at level 7+: simulacrum and clone).
• The only name change here is that dispell magic is now dictionary-corrected to dispel magic. Otherwise, all names are identical, including: phantasmal forces (plural), fire ball (with a space), etc.
• The order of water breathing and explosive runes is switched in the list (possibly a transcription accident being corrected?)
• Many spells list an Area Effect of "personal", both for spells that must be cast on others (strength, suggestion, geas, reincarnation, etc.), and spells that are clearly caster-only (polymorph self, wizard eye, magic jar, contact higher plane, etc.). Therefore it's unclear in other cases which way the intent was here (fly, levitate, teleport, etc.). 
• Under Duration, Gygax uses a dash ("-") both for instantaneous spells (fire ball, lightning bolt, teleport, etc.) and spells with special long-lasting durations (the charms). He specifies "until dispelled" both for spells limited in the text by concentration and those that are really permanent (making wall of fire and wall of ice look identical here when they're not in Vol-1).
• Finally: sleep gets a duration for the first time here (4-16 turns), clairaudience/clairvoyance get reduced (from 12 turns to 6), and some minor duration changes are made to detect evil, ventriloquism, and ESP (possibly typos; these are not re-used in any later editions such as Holmes, B/X, etc.)

Anything else you can see here?

Arena – Man vs. Man

Battling Random D&D Fighters in a Simulated Arena, So As to Assess Average Advancement and Ability Scores at High Levels

Problem

It's always been clear that higher-level characters in D&D should have higher ability scores; thinking statistically, seeing an NPC at higher level provides evidence that they likely have better-than-average ability scores that allowed them to survive and gain levels. But how much higher? This has long been hand-waved in D&D, with the publishers and players simply making something up on the fly that felt right, with little consistency. Obviously none of us has ever archived the play history of enough PCs to assess what their ability scores were, on average, at a given level.

Method

Let's simulate characters battling and gaining levels in a computer program; the one I've written in Java is called "Arena", broadly modeling gladiators battling to the death in a Roman-style Colosseum. The model runs like this: First, generate 10,000 1st-level basic D&D fighters with random ability scores (3d6 in order; bonuses OED-style +1 for 13-15, +2 for 16-18). Equip everyone with normal chain, shield, sword, and helmet. Pair them up randomly and fight (all fights are to the death at 0 hp). Award experience pro-rated for standard D&D treasure awards (for example: at 1st level, multiply base experience by 20, because about 95% of XP at the level generally comes from treasure; link); level-up if appropriate. Replace the dead part of the population with 5,000 new random 1st-level fighters and repeat.

Note that this is ever-so-vaguely scaled to the overall Roman gladiatorial system. Historical analyses suggest that a professional given fighter might battle about 3-4 times a year (i.e., once per season; certainly healed up between fights), for a fairly high monetary reward, and that about 8,000 deaths occurred in arenas, per year, throughout the Roman Empire (link). My guess is that about 100 "cycles" of this action (one combat quarterly) would be approximately equivalent to the maximum gladiator career of about 25 years. Yet for broader data I've let the simulator run for 1,000 cycles below and not bothered to "age out" anyone in the system. Obviously, D&D characters engage in far more combats than anyone in real life.

Results

Here are the average results after a run of 1,000 cycles of randomized, paired combat:

What you can see here is that the vast majority (96.6%) of the gladiatorial population will remain at 1st level over any amount of time. The largest single bump in ability scores is actually between 1st and 2nd level; basically, a starting character needs a special "something" (like high ability scores) to survive to 2nd level, and thereafter can mostly "cruise" by consuming more and more 1st level characters that they're paired against.

The ability scores that make the biggest difference for these fighters seem to be (in order) Strength, Constitution, and Dexterity. Strength average is 14 at 2nd level, and then around 15 for levels 3+; Constitution average is 13 at 2nd level, and higher by one or two points thereafter; Dexterity average is around 12-13 for any level above 1st. Since the combat here is entirely melee (sword vs. shield; no ranged combat), the Strength ability is highlighted. And since the Intelligence, Wisdom, and Charisma scores have no effect in this action, they remain around the natural 10.5 average at any level.

But another thing: The characters don't just benefit from high ability scores, they also benefit from higher than expected rolls for hit points. For example: You might expect that a 2nd level fighter with 13 Con would have around 11 hit points (2(d8+1) ~ 2(5.5) = 11); but in fact the average is 13 points (i.e., about 20% higher). At 8th level the naive expectation would be 44 hit points (8(d8+1) ~ 8(5.5) = 44), but the average seen above is actually 56 hit points (almost 30% higher). Recall that 3E D&D, for example, actually did stipulate that NPC's at advanced levels had the naive expectation for hit points, which is not actually correct according to this assessment.

Another thing we might look at is the demographic curve at advanced levels and regress to some predictive formula; here it is below (x-axis is level 2+; y-axis is in number per 10,000 gladiator population):

Note that the preceding was based on purely random pairwise combats between all 10,000 members of the population. An alternative that I tried was to first sort the population by level and have the 1st person fight the 2nd person, etc. (thus recreating a "championship" bout, where the two best characters pair off, and a bit more like D&D adventuring where it's expected that PCs fight monsters of about the same level). But the result of that was quite comical; one very strong character would initially graduate from 1st level, and thereafter advance by perpetually massacring everyone else who ever managed to pop into 2nd level. After 1,000 cycles this usually results in something like 9,990 fighters at 1st level; 9 fighters at 2nd level; and 1 fighter at 8th level (with scores like Str 16, Dex 15, Con 17, hp 62).

Conclusions

These results begin to suggest what we might do for reasonable and consistent ability scores at higher levels. As a base we roll 3d6 for each ability score, and then something like:

• 2nd level: Prime ability +3, secondary +2, tertiary +1 (total +6).
• 3rd-4th level: Prime ability +4, secondary +2, tertiary +1 (total +7).
• 5th-7th level: Prime ability +4, secondary +3, tertiary +2 (total +9).
• 8th+ level: Prime ability +4, secondary +4, tertiary +3 (total +11).

(Edit: Or actually, don't  do that, because adding a flat bonus like 3d6+4 gives a range of 7 to 22, and no high-level fighter is going to have values in either the lower or upper end of that interval; the variance is all off. Instead do something like roll 3d6 for the good abilities and apply a minimum of 4 on any single die, or re-roll values below that.)

Also, at level 2+ we should boost hit points by +20% to +30% over the naive mathematical expectation, as mentioned above. Or use the somewhat common mechanic: re-roll any "1"'s or "2"'s on hit dice. (Edit: Or apply a minimum of at least half any die roll, e.g., minimum 5 on any fighter's d8.)

Personally, my preference is not to give any special modifiers to PCs at 1st level (in ability scores, bonus or minimum hit points, etc.). But if we want PCs to automatically be "special", then I'd be very willing to start them at level 2+, assume to some extent that they've already proven themselves exceptional, and grant the aforementioned bonuses as representative (and helpful to the players). 

Here's some ways we might expand on this work. Obviously, the current code only pairs off fighters-vs-fighters, and only those relevant ability scores are selected for advancement. I wouldn't hope to simulate wizard-vs.-wizard battles any time soon. But we might run a grab-and-dash simulation for a Thieves' Guild, where an initial skill roll (pick pockets or open locks), if failed, devolves to a melee fight or flight. Or we could switch from an arena of all NPC fighters to fighters-vs-monsters according to the OD&D wandering monster table and see how that goes (monsters without special abilities, at least initially). Or I suppose we could go more "realistic" and find a way to award XP in our arena without death to the opponent in every case, such that more people could survive to higher levels.

What do you think? Any other observations or suggestions?

Appendix

Here's the code to the Arena simulator (Java ZIP file).

Spells Through The Ages – Suggestion

Here's one of the more rarely-used pieces of D&D magic: the 3rd-level wizard spell suggestion. In my recent assessment of long-lasting D&D spells (link), it's the only spell that I would recommend lengthening the duration (from a traditional 1 week to 1 month). Let's take a closer look.

Original D&D

Suggestion: A spell which works on the principle of hypnosis. If the creature which it is thrown at fails to make its saving throw vs. magic it will carry out the suggestion, immediately or deferred according to the wish of the magic-user. Self-destruction is 99% unlikely, but carefully worded suggestions can, at the referee’s option, alter this probability. Suggestions must be simple and relatively short, i.e. a sentence or two. Duration: 1 game week.

Now, suggestion isn't in the LBB's (Little Brown Books in the first-ever D&D boxed set). This spell and the text above were added in Original D&D Sup-I Greyhawk (by Gyagx & Kuntz). I find this new 3rd-level spell hard to parse, considering it just gives one short direction that lasts maybe a week -- especially when you've got charm person at 1st-level, allowing one to continually order the victim around, and lasting arbitrarily long (two weeks before an average-Intelligence victim gets a new save, as shown two pages prior in the same book). In what way is this spell stronger? Is it that the suggestion has potency even in the absence of the caster? Was it simply a glitch that the rules for the two spells were developed in parallel, not knowing how powerful the other was intended to be?

Generally speaking, I find most of the spells added in Sup-I to be puzzling, or hard to make sense of, or weaker than spells in the LBBs, or more complicated or just kind of queer. So suggestion is just one example of that. In fact, the very addition of spell levels 7 to 9 (also in Sup-I) I basically disagree with, and find those spells to be frequently weaker than the original level 6 spells (control weather, death spell, reincarnate, etc.)

Dave Cook left this spell out of his D&D Expert Rules (perhaps wisely), so now we'll just proceed directly to the AD&D line.

AD&D 1st Ed.

