It’s a Kind of Magic

Magic! Low or high, complex or simple, most fantasy games play with it. AD&D’s magic system is both glorious and complicated. I’ve discussed the spell casting part before, this post is about magic items. This is of interest as people regularly suggest that 1e and older edition games are “Monty Haul”, or that they can be, I’m interested in the baseline.

There are three sources of magic items for PCs in AD&D, random encounters, “dungeons” and magical item creation.

The magical item creation rules are fairly restrictive and require high level casters and a lot of gold. I don’t know many groups that use this route to get magic items. So they can be set aside for the purposes of this conversation.

Random Encounters

God bless Gary Gygax, but the work you have to do to piece together things in 1e can be daunting.

Wandering monster encounters hold the possibility of magic items. When a wandering monster is encountered the DM rolls to see if it is encountered “in lair”, if it is, then the monster’s treasure will be in the lair somewhere, protected, hidden or obvious, all up to the DM. She also rolls to see how many monsters appear.

If they roll that the monster was not “in lair”, then the treasure is in the lair and the adventurers have to find it, by no means easy or possible. Unless they have a ranger in the party to track the monster’s trail back to the lair, it’s really up to the DM how hard that will be.

But say they find the lair, and they find the treasure.

You then go to the treasure Appendix of the Monster Manual, and you roll on the tables. Say the party defeated a group of 25 lizard men, treasure type D.

D has the following odds of magic items coming up:

15% chance of any two maps or magic + 1 potion

So there is an 85% chance nothing will come up. But say you roll a 10%, so there is magic!

You will note that you were rolling on the “Maps or Magic” column.

You now go to the DMG, and go to the “Map or Magic Determination” table on pg 120 of the DMG, you roll 01-10: Map, 11-100 - Magic Item Table

So there is a chance that one or both of your rolls for magic in this case could produce maps instead.

Bummer eh?

Not really, if you get a map, then you roll for what kind of map it is, it could be a false map (5% chance) so you would waste hours of gameplay for nothing, or a money treasure map, a magic treasure map or a combined hoard map. The tables you roll on, which appear down the page, are VERY generous, ridiculously so, thus getting a map rather than magic items is winning the lottery. Of course, the challenge to get to the treasure should be proportional to the hoard, but still, this is a typical example of how Gygax buries something important, a random roll here could produce mountains of treasure.

But say you don’t roll a map, what then?

You go to table III Magic Items.

Gygax pointed out in the MM that if you rolled low for the number appearing you could adjust the treasure results downwards (he didn’t specify how to do this) and vice versa if you rolled up a larger number appearing (didn’t specify how to do this either). But that is only for magic items from random encounters.

Now, before going to the magic item table in more detail, I want to look at the system for dungeons, as it also leads you back to the magic items table.

Dungeons

Whenever you have treasure protected for some reason, or a structured environment like a dungeon, crypt, fortress, whatever, you are not to use the system for random encounters, you are to place that treasure yourself. Gygax gives you a lot of warnings about how to avoid overdoing it when assigning magic treasure, without being very clear on what “overdoing it” would constitute.

So most people would have looked for guidance somewhere. Modules are an obvious idea, see how much treasure is in there and try to do something comparative. Pregens are another way to assess this, as is the Creating a Party on the Spur of the Moment Appendix (Appendix P), which can be used to create pregens and calibrate treasure, and the Humans subtable in the Random Monster Encounters section (Appendix C).

However, it is worth a look at Gygax’s treasure system for dungeons, tucked into his random dungeon creation section.  Essentially, if you are stuck creating a dungeon, or stocking it with treasure or monsters, you can use Gygax’s random generation system.

Assessing this is difficult, as the randomness means you could get a big or small dungeon in terms of rooms. The number of rooms determines the number of times you get a chance to roll for treasure being in the room, so the number of rooms determines in a rough way the odds that you will have magic treasure. Every time you use the system it produces a different dungeon layout and number of rooms.

So let’s assume a small dungeon as a test case, about 40 rooms.

For each room you would roll on the contents table, Table V:F

It gives you a 20% chance of getting treasure, with or without a monster (15% chance with a monster, 5% chance without).

20% of 40 is 8, so let’s say 8 rooms have treasure, and ¾ of those would have monsters, so you would have 6 monsters, 8 treasure rolls (6 with monsters, 2 without), let's assume the party defeats 6 monsters and finds the treasure for 6 of the 8 rolls, as it is not automatic that they will find all the magic treasure (there are further tables for traps and concealment of treasure for example). Assume that all of the treasure rolls here are for monster encounters.

