For Chainmail missile fire, you batch up the firing troops in blocks of up to 10 or 20, roll a d6, and check a table for one of two possible results (or sometimes only one possible result regardless of the d6 roll). There are three armor classes: unarmored, half armor, and full armor -- which we might broadly correlate to leather, chain, and plate in the D&D system that came later. As shown in the table above, taken as an average, and rounding to the nearest d6 pip, the expected kills per figure firing is appreciably close to 3/6 (50%) vs. unarmored, 2/6 (33%) vs. half armor, and 1/6 (16%) vs. full armor. That's actually identical to what I set in Book of War, prior to ever doing this analysis, so that's great. (Note that Chainmail has no range modifiers for mass missile fire, which I now think is a realistic choice.)
So far so good. Now let's look at the melee combat chances.
Chainmail mass combat has a separate, unique chart for every troop type attacking vs. every other troop type. There are three foot types (light, heavy, and armored), and three horse types (light, medium, and heavy). There are also special modifiers for pikes that I'm ignoring here. I would think that the three weight-classes correlate with the three armor-classes above in missile combat, but I've recently seen original players giving mixed signals about that. In the book, attack levels are given in the fashion of "1 die per two men, 6 kills", or "4 dice per man, 5, 6 kills".
Now, looking at the table above, the results here are not aligned so well with either by-the-book D&D or Book of War (BOW); the attacker weight classes have significantly different chances to hit and kill, apparently reflecting overall discipline and density, even if we would think they are all "normal men" or something close to it. Let's assume that footmen can take 1 hit, and horsemen 2 hits, before dying, as we do in BOW. Then the Armored Footmen land hits on average about the same as missiles do: about 1, 2, or 3 chances in 6, vs. armored, heavy, or light types. As we've said, that's also approximately what we see in standard D&D or BOW for normal or 1st-level men. But Heavy Foot only land hits about half as often, and Light Foot only about one-third as often. In other words, by collapsing the attack proficiency classes, D&D and BOW give about triple hits to Light Foot, and double hits to Heavy Foot, as compared to Chainmail (noted in the last column in the table above).
That's not something I would expect to "fix" in BOW at any point. Note that this shift seems also to have been acceptable to Gygax in his later Swords & Spells supplement for D&D, which does the exact same thing (following suit from the core D&D rule). There is a small percentage adjustment to damage by troop classifications in the Melee Bonuses and Penalties on p. 24: ±10-20% for troops that are non-Regular (Guards, Elite Guards, Levies, or Peasants). But when one starts with only a 1-3 in 6 chance to hit, those percentages never amount to a whole pip of difference on a d6.
As a final thought experiment, let's see how the difference in attack types from Chainmail would appear when up-converted to D&D. Let's say "Armored Foot" is equivalent to a full 1st-level, Veteran Fighter, with (as seen above) about a 50% chance to land a hit on an unarmored/light foot opponent, and so forth. Then the "Heavy Foot" type is somehow restrained from landing hits, about 25% less likely, i.e., 5 pips in 20; so a -5 penalty, or effectively Fighter level -4? And the "Light Foot" is deficient by about 40%, or 8 pips in 20; so fights at -8, or Fighter level -7? Or we could turn it around, assert some sort of reduction for fighting en masse generally, and say Light Foot is otherwise fighting as Normal Men, Heavy Foot as 3rd-level Fighters, and Armored Foot as 8th-level fighters. Something like that. Interesting, but probably not something we want to port into our D&D games.
Git yer speadsheet here.
Does it make sense to call any figure a “fighter -7”? There’s no analog in d&d at 1:1. Not by a very large margin.
ReplyDeleteSuggestion: make one figure (20 men) light foot equal to NM, heavy foot equal to Veterans and armored foot equal to Heroes.
Then you can give defensive values of 7/5/3 and use the normal attack tables. Each hit deals 1 and horse have 2 hits to kill.
IMO 10 men to a figure better but YMMV.