Suggestion (Enchantment/Charm)
Level: 3
Range: 3"
Duration: 6 turns + 6 turns/level
Area of Effect: One creature
 
Explanation/Description: When this spell is cast by the magic-user, he or she influences the actions of the chosen recipient by utterance of a few words - phrases, or a sentence or two - suggesting a course of action desirable to the spell caster. The creature to be influenced must, of course, be able to understand the magic-user’s suggestion, i.e., it must be spoken in a language which the spell recipient understands. The suggestion must be worded in such a manner as to make the action sound reasonable; a request asking the creature to stab itself, throw itself onto a spear, immolate itself, or do some other obviously harmful act will automatically negate the effect of the spell. However, a suggestion that a pool of acid was actually pure water, and a quick dip would be refreshing, is another matter; or the urging that a cessation of attack upon the magic-user‘s party would benefit a red dragon, for the group could loot a rich treasure elsewhere through co-operative action, is likewise a reasonable use of the spell’s power. The course of action of a suggestion can continue in effect for a considerable duration, such as in the case of the red dragon mentioned above. If the recipient creature makes its saving throw, the spell has no effect. Note that a very reasonable suggestion will cause the saving throw to be made at a penalty (such as -1, -2, etc.) at the discretion of your Dungeon Master. Undead are not subject to suggestion. The material components of this spell are a snake’s tongue and either a bit of honeycomb or a drop of sweet oil.

Basically the same spell here with some additional verbiage. A small modifier is given for "a very reasonable suggestion" (-1 or -2 to save), and an increased chance seems to be given for getting a creature to kill itself through trickery (i.e., hallucinating that acid is bathwater). Although the spell text says that is has "a considerable duration", the time the spell lasts has actually been radically decreased from 1 game week (in OD&D Sup-I) to only 6 turns + 6 turns/level (that is, by the written rules here, 1 hour/level + 1 more). The charm person spell still retains its arbitrarily long schedule of saves (in units of days, weeks, or months) , so if anything it's even harder to see how that 1st-level spell is not more useful than suggestion.

AD&D 2nd Ed.

Suggestion
(Enchantment/Charm)
Range: 30 yds.
Duration: 1 hr. + 1 hr./level
Area of Effect: 1 creature 

When this spell is cast by the wizard, he influences the actions of the chosen recipient by the utterance of a few words--phrases or a sentence or two--suggesting a course of action desirable to the spellcaster. The creature to be influenced must, of course, be able to understand the wizard's suggestion--it must be spoken in a language that the spell recipient understands.

The suggestion must be worded in such a manner as to make the action sound reasonable; asking the creature to stab itself, throw itself onto a spear, immolate itself, or do some other obviously harmful act automatically negates the effect of the spell. However, a suggestion that a pool of acid was actually pure water and that a quick dip would be refreshing is another matter. Urging a red dragon to stop attacking the wizard's party so that the dragon and party could jointly loot a rich treasure elsewhere is likewise a reasonable use of the spell's power.

The course of action of a suggestion can continue in effect for a considerable duration, such as in the case of the red dragon mentioned above. Conditions that will trigger a special action can also be specified; if the condition is not met before the spell expires, the action will not be performed. If the target successfully rolls its saving throw, the spell has no effect. Note that a very reasonable suggestion causes the saving throw to be made with a penalty (such as -1, -2, etc.) at the discretion of the DM. Undead are not subject to suggestion.

The material components of this spell are a snake's tongue and either a bit of honeycomb or a drop of sweet oil.

I think this is the usual copy-and-paste job from 1E to 2E; I can't see any functional differences at all. On to 3E...

D&D 3rd Ed.

Suggestion
Enchantment (Compulsion)
[Mind-Affecting, Language-Dependent]
Level: Brd 2, Sor/Wiz 3
Components: V, M
Casting Time: 1 action
Range: Close (25 ft. + 5 ft./2 levels)
Target: One living creature
Duration: 1 hour/level or until completed
Saving Throw: Will negates
Spell Resistance: Yes

The character influences the actions of the enchanted creature by suggesting a course of activity (limited to a sentence or two). The suggestion must be worded in such a manner as to make the activity sound reasonable.

The suggested course of activity can continue for the entire duration. If the suggested activity can be completed in a shorter time, the spell ends when the subject finishes what he was asked to do. The character can instead specify conditions that will trigger a special activity during the duration. If the condition is not met before the spell expires, the activity is not performed.

A very reasonable suggestion causes the save to be made with a penalty (such as –1, –2, etc.) at the discretion of the DM.

Again, this looks the same as the other AD&D spell versions. Duration has been justified to simply 1 hour/level (no bonus extra hour). But at least in this ruleset charm person has been reduced to the same 1 hour/level duration, so suggestion is not obviously deficient in that particular statistic. But I still can't see any advantage to justify it being two spell levels higher.


Personally, I think what I'm going to house-rule in my games is that charm person lasts 1 day, and suggestion lasts 1 month (again, the only extension to spell duration that I make), so there's a clear long-term advantage to this latter spell. But what do you think? Is there some explicit functional advantage to suggestion that I'm missing out on? Or was the spell simply a waste in Sup-I and the rest of classic D&D?