6 treasure rolls, each one has a 13% chance of coming up with magic treasure. When the treasure is guarded by a monster you get two rolls, the odds of a magic item coming up with two 13% rolls is about 24%.

So there is a 24% chance you will roll a magic treasure, you get 6 rolls.

What are the odds of getting at least one treasure with six rolls of 24% chance each? About 80%, so your odds of getting at least one magic treasure roll with this table in a dungeon with 40 rooms are around 80%. That’s pretty good. The odds of you getting a magic item on the random encounter treasure tables are a lot lower for all of the monsters.

Back to the Table

So in both cases, the random encounter and the planned, dungeon encounter, you are led back to the magic items table.

As Gygax points out below the table, “potions, scrolls, armor and arms are plentiful”, “Rings, rods and miscellaneous items of magic represent only a 25% occurrence on the table.”

So although dungeons and similar environments will produce more treasure as they increase in size, random encounters produce treasure based on the monsters in question and the number appearing. However, in dungeons the treasure is tied to the dungeon level, which is roughly keyed to the monster difficulty, as well, magnified by number of rooms. So on the overall most dungeons provide a greater potential reward than random encounters.

Potions and disposable magic like scrolls come up 35% of the time, just over ⅓.

Misc magic items, of wildly varying power, come up about 15% of the time, rings 5%, Rods, Staves and Wands 5%, approximately 40% armor, weapons and swords.

Now, a quick look at the individual tables.

Potions are almost all equal odds, so it’s pretty random.

Scrolls are RIDICULOUSLY good. You have an nearly 40% chance of rolling a protection scroll, 2% chance of a cursed one, and around 60% chance of a scroll with spells.

But check this out, you have about a 40% chance of rolling a multiple spell scroll, and on each one there is the potential for spells above first level up to 9th (for magic users). These can be game changers.

The ring table is pretty even, all useful.

The Rod/Staff/Wand table favors wands, but the odds of any individual rod/staff/wand coming up vary widely based on power.

The misc magic tables are all over the map, table one includes the subtable for artifacts and relics, all tables have cursed items on them, representing around a 10% chance of a cursed item from each table. Each also has very powerful items on it, e.g. items of 6000 xp +, in this case being around 5% of the table.

The armor table favors plate the most with the odds of rolling a +1 and +5 suit of plate mail being 6% and 1% respectively, with an overall 18% chance of magical plate.

Swords are relatively balanced between options, with a 76% chance of getting more than just a +1 sword.

Finally misc weapons are spread out fairly evenly between a few major weapon types.

Now, just to give perspective, say you were a fighter who used a broadsword, what are the odds of finding a magical broadsword in a random encounter with say two manticores and the case mentioned above of 6 rolls on the treasure table, e.g. when the fighter and his party clear out the hypothetical 40 room dungeon. In the case of the manticores let's assume it is either in lair or the party found the lair.

Manticores are treasure type E, which is a 25% chance of any three plus scroll. “Any three” here means you roll on the “Map or magic” table, a 90% chance of magic rather than map. So there is a 25% chance of three rolls, each one has a 90% chance of moving on to magic. Let’s keep this simple and say there is an approximately 25% chance of three magic table rolls. So say the 25% comes up. Since you have 2 of 4 potential manticores in the fight, you don’t adjust the treasure amount.

So for that fighter, the odds of a magic broadsword coming up are based on the odds of magic swords coming up, which are 10%, so the odds of a magic sword coming up on three rolls on the table are 27%.

Now, what are the odds of a magic broadsword coming up with a swords roll? 20% fixed that whatever magic sword comes up, it will be a broadsword.

So what are the odds of the fighter getting a magic broadsword of any kind after the manticore encounter?

25%x27%x20% = approximately 1%

What about the dungeon mentioned above?

Well, with 6 attempts on the table the fighter would have a 27% chance of getting a magic sword result, and a 5% chance of getting a magic broadsword of any kind.

Obviously the result here is tied to the specific conditions, but generally this supports the idea that random encounters produce less magic.

Observations

Gygax stated that you shouldn’t use the random encounter magic item system for the dungeon. However, both systems get you to the magic item table. Because there are multiple rooms in a dungeon, and the odds of getting magic items are tied to the number of rolls you get, and thus the number of rooms, dungeons present the best bang for your adventuring buck.

So why not use the treasure system for random encounters in the dungeon? If you did that, you would roll for every monster, so in our 40 room dungeon where the party defeated 4 monsters and got 6 treasure rolls, let's say it’s a 2nd level dungeon. And let’s take 4 random monsters from the 2nd level dungeon tables to see what comes up.

For a 2nd level dungeon, you roll on a table on pg 174 to determine what table to consult, you don’t just consult the 2nd level monster table for the second level of the dungeon. You have roughly 60% chance of rolling on monster level 1, 20% chance of rolling on level 2, and a 20% chance of rolling on levels 3 or higher.

So let’s assume level equivalence just to keep it simple. And further assume we don’t roll “characters” as an encounter as that can skew the magic significantly (11% chance of characters). Lastly assume average numbers of monsters appearing.

Up above we determined that the odds of getting at least one magic item for this 40 room dungeon would be around 80%.

So four random picks from the table:

8 giant centipedes

12 giant rats

2 piercers

5 troglodytes

What would be your odds of getting magic items from these as random encounters outside the dungeon?

Giant centipedes and piercers have no treasure (for real) based on treasure type, so the first important difference to note is that all rooms with monsters in the dungeon system for creating magic treasure get a chance for magic treasure as an option, as do rooms without monsters. 

Some MM monsters have no treasure type at all.

But for the trogs and the rats, there is a 30% of any three magic items and a 10% chance of any two, so the odds of getting at least one magic item in a random encounter with these four monsters is around 37% or a bit more than a third..

So the biggest difference here between systems is that the dungeon magic item system is far more generous than the random encounter system. The bigger the dungeon the better. The dungeon also compresses time, so rather than having to travel enough to trigger a sufficient number of random encounters, they are all organized in one place. That and the better treasure odds make them far more lucrative.

Last thing to check, which is more dangerous?

The dungeon system keys you to pre-selected levels, but there is some variance. There is a 10% chance of getting above your current dungeon level for the table roll, e.g. the monsters being more powerful than usual for that level, and about a 60% chance of getting under. So for the *most part* the dungeon tables are calibrated to level.

What about the random encounter tables?

Take a sample, the Tropical and near Tropical uninhabited tables.

There are about 15 monster entries for dangerous monsters like shedu, lamia, rakshasas, nagas, weretigers or coatls. The remainder are divided up between animals, giant animals and pack animals. spread across multiple tables for different terrains, duplicates abound. So your chance of encountering something fairly difficult for low to mid level parties, is around 15% or so of the tables for each terrain type. So 85% of the time, it's an animal, a giant animal or men. 

Men can be dangerous in large numbers, as bandits for example.

So the main difference I can see here is much greater variance in monster strength for random encounters than dungeon encounters, the potential for a full range of encounters from very easy to very hard, and a smaller chance of magical reward from random encounters.

Keep in mind that the stats here are linked to a hypothetical 40 room dungeon, there is a lot of swing in my numbers because of this. The size of the gap is what makes it notable, otherwise the swingyness would have drowned it out.

So it appears that Gygax wanted dungeons to be the more direct source of treasure. And thus that dungeon crawling games would be more magic heavy.

The Experiment

Three years ago I decided to start a D&D business, and in that time I have run 3 concurrent campaigns for 2 years time, playing a total of 160 hours at the table between them. I am now extracting information to see where I’m at, running a more or less BTB game. My house rules are pretty combat focused, otherwise magic and experience and such are about 95% BTB.

Here I was interested in how many magic items we have picked up, but a word to how I did this.

I don’t really do dungeon type exploration very often, I run an open, sandbox game with a city setting, a nearby coast, and lots of freedom. They have explored some temples, some crypts, a sunken tower, an underground river, that sort of thing.

But I certainly don’t do staged difficulty, where the further you delve the greater the challenges and the greater the reward, the increase or decrease in challenge is related to the environment and how the PCs navigate it. That isn’t organized in the way you need to be to use the system for dungeon treasure.

So I decided I would use the treasure allocation system meant for random encounters to help me stock my game world, not the dungeon one. This meant that the party could find a map rather than a magic item and be led to a huge hoard. It could also mean that they roll nothing, as the percentages in the MM are fairly low.

However, I followed Gygax’s advice and if the number appearing of a monster or opponent was at the top end, I would increase the odds of success, if below I would decrease it. I decided to double and halve, e.g. if the number appearing of a monster was 2-8 and I rolled 8, then the treasure roll is made twice, you keep the best results. If I roll a 2, if the odds of getting a magic item were 20%, they would be 10%

When the party fought monsters I rolled on the MM tables as appropriate, when they fought characters I rolled using the “Creating a Party on the Spur of the Moment” Appendix or the Humans subtable of Appendix C,  with some flourishes for my campaign world to determine what they had. When they met NPCS and fighting happened, they sometimes got access to their opponent’s items at the end.