No, "fighter -7" certainly wouldn't make sense. I think I was trying to highlight how the probabilities in the two systems are essentially incompatible; maybe I should have made that point more clearly.
DeleteMy assumption was that the differences in fighting capability partially reflected how many men out of the 1:20 figure were actually fighting effectively. So for a group of Armored Foot with hand weapons, 10 of the 20 men can get into the thick of things, with the other 10 providing backup but not able to effectively contribute much lethal force - unless armed with halberds or pikes, in which case the second rank is fighting as well and the figure gets to roll two attack dice instead of one. On the other hand, HF and especially LF could be construed as being more cautious due to their lesser armor or lack of armor and therefore fighting below maximum potential. Being primarily a category used for peasant conscripts and archers, LF can additionally be reasonably supposed to lack enthusiasm for melee and be less aggressive for that reason as well.
ReplyDeleteI think weaponry also plays into it; LF are likely assumed to only have a single weapon, either a dagger or a one-handed sword. Meanwhile, most of the troop types described as HF fought with spear and shield, which has significant offensive and defensive advantages against LF, as well one or two backup weapons chosen from among swords, maces, battle axes, long dagger, and/or hurled missiles. Meanwhile, AF would historically be using two-handed weapons by default, poleaxes or bastard swords for dismounted knights, with a greater variety seen among the condottiere types. Then they would additionally have a couple of backup weapons. Their plate armor generally negates HF's previously-mentioned spear-and-shield advantage, so whereas 1 HF can go toe-to-toe with 2 LF, the HF only performs 50% better than LF when pitted against AF.
I agree that's the most likely defense of the difference. However, if you correlate the difference in weapons in standard D&D terms (maybe AD&D modifiers), they don't make as much difference as we see in the Chainmail tables. And the troop-density issue seems like it would be a wash on either side (if anything, light troops as seen in Swords & Spells get more room to fit more people into the combat).
DeleteSide note: Assuming a man takes up 3' and the infantry base is 3/4" (20mm), then actually only 7 men abreast can fit there, so 3 ranks? (3/4 in × 30ft/in × 1 man/3 ft = 7.5 men).
Would the bases only be 20 mm in CM as written?
DeleteI know that WH used that size extensively, but Citadel miniatures are either 25 mm scale for the older ones, or 28 mm scale later on - and even in that scale, minis representing beefy types like chaos warriors, lizardmen, or orks use 25 mm square bases for fantasy, or 32 mm round bases for WH40K.
Meanwhile, CM states that it was written with 40 mm scale miniatures in mind. I've never owned any minis of that scale, but given my experience with 25/28 mm scale minis, I assumed the bases would be at least 25 mm square, if not 30.
I'm guessing based on the specifications in Swords & Spells. Now (a) Chainmail gives figure heights but not bases, while (b) Swords & Spells specifies bases but not figure heights. But I'm broadly guessing that Gygax used the same miniatures for both. S&S actually has even smaller bases, starting with 5/8" (i.e., 15mm) for basic men/orcs/dwarves, then 3/4" (20mm) gnolls, battle axes, etc., and on up. I could be wrong but that's my best stab at it.
DeleteAlso I'm kind of averaging that 30mm figures are actually in use because (a) it's between the 40mm and 25mm given on Chainmail's first page, and (b) it's the first thing given in the Fantasy section. If true, that would pretty close to Warhammer scale figures.
I'm wary of using Swords & Spells as a guide because it uses 1:10 figure ratio, and the beginning of CM suggests the 1:10 figure ratio only if figures smaller than 30 mm are used. EGG keeps the ground scale the same between the two, at 1" = 10 yards, so my presumption is that he intends for figure basing and frontage to be smaller when using a 1:10 troop ratio in order to maintain parity of scale - though he does admit in S&S that he can only control the width, while "the scale of the game is such that the depth of the stand must simply remain a function of the casting itself."
DeleteI also find it interesting that Warhammer 1st Edition also has a 1-in-6 chance of scoring a kill in a melee between two 'Light Foot' types, though this is split into multiple rolls. