Chainmail And Weapon Modifiers

In my last post I took a look at the Chainmail Man-to-Man tables and distilled each armor class down to a standard to-hit value. To spare you the pain of looking back over that rambling mess, here’s the table of to-hit values for every armor class, listed for 2d6 and d20 combat, as well as a straight percentage chance of hitting.

	9	8	7	6	5	4	3	2
2d6	7	7	8	8	9	9	10	11
d20	10	11	13	13	14	15	18	19
d100	0.57	0.5	0.42	0.42	0.35	0.29	0.15	0.1

Recall that, in Chainmail, AC 9 represents an unarmored opponent, AC 8 represents leather armor, AC 7 a shield alone, AC 6 leather and shield, AC 5 chain mail, AC 4 chain and shield, AC 3 plate armor, and AC 2 plate and shield.

This table shouldn’t be too big of a shock. It follows more or less the progression suggested in the “Alternative Combat System” in Original D&D, with the caveat that plate is significantly better than chain mail. Now we’re going to move on to weapon modifiers. In Chainmail, every weapon has a certain chance to hit each armor class. In this distilled system, this is represented by a table of to-hit modifiers, where the weapon is compared against the armor class, and a certain bonus added to the to-hit roll. Below is the table used when the 2d6 combat system is used.

	Reach	9	8	7	6	5	4	3	2
Dagger	1	1	0	0	0	0	-1	-2	-1
Hand Axe	1	0	0	0	-1	-1	-1	-1	-1
Mace	3	-1	-1	0	-1	1	1	3	3
Sword	4	0	-1	0	-1	1	0	0	0
Battle Axe	5	-1	-1	0	0	2	2	1	1
Morning Star	6	1	1	1	1	3	2	2	3
Flail	7	0	0	1	1	3	2	4	4
Spear	8	-1	-1	-1	-1	-1	-1	-1	-1
Polearm	9	1	1	2	1	1	1	1	1
Halbard	9	-1	-1	0	1	3	3	3	3
Two-Handed Sword	10	1	1	2	2	4	4	4	4
Mounted Lance	11	2	2	3	3	2	2	2	2
Pike	12	-1	-1	0	0	1	1	1	1

Any positive numbers are simple bonuses to be applied to the to-hit rolls, while negative numbers are penalties. Using this table, we get exactly the same combat odds as in the original Chainmail. It’s obvious from the above table that some weapons, especially bigger ones such as the two-handed sword and the mounted lance, are simply better than their counterparts. To get a feel for exactly how much bigger, in a language that D&D players are more likely to understand, here is the same table converted to the d20 system (following the to-hit numbers stated earlier in this post).

	Reach	9	8	7	6	5	4	3	2
Dagger	1	3	2	0	0	-1	-3	-2	-1
Hand Axe	1	0	2	0	-3	-4	-3	-1	-1
Mace	3	-3	-2	0	-3	1	2	9	6
Sword	4	0	-2	0	-3	1	0	0	-1
Battle Axe	5	-3	-2	0	0	5	6	2	1
Morning Star	6	3	4	3	3	7	6	5	6
Flail	7	0	2	3	3	7	6	11	10
Spear	8	-3	-2	-2	-3	-4	-3	-1	-1
Polearm	9	3	4	6	3	5	2	2	1
Halbard	9	-3	-2	0	3	7	9	9	6
Two-Handed Sword	10	3	4	6	6	10	11	11	10
Mounted Lance	11	5	6	8	8	7	6	5	4
Pike	12	-3	-2	0	0	1	2	2	1

Now it should be a lot clearer exactly how much better than your standard weapons the two-handed sword, for example, is. To penetrate plate and shield a standard character with a two-handed sword needs to roll a 9 or higher on a d20.

You’ll notice also that weapons have a “reach” value (this is called “class” in Chainmail, but I think “reach” is more descriptive). This has the following effects:

• In the first round of melee between two opponents, the attacker (being the one who moved into melee) strikes first unless the defender has a weapon whose reach is 2 greater than the attacker’s. This simulates the defender setting his spear or whatever against the charge.
• In the second and each subsequent round of melee, the same person who struck first last round does so again, unless the opponent has a weapon whose reach is 2 lower than the first combatant’s. This simulates the added speed and maneuverability that having a lighter weapon gives you.
• If combatant A’s weapon has a reach of anywhere from 3 lower than combatant B’s to 1 higher than combatant B’s, combatant A can parry his opponent’s attack, forcing him to subtract 2 from his to-hit roll, though combatant A can not make his next attack.
• If combatant A’s weapon has a reach from 4 to 7 lower than combatant B, then combatant A can either choose to strike first or parry combatant B’s blow. If the parry is successful, combatant A still gets to make his counterattack.
• If combatant A’s weapon has a reach of 8 lower than combatant B, then combatant A gets the first blow, plus he has the option of striking again or parrying.
• Any combatant whose weapon’s reach is at least 4 lower than his opponent’s gets another blow in addition to the benefits listed above.

So now we have different weapons that feel completely different, so while a burly fighter wielding a two-handed sword might make mincemeat out of a scrawny magic-user with his dagger, the magic-user still gets two chances to strike the fighter before the fighter even makes his first attack roll. Under the Chainmail rules, weapons are all different, many weapons having certain advantages over others. We’ve given weapons character even though they all deal 1d6 damage with a successful hit.

This also goes a long way towards differentiating classes at lower levels. I’ve often heard the complaint that, at low levels, there is no difference in fighting capability between fighters, clerics, and magic-users. Now the difference is clear: fighters can use any and all weapons, from the lowly dagger to the mighty two-handed sword, while magic-users can only use a knife. Not only would these two classes have different results in combat, playing them would feel very different.

Now what of when PCs are fighting monstrous creatures that have no immediate analogue in the weapon vs. AC table? I think the best solution is to give each weapon a simple modifier to hit to be used when facing monstrous foes. This modifier would be used regardless of the opponent’s armor class, and would simply be a reflection of the overall effectiveness of a weapon. Or you could just leave that system the way it is, and give those magic-users a fighting chance against a dragon.

You may notice I haven’t covered ranged weapons. This is because ranged weapons suck, and the math involved sucks, and I haven’t had the drive or opportunity to do it yet. I also haven’t talked about combat progression yet. I’ll get to that as well, but again the math is a bit wonky, or at least it seems that way to me. As it is this post pretty much outlines an entire combat system that you can plop into a D&D game. I plan on using this in my next game, whenever that happens, either in a 2d6 or a d20 form.

Whew. That was a lot of post with very little fluff. Here are some pictures to make everybody chill.

My latest D&D-related acquisition.

My current sci-fi reading.

Lumpy Space Princess

This Is Chainmail. This is D&D.

[WARNING: This post contains nerdery and numbers. Proceed at your own risk.]

About a year ago I wrote this analysis of the Chainmail mass combat rules. I have this on-again off-again project I’m working on, which basically amounts to a retroclone. Basically I’m looking to write a system that stays very true to the original rules as written, with cool things added on from later iterations of the game in a way that they don’t upset the original spirit. Yes, this is more or less a set of house rules, that I at one point tend to write up in some sort of organized book-like file, more for my own gratification and the people I game with than for public consumption.

This is pretty long and rambly, so here's a picture of what Google assures me is Harold's death as presented in the Bayeux Tapestry.

One of my primary concerns in the crafting of this system is adhering to the rules as presented in Chainmail since the original rules for Dungeons & Dragons mentioned explicitly many times that the rules were to be used in conjunction with Chainmail. For those unfamiliar with that game, there are basically three different levels of combat detailed in Chainmail. The “standard” game is played between armies with figures representing twenty men apiece. The “Man-to-Man” rules deal with skirmish-level combat, where each model is an individual combatant. Fantasy combat deals with contests between fantastic creatures, such as dragons and trolls and the like.

In the post I referenced a year ago, I dealt with the standard 20:1 rules. That post was more-or-less a distillation of the base system, but my thinking at the time was to use that, along with the “Fighting Capacity” designations for each class as detailed in the LBBs, to fight smaller-scale conflicts. I dedicated a great deal of time and effort into making a combat system that would closely follow the guidelines laid down in that post. However, in so doing, I gradually came to the realization that, for the typical combats that my players tend to get into, those rules are just not fun, and any alteration I made to them took them farther from the spirit of the original. I often found myself stuck between that which is true to Chainmail and that which is fun.

So I’m abandoning that train of thought for the time being to focus more on the Man-to-Man rules. Looking at these, it’s much closer to the standard D&D combat with which we’re all familiar. In fact, one could easily just pick up the rules wholesale and drop them into D&D and have a perfectly workable system. So why not just do that? There are two problems.

• The rules assume all combatants are of equal skill, with the only difference being equipment. Keeping this standard would basically make leveling up a simple matter of better saves and better hit points for the fighter, with no improvement in fighting capability whatsoever.
• All attacks are rolled on a table, where the attacker’s weapon is cross-referenced with the armor of the defender. This is wonderful for two humanoids in battle, but it is completely unworkable for creatures like dragons, or creatures wielding exotic gear, or bears.

OK, so that first problem can probably just be fixed by giving higher-level fighters some sort of bonus to their die rolls or whatever. I’m not dealing with it at the moment.

The second is more problematic, seeing as how the game is Dungeons & Dragons, and a combat system that doesn’t let you fight dragons doesn’t really gel. So how do we deal with it? One option is simply to assign those monsters the armor that most closely approximates their natural defenses. A troll, with its thick, rubbery hide, might be given a leather armor class, whereas a scaly dragon would get plate. I imagine most animals would be assigned leather or no armor under this system, and we can tack on additional rules, such as small creatures forcing the attacker to subtract 1 from his to-hit roll.

That might be one way to go, but then what happens when those creatures attack? Do we likewise assign them a weapon? What would a dragon’s bite be? The system as written has the effectiveness of certain weapons versus certain armor deeply entrenched, and it seems like trying to shoehorn a bunch of non-humanoids into the mix takes away from the original intent.

The alternative that I propose is to assume that the system as it stands represents not only a weapon’s inherent usefulness against a certain armor type, but a combatant’s knowledge of how exactly to use such weapon to its greatest advantage (and perhaps the defender’s ability to counter those weapons which are easily countered). If we give each armor type an armor class (say, between 9 and 2), we can give each armor class a certain percentage chance to hit, which is used universally. If the combat is between two humanoids, we then have a separate set of modifiers for each weapon against different types of armor. When humanoids fight fantastic creatures, there are no modifiers, representative of the fact that the combat tactics to fight, say, a dragon, are quite exotic to the warrior, so he can’t make use of any inherent advantages his weapon may have over the opponent. If this seems familiar, it’s because it is basically the system used in 1st edition AD&D.

To find the baseline probability to-hit for each armor class, I took the average of every weapon’s ability to hit every armor class, percentage-wise. The results are as follows:

Unarmored (AC 9): 57%
Leather (AC 8): 50%
Shield (AC 7): 42%
Leather & Shield (AC 6): 42%
Chainmail (AC 5): 35%
Chain & Shield (AC 4): 30%
Platemail (AC 3): 15%
Plate & Shield (AC 2): 10%

You’ll note that in the original rules wielding a shield alone is on average better than wearing leather armor with not shield. These percentages can be converted into target numbers. Here I converted these into the number one would need to roll on 2d6, since this is the type of roll used in Chainmail.

AC 9 and AC 8: 7
AC 7 and AC 6: 8
AC 5 and AC 4: 9
AC 3: 10
AC 2: 11

Since the range of results is fairly limited using only 2d6, there is not much variation in targets. These are really approximations of the true average odds, but I think they are the best approximations we can have. Here’s the same table with target numbers using the more familiar (and more granular) d20.

AC 9: 10
AC 8: 11
AC 7: 13
AC 6: 13
AC 5: 14
AC 4: 15
AC 3: 18
AC 2: 19

Basically, we have the standard combat table for 1st level characters, with the exception of AC 7 and the jump to plate mail.  I think this goes to show that Gygax and co. were either very careful to map out the combat table so that it fit with Chainmail, or very careful when writing Chainmail in the first place to have a steady progression. I believe the jump between chain & shield and plate mail is a product of the bell curve distribution of 2d6 versus the linear distribution of the d20, and likewise with the blip at AC 7.

So that’s everything you need to fight those fantastic creatures and bears and whatnot. Roll a die, find their AC, and see if you score a hit! Then, of course, roll 1d6 damage and deduct it from the opponent’s hit points. Rinse and repeat. Soon I’m going to write a post looking at the weapon modifiers, once I figure out how to make nice-looking tables. That’s where things get interesting.

Variant Rules: Curvy Battleship

Here’s one for all you Battleship afficionados out there.  A nice set of variant rules to spice up an old classic and add all new levels of strategy and depth to an already highly entertaining game.  Battleship, I feel, is an underappreciated gem of old school wargaming, and one with a general lack of variants and mods.

The way Curvy Battleship works is thus. Each player still has the 5 classic ships (I will use the following names and numbers of pegs, Patrol Boat – 2, Submarine – 3, Destroyer – 3, Battleship – 4, Carrier – 5).  However, rather than representing the ships with the traditional game pieces, each ship is represented by white “miss” pegs. This requires that the player remembers exactly which peg corresponds to spots on which ship, but this is a mild inconvenience, and it is necessary considering the one major difference between Classic and Curvy Battleship.

Curvy Battleship ships, as may be deduced from the title, need not conform to the traditional straight shape, but can turn, at will, to any adjacent square.  That is to say, a carrier may occupy any 5 squares, as long as each peg is adjacent to the previous peg. This path can curve diagonally as well.

I played a great game of this earlier today with Kevin, with whom I developed this variant, and lost handily, falling to his strategy of mostly ignoring the curvy rules, and misleading me with straight ships (which quite goes against his character). The fact that my ships resembled a certain recognizable yet mostly inappropriate pattern probably also helped spell my demise.

I am only half-joking.

Parsing the Chainmail Combat System

The original rules for Dungeons & Dragons claim to be intended for use with the combat system put forth in the earlier set of medieval wargames rules Chainmail, co-written by Gary Gygax and published by Tactical Studies Rules.  Whether or not Gary ever actually used these rules in the context of D&D is debatable, but I believe it worthwhile from both a historical and a potential gameplay perspective to take a closer look at the “intended” combat ruleset of the original game.

As such I have been spending some time between working on my Megadungeon looking over the charts and tables at the back of the book, and I’ve taken the liberty of parsing out a set of algorithms to use to determine the necessary combat odds.  To understand the following, you need to know the basic system.  Whenever a group of men (stands of 20 in the original mass combat system) attacks another, a number of dice is rolled based on the number of men attacking and the types of the opposing units (expressed hereafter as x/y, where x is the number of dice rolled per y men attacking), with each die needing to fall within a certain range to score a kill (or a hit dealing 1d6 damage in D&D). So if eight men are attacking, with a 1/2 ratio, needing a 5 or 6 to kill, 4 dice would be rolled, and every 5-6 would score a kill or a hit (meaning, on average, 1.3 hits will be scored in such an attack).

Now I’m not entirely sure if the actual ratios and ranges required can be reproduced if there is no clear algorithm, so I will simply present the algorithms I have found that underlie the original system and note where there are exceptions.  For my purposes, there are three levels of foot soldiers and three levels of cavalry, light, medium, and heavy, which is a slight deviation from the original terminology.  Here is what I have found:

There are four basic strengths of unit, ranking from A to D, D being the worst and A being the strongest:

• D includes only light foot
• C includes medium foot and light cavalry
• B includes heavy foot and medium cavalry
• A includes only heavy cavalry

Based on this, mounting a footman on a proper horse appears to improve its combat capability by one step.

To determine the ratio range of rolls required to kill, a few guidelines are:

• To kill a unit within the same class or a higher class, a 6 is needed
• For each difference in steps between the attacker and the defender, when the attacker is stronger, the kill-range is increased by one (so a heavy foot attacking a light foot would require 4-6 on a d6 to kill)
• Foot attacking cavalry always requires a 6 to kill
• Heavy cavalry is an exception in the original rules, in that in regards to the range it is underpowered compared to the algorithm

To determine the x/y ratio, the following guidelines can be used:

• Foot cannot have a higher ratio than 1/1
• Within a category (that is, foot attacking foot or cavalry attacking cavalry) every positive difference in step increases the y value by 1
• For foot attacking cavalry, if the cavalry is in the same class or higher, y is increased by 1
• For cavalry attacking foot, if the cavalry is in the same class or higher, x is doubled
• Light foot are an exception in that they are slightly overpowered against cavalry compared to the algorithm
• Heavy cavalry is an exception in that it is slightly overpowered compared to the algorithm

For use in D&D, each character can be assigned a Fighting Capability based on class and level (you can use the values given in the original booklets, or use some other system, possibly using the hit dice proposed in the Swords & Wizardry White Box with bonuses being applied to only one die) which indicates how many men he or she represents using this algorithm.  Whether they are light, medium, or heavy foot, can be based on their equipment at the purview of the GM.  There are other, better thought-out guides to using Chainmail with OD&D, but as those mostly require the original rules of Chainmail to use, this guide can supplement or supplant those tables.

I hope the above was clear enough to read. They are basically transposed from my own notes, which were written pretty much for my own reference. I have yet to closely examine the man-to-man or fantasy combat rules.  Ultimately I hope to come up with a separate combat system to plug into the original D&D rules (or their simulacra) based on concepts in the Chainmail ruleset that I can then release for others to use.  I’m not sure yet whether I want to use the Chainmail system or the now-standard “Alternative Combat System” for my OD&D game.  I should probably get on that, as it begins tomorrow.

Blogs of the Round Table: Denouement

Denouement, besides being one of the rare words to not return pornography on page 12 of a Google Image search (although that status may have been lost by the time of your reading this, so proceed with caution), is also one of the many things that video games do poorly.  This month’s topic for Blogs of the Round Table is also probably the worst one for me to write a post about, especially as my first entry into the discussion.  To quote the website:

This is the closest I could get to a fitting image

How can the denouement be incorporated into gameplay? In literary forms, it is most often the events that take place after the plot’s climax that form your lasting opinion of the story. A well constructed denouement acts almost as a payoff, where protagonists and antagonists alike realize and adjust to the consequences of their actions. Serial media often ignored the denouement in favor of the cliffhanger, in order to entice viewers to return. Television has further diluted the denouement by turning it into a quick resolution that tidily fits into the time after the final commercial break.

But the denouement is most neglected in video games where it is often relegated to a short congratulatory cut scene, or at most–a slide show of consequences. This month’s topic challenges you to explore how the denouement can be expressed as gameplay.

I really should be doing schoolwork right now…

How the denouement is factored into game design is highly dependent on the type of game we are working with.  For our purposes there are three big categories of games, those for which story is a major factor (“Story Games” from here on out, e.g. Final Fantasy, Elder Scrolls, Fallout), those for which story is there to support the gameplay (“Action Games” for lack of a better term, e.g. Halo, Gears of War, Command & Conquer), and those for which story is either non-existent, or a purely emergent part of gameplay (“Crunch Games”, e.g. sports games, Civilization, Total War).

Currently, we mostly see denouement in the story games, for the obvious reason that they are trying to tell a story, and thus conforming to traditional story-telling techniques.  More often than not, however, we see a cutscene at the end, or an opening up of a freeform world which is left relatively unchanged from prior to the climax.  In a book or a film, the denouement is where we see the plot resolved, where the characters get their rewards, and where we see the world as affected by the events of the story.  Limiting the big payoff to a non-interactive cutscene or freedom to roam the countryside takes away a lot of the emotional impact of the denouement.

Implementing denouement into gameplay for a story game should be relatively easy.  Most story games have some sort of conversation system.  Gameplay does not really need to be affected in any way to bring denouement into the picture, simply change the gameworld as appropriate (raze cities, bring magic back to the Mystical Forest of Lor Leliel, bring the guards back from the front lines to the cities, whatever), and make sure that NPCs can describe or comment on the changes substantially.  The reason this is not done very effectively now is because it involves creation of content, which is already difficult enough with today’s graphics systems.  If the story calls for a town to be destroyed, nobody wants to pay an artist to create a new, destroyed version of the town, as well as the writer to rescript several dozen dialogue trees to take into account the change.  In this case, either the story would be rewritten, the changes would be made much less drastic in game than portrayed in the story, or the game will simply end in a cut scene.  Really, there’s not much that can be done to improve the denouement of a story game that is not being done already on a much cheaper scale.

“Action Games” (and I will use the term in quotes to distinguish from the recognized genre) are a different beast.  I don’t know about you, but I don’t play Halo for the deep and involving story (actually I don’t play Halo at all, but that’s another post).  Short of implementing new gameplay mechanics, there really isn’t much option for expanding the story beyond the thrilling conclusion.  Once you destroy Halo (I’m speculating here, I assume this is how the first game ends), there’s not much else for you to do gameplay-wise.  I suppose one could walk around the mothership or start training new recruits, but very few people are going to want to keep playing if the game becomes “talk to the officers, they will congratulate you and let you know how Earth is doing.”  For “action games,” cutscenes are appropriate.

Crunch games are another beast entirely.  I use the term here mainly to refer to turn-based-strategy games and sports games, although I’m sure there are other types of game in the category.  Take for example the Madden games.  The biggest draw to these games for me has always been the Franchise mode.  There is no overarching story here, other than “you are coaching a football team,” but through gameplay, wins and losses, trades and recruits, the player builds the story of his coach.  Similarly, for TBS games, there is very little story to begin with, but the player creates the story of his empire through victories on the battlefield (or techs researched, city improvements, what have you).  With sports games, there is no real denouement possible beyond “you have retired, and now you are a sportscaster” or “you have died.”

For a strategy game, however, I often find that the game ends as things become interesting.  Once I build my perfect empire, what happens next?  With my enemies crushed or allied to me, what do I do with my military?  Do I maintain the troops, but use them as a purely defensive force?  Do I occasionally send forces in to keep the other countries in check?  What kind of economy does my country support now that I am the only global superpower?  What exactly does a “space race” victory entail?  For large scale 4X games like this, there is a lot of unexplored territory for what happens next.  Perhaps it is beyond the scope of a game like Civilization or Rome: Total War to portray not only the rise, but the fall of a glorious empire.  A game where great powers arise and then stagnate from the inside before falling to outside invasion, or descend into civil war, would be extremely satisfying to the history buff in me, and would bring closure to a game which would normally end with total conquest and immediate victory.

I don’t know of any video games that do this successfully, but if one exists, it could be my favorite video game of all time except for Portal.

Please visit the Blog of the Round Table’s main hall for links to the rest of this month’s entries.