Take one example, my groups went up against a lot of warlocks, one, Moon Abithin, was generated using Appendix P tables, he was 7th level.

He had a 105 % chance of a ring of protection over a 28% chance of bracers. That gave him a 7% + 15% = 22% chance of the ring being “exceptional”, e.g. beyond +1. I rolled that, so it was a +2 ring, I rolled then a 7% chance of it being greater than that, and failed. So it was a +2 ring of protection.

He had a more than 100% chance of a scroll, he got one of three spells with Death Spell, Conjure Elemental and Cone of Cold.

There is a 105% chance of having a magical dagger, that made it a 22% chance of exceptional, I rolled that, and a 7% chance of exceptional again which I also rolled, so he had a dagger +2/+3 versus larger than man sized creatures.

He had a potion of healing.

I fiddle with the table for miscellaneous items in Appendix C as it is too small, I roll instead on the misc tables in the DMG 1-4 times, in this case I rolled twice and got a Bowl of Commanding Water elementals, and a broach of climbing, a home brew replacement for a broach of shielding in the DMG.

Insofar as they go up against warlocks and mid to high level fighters in my setting, they will occasionally get their items.

I’ve run two of the campaigns using these rules with the same or continuous PCs, there have been some deaths and replacements. But using the random encounter treasure rules for in combination with Appendix P and the subsection of Appendix C, this is where we are at after 120 hours between two campaigns (I ignore my home game as it has had much higher PC turnover).

The PCs in both games are 3-7th level after 2 years and 120 hours of play. They have had 6 fatalities, there are 12 players between the two games. You will see the items from Moon Abithin spread throughout the party, Moon did not survive the encounter with the party...

Party 1

MU 6th - Scroll with Death Spell, Conjure Elemental and Cone of Cold, Bowl of Commanding Water elementals +2 ring of protection. Scrolls of 3, 6 and 5 spells were either cast or taken to spell books, so that’s 17 scroll spells in two years of play, not including “level up” spells.

Fighter 3rd - Potion of Invulnerability, +1 spear

7th level thief - +1 short sword, potions of Water Breathing and ESP, dagger +2/+3 versus larger than man sized creatures

Druid 6th - +1 scimitar, +1 armor, potion of fire resistance, broach of climbing (climb walls 95%)

Bard 1st (F5/T6) - +1 bow, potion of extra healing, oil of etherealness, magic fish - creates water 3x per day as 10th level priest

Totals: 17 spells, one ring +2, 6 potions, 4 +1 weapons, one exceptional weapon, 1 magical armor, 3 misc magic items

Party 2

Fighter 6th - Sword +1 / +3 versus giant creatures, potion of healing

MU 6th - Scroll: Invisible Stalker, Monster Summoning 2, Globe of Invulnerability, Ring of Protection +1, Potion of healing. Similarly found scrolls of 2, 6, 8 and 5 spells, some went into spell books, some were used. 24 spells found in 2 years of play.

Thief 7th - Ring of feather falling, +1 short sword, +1 leather armor, oil of slipperiness

Paladin 5th - +1 chain, +2 shield, +3 longsword, potion of growth

Ranger 5th - +1 ring of protection, +1 bow, Potion of plant control, potion of clairvoyance

Monk 5th - Ring of protection +1, ring of free action, fire bracer - allows three fiery fist punches per day to his open hand attacks, extra 1-4 damage for each.

Fighter 6th - Bastard Sword +1/+3 versus magic using creatures, potion of frost giant strength

Totals: 24 scroll spells, five rings, seven potions, 3 magical armor/shield, 2 +1 magical weapons, two exceptional weapons, one miscellaneous weapon.

They have lost two scrolls of multiple spells (to a cone of cold spell), two potions to a lightning bolt, lost armor to critical hit damage, and sacrificed one +1 shield to deflect a deadly blow.

I have no idea what running the game with dungeon style magic would do to this, but my guess is that the raw number of magic items would increase about 100%, if not a bit more.

I’m quite comfortable with this level of magic, the players are as well, I do not get requests for “more magic items” from either group. Interestingly, when I look at module pregens as examples I see that this level of magic is roughly around what pregens have in most of the major 1e modules. I’m a bit higher magic than these pregens, but not that much. It makes me wonder if early TSR designers used the DMG tables to generate characters for modules.