Go to Part 1

There are literally thousands of RPGs. I, personally, own hundreds of them. So, obviously, I’m not going to attempt to catalogue every single stat block that has ever existed, but I think it might be useful to take a step back from D&D’s paradigm and look at how other games have approached this problem.

Sufficiently simple games, of course, can largely sidestep it entirely. It’s not unusual to find an RPG where NPCs are entirely defined by just two or three numbers. In some player-facing systems, many NPCs may not have any stats at all!

Although it can be somewhat obfuscated, the 1974 edition of D&D more or less fell into this paradigm. Most combat stats were derived from Hit Dice, and so for most monsters the only relevant stats were:

• Armor Class
• Move
• Hit Dice

Even in Monsters & Treasure (the 1974 version of the Monster Manual), this allowed monsters to be presented in purely tabular format:

In practice, special abilities and exceptions in each monster’s description complicate the placid simplicity projected by this table for many creatures, but this is an approach which has been more coherently synthesized in recent games.

In Over the Edge 3rd Edition, for example, Jonathan Tweet, another co-designer of D&D 3^rd Edition, created a system in which NPCs can be given a complete stat block with a single number:

• Sub-Par (0)
• Competent (1)
• Expert (2)
• Elite (3)
• World Class (4)
• Superhuman (5)

Important NPCs, however, can be given additional specialties and abilities that operate at a higher tier:

MADELINE VORE

Business Leader, Elite (3^rd): Madeline has an easy and firm grasp on how to create and leverage brand for maximum profit and how to run the enterprise that does so.

Energy Vampirism, Expert (2^nd): With a touch, Madeline can drain the life energy of a human and save it for herself, extending her life and restoring vigor.

DEFAULT, COMPETENT (1^st)

Monte Cook’s Numenera is another game that takes this approach, with stat blocks like this:

Warlord: level 4

Reugar Darkglow: level 3, tasks related to metalworking as level 5

Nieten: level 4, attacks with stronglass sword as level 5; health 20; Armor 1

Dread Rider: level 4, Might defense as level 6; health 15; damage 6 (4 blade, 2 electricity); Armor 3
Lightning Storm: Immune to electricity and +2 electricity damage. Others in immediate range suffer 1 damage per round.
Recall: Teleport to nearest dread destroyer.
Summon: Summon dread destroyer (arrives within a few hours).
Death Trigger: If slain, dread destroyer is summoned.

I refer to these as fractal NPCs: A simple default, but with the ability to add as much complexity, specificity, and crunch as desired when the occasion calls for it.

At the other end of the spectrum, of course, we have crunchier systems where NPCs have a ton of detail baked in by default. There are a lot of reasons why a system might want (and benefit from) this crunch, but a very common are systems where NPCs and PCs are built using the same rules. (And, therefore, all the complexity of creating a PC is inherited by the NPC stats.)

Like D&D, it’s not unusual for these crunchier systems to simply give up on the idea of a concise stat block. In The Succubus Club, for example, Vampire: The Masquerade just threw its hands up and printed the NPCs using full-blown PC character sheets:

It’s more typical, however, for crunchier systems to still try to achieve some degree of compactness, as seen in this stat block from GURPS Fantasy Adventures:

This relative compactness is often achieved by relying on some degree of system mastery: The spells, advantages, and disadvantages are all simple lists, requiring the GM to look up the rules for each individual feature. This is incredibly effective if the GM is familiar enough with the rules to know what, for example, Magery 2 means without looking it up, but brutally punishing to a new GM who has to look up a dozen or more abilities just to figure out what the NPC’s options are.

To square the difference, the designer can attempt to anticipate which abilities the GM is least likely to be familiar with and/or which abilities are relevant to the current encounter and provide a more detailed reference only for those abilities. For example, Literacy, Charisma +2, Compulsive Liar, and Greedy are probably all relatively self-explanatory, so maybe we can just focus on explaining what Voice and Magery 2 mean.

This is what I refer to as a hierarchy of reference. If you’re a GM doing this for yourself, it can be time-consuming but relatively straightforward (and very useful!). After all, you know exactly what information you do and don’t know. If you’re doing it for a published book, on the other hand, it’s often far trickier. You can’t even really target a specific level of system mastery, because every GM will learn different parts of the system in different ways and at different times. It can even be deceptive. (Compulsive Liar and Greedy look like roleplaying prompts, particularly if no further explanation is given. Did you remember to make the Will rolls they require?)

The other technique, of course, is to find ways to condense the presentation of the information. We saw variations of this earlier with the AD&D and BECMI D&D stat blocks. When Wizards of the Coast redesigned the D&D 3^rd Edition stat block in 2006, I thought their utility-based division was nifty, but the specific execution too large, so I revised the stat block into a short form that required a fraction of the space:

NAME (CR #) – [Gender] [Race] – [Class] [Level] – [Alignment] [Size] [Type]
DETECTION – [special], Listen +#, Spot +#; Init +#; Aura …; Languages [list], [special]
DEFENSES – AC #, touch #, flat-footed #; hp # (HD); Miss #%; DR #; Immune …; Resist …; Weakness …
ACTIONS – Spd # ft.; Melee attack +# (damage); Ranged attack +# (damage); Space # ft.; Reach # ft.; Base Atk +#; Grapple +#; Atk Options …; SA …; Combat Feats …; Combat Gear …
SQ …
STR #, DEX #, CON #, INT #, WIS #, CHA #
FORT +#, REF +#, WILL +#;
FEATS:
SKILLS:
POSSESSIONS:

Which could be combined with short references for pertinent abilities (and, when in doubt, page references for quickly looking up for the full rules when necessary).

This approach of condensation also requires a certain degree of mastery, but this time only of how the relatively dense stat block is designed to be read. (As opposed to every unique ability in the entire game.)

And, of course, these two techniques for dealing with crunchy stat blocks can also be used in combination — both compacting the information and also making decisions about which abilities need to be fully described.

IN CONCLUSION

So when it comes to stat blocks, there are broadly two approaches you can take.

First, you can choose to design your system to make the presentation of adventure stat blocks more compact, which often also has the effect of making them easier for GMs to create, particularly on-the-fly when something unexpected happens in the middle of a session.

Or you can try to figure out some way to condense the presentation of information, often relying (to at least some extent) on the GM’s system mastery to help carry the load.

Or you can just give up and tell people to look it up. If you’re feeling really apathetic, you can do that without even the common courtesy of a page reference.

This article is a revision and expansion of an older article. The original version can be found here.

Go to Part 1

This brings us to 3^rd Edition in 2000. In this new edition, every monster was made as customizable as a PC and their stat blocks ballooned as a result. For example, from The Sunless Citadel:

Jot: Quasit; CR 3; Tiny outsider (chaotic, evil); HD 3d8; hp 18; Init +3 (Dex); Spd 20 ft., fly 50 ft. (perfect); AC 18; Atk +8/+8/+3 melee (1d3-1 and poison, 2 claws; 1d4-1, bite); Face/Reach 2 1/2 ft. by 2 1/2 ft./0 ft.; SA Poison; SQ Spell-like abilities, damage reduction 5/silver, poison immunity, fire resistance 20, alternate form, regeneration 1 (normal damage from acid and holy or blessed weapon); SR 5; AL CE; SV Fort +3, Ref +6, Will +4; Str 8, Dex 17, Con 10, Int 10, Wis 12, Cha 10.

Skills and Feats: Hide +15, Listen +6, Move Silently +6, Search +4, Spellcraft +4, Spot +6; Weapon Finesse (bite, claw)

Special Attacks: Poison: Claw, Fort save resists (DC 13), initial damage 1d6 Dex, secondary damage 2d4 Dex

Special Qualities: Spell-like abilities: At will, as 6th-level sorcerer (save DCs, where applicable, are 10 + spell level), detect good, detect magic, and invisibility; 1/day cause fear (as spell, except its area is a 30-ft. radius from quasit), 1/week commune (six questions) as 12th-level cleric; speak telepathically to any creature within 100 ft.; Alternate form: polymorph self to one or two Medium-size forms.

BACK TO BASIC PRINCIPLES

These early 3^rd Edition stat blocks were, frankly speaking, terrible. This was primarily the result of four things:

1. Minimizing the amount of space the stat block takes up. This is legitimately motivated by the desire to make sure the information stays relatively compact. The entire point of standardizing things into a stat block, after all, is to move away from the early, casual style of describing stats narratively, and you want to keep the stat block as small as possible so that you have more room for the rest of the content in an adventure.
2. An early failure to prioritize key information. This didn’t matter in early stat blocks because they were only trying to code roughly a dozen pieces of information. When there’s only a handful of entries in the stat block, it’s easy to find anything you’re looking for pretty much instantaneously. By the time 2^nd Edition arrived, however, the stat block was commonly trying to code twice as much information. In 3^rd Edition, the information had more than quadrupled. And the amount of information was increasing because of…
3. Including all of a monster’s stats. This is obviously advantageous because you’ll avoid needing to look up information somewhere else. Replacing a chart look-up with THAC0 is one example of this, but you can see a general trend by simply glancing through the stat blocks we’ve looked at.
4. A failure to minimize the presentation of certain information. For example, is it necessary to include the “SV” abbreviation to prelude the saving throws? Probably not. Another example would be “speak telepathically to any creature within 100 ft.” Couldn’t you just write “telepathy 100 ft.”? As with the 2^nd Edition stat block, this is being complicated by another trend dating back to 1977: A desire in the advanced rules to codify effects as precisely as possible to eliminate any doubt, confusion, or interpretation and create a “standard” version of the game. This combines poorly with a simultaneous desire to “include all the info you need in the stat block.”

Regardless, what you had was a stat block that was very difficult to use and very easy to make mistakes with. So in July 2006, Wizards of the Coast debuted a new stat block format that was focused overwhelmingly on making it as clear and easy to use NPCs and monsters as possible. To this end it stuck with the commitment to include all the information needed to run the encounter in the stat block, and further split the stat block into five utility-based sections:

• Section 1: The information needed to begin an encounter. (What is the monster? How does it detect the PCs? Will the PCs be able to speak with it? What’s its initiative? And so forth.)
• Section 2: The information you’ll need to know on the PCs’ turn. (What’s its AC? Hit points? Saving throws? Resistances and immunities? And so forth.)
• Section 3: The information you’ll need on the monster’s turn. (What can it do? What attack options does it have? What special actions can it take?)
• Section 4: The information you don’t need to know during combat. (Or, at least, generally won’t need to know.)
• Section 5: Explanatory text. If an unusual ability is mentioned in the first four sections, its full explanation is given at the bottom of the stat block.

There was just one thing, though…

It was HUGE!

Wizards, it should be noted, wasn’t alone in this. That same year, Monte Cook, the co-designer of 3^rd Edition, introduced a revised stat block in his Ptolus sourcebook. It used a similar amount of white space, but chose to focus on separating out “critical” information. A couple years later, Paizo would further tweak the schema for Pathfinder, keeping the key concept of dividing the stat block into multiple sections, but specifically labeling the sections and integrating a section describing tactics right into the middle of the stat block (Defense, Offense, Tactics, Statistics).

What strikes me as interesting about all of these efforts, however, is the degree to which they effectively concede the battle. All of these revised stat blocks were essentially indistinguishable from entries in the Monster Manual. In other words, by 2006, D&D had basically given up on the idea of the “adventure stat block” a concise summary and were basically just copy-pasting bestiary entries.

D&D 4^th Edition (2008) was a radically different game from any other edition of D&D, but its stat blocks were just as chonky and continued to be duplicated directly into the adventures:

The problem was that these stat blocks were so large that they began to warp space around themselves. An encounter with two or three different stat blocks would chew up page space, creating an incredibly difficult or even impossible problem for layout to solve while flowing the text. Which, in turn, meant that DMs would have difficulty parsing the adventure.

The stat blocks were easy to use (everything you need, right at your fingertips!), but they were making everything else more difficult.

So the designers at Wizards of the Coast began looking for a solution to this new problem.

First, they experimented with putting all the stat blocks for the adventure in the back of the book. That way they wouldn’t interrupt the flow of the adventure, but they’d still be conveniently gathered together in one place for the DM. (Since even the early 3^rd Edition stat blocks could get quite large, they’d started doing this with some of their adventures even before the 2006 revision.)

But what they eventually settled on was the Delve format:

The basic concept was that every encounter would be presented as a two-page spread, allowing the stat blocks to be presented without, at least theoretically, disrupting the flow of the text. The presentation varied a little from one adventure to the next, the Delve encounters were generally presented at the end of the adventure (or, in some cases, as a separate pamphlet). The idea was that the main text of the adventure would say something like, “Now run Encounter L1,” and you’d flip the to appropriate Delve encounter.

This was another fascinating capitulation: First the stat blocks had gotten too large, so they’d been banished to the back of the book. Now the encounters were too large, so they were banished, too.

The selling point of the Delve format was, once again, that everything you needed was right on the page! It had never been easier to run an encounter! In actual practice, it was an utter disaster:

• The two-spread requirement was a classic “you gotta fill the space!” trap for designers, encouraging lots of empty verbiage that offered little or no value to the DM running the adventure.
• Conversely, if you did have an encounter that needed a lot of space to do properly, the format would force you to cram it into the limited space.
• It encouraged My Precious Encounter™ design, resulting in railroaded scenarios filled with monsters politely waiting around for the PCs to show up. (Part of this was a separate design ethos of trying to program encounters so that DMs could be reduced to dumb machines.)
• No matter where you actually put the Delve encounters in the book, they remained fundamentally bloated, chewing up space and drastically reducing the amount of playable content per page in scenarios using it. (This became painfully apparent in the few cases where Wizards of the Coast attempted to release pre-Delve scenarios in the Delve format: Often they’d be presented in books with two or three times as many pages, but nevertheless be harshly abridged to a fraction of their original content.)

When 4^th Edition finished the painful process of crashing and burning in 2011, a strategic decision was made to make a firm break with its failure for the next edition of the game. Among the things summarily tossed on the rubbish bin was the Delve format.

5th EDITION

From 2012 to 2014, D&D was effectively out of print. Instead, a very public playtest called D&D Next was conducted, using a combination of PDF rules packets and published adventure books. These adventures featured some light experimentation with the structure used for location keys, but as far as monster stats were concerned, they all used the same format which remains the standard approach for 5^th Edition:

• Stat blocks that do not appear in the Monster Manual are printed in an appendix at the back of the adventure.
• NPCs and monsters with stat blocks (with in the Monster Manual or the appendix) are listed in bold

Sometimes it will be indicated that the stat block appears in the adventure’s appendix (e.g., “six derro (see Appendix C)”), but frequently the DM is left to just guess where they should be looking for the stat block.

Which is just wild! Particularly since the stat block appendix is sometimes only sort of vaguely alphabetical: In Dragon Heist, for example, “Griffon Cavalry Rider” appears under the “City Guard” sub-heading and some named NPCs appear alphabetically, while others are instead grouped by the faction they belong to.

In other words, 5^th Edition kinda just gave up.

As far as adventure stat blocks containing the essential information for running an encounter, D&D has more or less returned to the earliest and most primitive days of the hobby, leaving the DM to flip madly back and forth between pages and/or browser tabs in search of the stats they need.

But maybe that’s just the way it needs to be. From a certain point of view, you can argue that they’ve never managed to truly get it completely right: I’d argue the closest they got was probably during the ‘80s when the adventure stat blocks were short, concise, and could be easily integrated into an adventure key, but those also relied heavily on technical jargon and could easily baffle a new player.

Of course, D&D is not the only game in town.

Go to Part 3: Other Options

I think a stat block can tell you a lot about a roleplaying game. What types of information do you need to juggle? What does the game consider important in distinguishing one character from another? How complicated will the game be to prep and run?

Holding that thought in mind, let’s take a brief tour of the D&D stat block.

(If you’re not already familiar with the history of D&D, you might want to take a quick tour to orient yourself.)

THE 1970’s

The first published adventure module was “Temple of the Frog” by Dave Arneson in Supplement II: Blackmoor (1976). Everyone was still trying to figure out the entire concept of a “module” and how it should be presented, and “Temple of the Frog” didn’t actually contain stat blocks, instead describing everything narratively. For example:

Room 7: (Company Office) This contains two desks and chairs, a locked foot locker (and an old crypt in the walls of the room which contains five skeletons of 2 hit dice that have armor class 7 and move 6” per turn.

The next published adventure module was Palace of the Vampire Queen (1976) by Judy and Pete Kerestan, which was published by Wee Warriors. (As we’ll see, during this time period the state of the art was being pushed forward by a lot of third party publishers.) The entire adventure key for Palace of the Vampire Queen was presented in a semi-tabular fashion:

There’s not a really “stat block,” per se, but you could interpret one of sorts between the “Creatures” Encountered” and “Max. Damage” columns, e.g. “2 vampire guards — 23, 26.”

Later that same year the Metro Detroit Gamers published The Lost Caverns of Tsojconth, a convention scenario designed by Gary Gygax (based on an original adventure by Rob Kuntz) for WinterCon V. This mostly continues the “creature name + hit points” formatting, but formalizes it a little bit:

L.
10 SAHUAGIN: HP: 16, 14, 12, 11, 11, 10, 10, 9, 9, 7. These creatures appear at the eastern edge of the island within 1 turn of the voice (#7) speaking. They ATTACK. No treasure.

Notable here, however, is the first appearance of NPC spell lists:

K.
COPPER DRAGON: HP: 72. Neutral, intelligent, talking, has spells: DETECT MAGIC, READ MAGIC, CHARM PERSON, LOCATE OBJECT, INVISIBILITY, ESP, DISPEL MAGIC, HASTE, and WATER BREATHING. It is asleep and will waken in 3 melee rounds or if spoken to or attacked.

Still in 1976, we also have both the Gen Con IX Dungeon (by Bob Blake) and City-State of the Invincible Overlord (by Bow Bledsaw and Bill Owens) from Judges Guild. The latter used a tabular presentation for many NPC stat blocks:

These Judges Guild modules, however, also include what are likely the first true stat blocks with entries like this:

Two Mercenaries: FTR, N, LVL: 3, HTK: 1
Bartender Koris Brightips: FEM, FTR, CG, LVL: 2, HTK: 4, AC: 9, Dagger, sings.

Displacer Beast AC: 4 Move: 15” Hits: 22 Fights 5^th Column

(“HTK” here is Hits To Kill. The original 1974 edition of Dungeons & Dragons notably used many different synonyms for what we now refer to as hit points. Thus HTK, Max. Damage, and similar entries can be found in many early stat blocks.)

We similar stat blocks from Judges Guild throughout ’77 in Tegel Manor, First Fantasy Campaign, and Modron.

In 1977, Wee Warriors also returned with The Dwarven Glory, another adventure by Pete & Judy Kerestan. The monsters here were still being described narratively, as they were in “Temple of the Frog,” but this adventure notably put the hit point totals and other stats in parentheses:

Room #6

2 ore carts, 10 lizard men (HP—5, 7, 10, 11, 10, 9, 8, 6, 7, 12) (AC—5) will try to ambush the party from ore carts. If unsuccessful, will fake panic and try to lead party into Room #5. Are on friendly terms with Cave Troll and Minotaur. Gem in dirt of floor with praying hands etched in surface (raise dead, usable once pe day).

I say “notably,” because in 1978 TSR figured out that there was a bunch of money to be made selling adventure modules and they came roaring into the market, releasing the G series, D series, S1 Tomb of Horrors, and B1 In Search of the Unknown. And this use of parentheses quickly became TSR’s house style.

G1 Steading of the Hill Giant Chief and the rest of the G series was written by Gygax as a way of taking a break between working on the Monster Manual and the original Player’s Handbook, and the parentheticals are still limited to hit points:

3. DORMITORY: Here 12 young giants (H.P.: 26, 24, 3 x 21, 18 x 17, 2 x 16, 14, 13) are rollicking, and beefy smacks, shouts, laughter, etc. are easily heard. All these creatures have weapons and will fight as ogres.

After finishing work on the Player’s Handbook, Gygax took another break and produced the D series. By this point, more information is being dropped into the parentheses (although the overall presentation remains fairly narrative):

Drow Male Contingent: There are 10 male fighters of 3^rd level to the southwest, 2 of whom are on guard duty and will report the presence of any creatures moving along the passage. Other than having 13 hit points each and AC 0 (because of 16 dexterity each), they are the same as a male Drow patrol, i.e. +1 short swords and +1 daggers and carrying hand crossbows and using dancing lights, darkness, and faerie fire (at 3^rd level) once per day per spell. There are 2 4th level fighters as leaders (H.P.: 18; AC -2) with +2 short sword, +2 dagger, and atlatl and 3 javelins. The commander of the unit is a 6th level fighter (H.P.: 28, +3 chain mail, +3 buckler, +3 for dexterity of 17, for an overall AC of -3) armed with +2 dagger, +4 short sword, and hand crossbow with 10 poisoned bolts.

By ’79, Judges Guild had firmed up their presentation of NPC statistics into definite stat blocks. For example, here’s the text from key V-5 in Dark Tower:

Avvakris: 10th level cleric of Set, Align: CE, chainmail, AC: 5, HP: 50, S: 14, I: 14, W: 15, D: 14, C: 11, CH: 15, weapon: mace, spells: bless (reverse), create water (reverse), detect good, detect magic, hold person (x3), silence 15′ radius, animate dead, dispel magic, speak with dead, cause serious wounds (x2), divination, flame strike (x2.)

Seth the Huge: 6th level fighter, align: CE, chainmail and shield, +3 dexterity bonus, AC: 1, HP: 34, weapon: longsword, S: 16, I: 8, W: 5, D: 17, C: 11, CH: 12. Large and cruel looking.

Wormgear Bonegnawer: 6th level fighter, align: CE, chainmail and shield, +1 dexterity bonus, AC: 3, HP: 48, weapon: longsword, S: 16, I: 11, W: 14, D: 15, CH: 11.

Over at TSR, Gygax was also firming things up. In T1 Village of Hommlet (1979), stats are still being dropped into the middle of paragraphs, but they’re being completely contained in parentheses (instead of leaking out as we saw with D1) and becoming increasingly standardized. Examples include:

Canon Terjon (6th level cleric — S 11, I 10, W 16, D 12, C 16, Ch 8 — chain mail, shield +1, mace; 41 hit points; invisibility and mammal control rings; typical spells noted hereafter)

Jaroo Ashstaff (7th level druid — S 11, I 11, W 18, D 9, C 15, Ch 15 — HP: 44, padded armor, cloak of protection +2, staff of the snake, +1 scimitar, ring of invisibility; spells given below)

black bear (AC 7; HD 3+3, HP: 25; 3 attacks for 1-3/1-3/1-6 plus hug for 2-8 on a paw hit of 18)

two dogs (AC 7; HD 1+1, HP: 5, 4, 1 attack for 1-4 hit points of damage)

THE LONG, SLOW EXPANSION

By 1980, therefore, stat blocks were assuming standardized forms and TSR was beginning to create editorial standards which applied to all of their books. (This is a trend you can also see with TSR’s dungeon keys.)

When T1 was republished in 1985 as part of T1-4 Temple of Elemental Evil, the stat blocks were revamped for this standard:

Black Bear: AC 7; HD 3 +3; hp 25; #AT 3; D 1-3/1-3/1-6; SA Hug (if paw hit 18+); Dmg 2-8; XP 185

Jaroo Ashstaff: AC 6 (padded armor); Level 7 Druid; hp 44; #AT 1; D by weapon or spell; XP 1427; cloak of protection +2, ring of invisibility, staff of the serpent (python), scimitar +1
S11 I 11 W 18 D 9 Co 15 Ch 15

Standard druid abilities: identify plant type, animal type, pure water; pass without trace; immune to woodland charm; shapechange 3 times per day; +2 bonus to saving throws vs. lightning; q.v. PH page 21.

Spells normally memorized:
First level: detect magic, entangle, faerie fire, invisibility to animals, pass without trace, speak with animals
Second level: barkskin, charm person or mammal, cure light wounds, heat metal, trip, warp wood
Third level: cure disease, neutralize poison, summon insects, tree
Fourth level: cure serious wound, plant door

Spell lists are obviously, by their very nature, lengthy. But you can see how compact and, arguably, even elegant the standardized “stat-line” format is for the stat block. Albeit nearly incomprehensible when you first look at it, it doesn’t take much familiarity with the rules before this stat block becomes very easy to use by virtue of taking every piece of necessary information and putting it right at your fingertips.

BECMI stat blocks were largely identical:

Goblins. (2d4) AC 6; HD 1-1; hp 3 each; MV 90′ (30′); #AT 1; D 1d6; Save NM; ML 7; AL C; XP 5 each. Each goblin carries a spear and 2-12 ep.

Champion (7th level Fighter): AC 6; F7; hp 42; MV 120′ (40′); #AT 1; D 1d4 (+2 for magic weapon); Save F7; ML 9; AL L; XP 450.

And the formatting of these stat blocks was largely unaltered when 2^nd Edition rolled around. Here’s a sample from a late-2^nd Edition module:

Behir: AC 4; MV 15; HD 12; hp 70; THAC0 9; #AT 2 or 7; Dmg 2d4/1d4+1 (bite and constriction) or 2d4/1d6 (bite/6 claws); SA once every 10 rounds can breathe bolt of lightning up to 20 feet long that inflicts 24 points of damage (save for half), swallow whole on an attack roll of 20 (victim loses % of starting hp until death on the 6th round, can cut himself out by attacking AC 7, but each round the victim spends inside the behir he faces a cumulative -1 damage penalty); SD immune to electricity and poison; SZ G (40′ long), ML Champion (15); Int Low (7); AL NE; XP 7,000

Here, though, we can start making two additional observations. First, as the rules for a creature become more complicated, the short simplicity of the stat block begins to decay into a mass of incomprehensible text.

Second, the earliest stat blocks were kept minimalist in part because many stats were standardized by Hit Dice and keyed to a unified chart. As the rules for creatures became less standardized, more information needed to be coded into the stat block (like THAC0), directly contributing to the “mass of text” feel.

Go to Part 2: The 21st Century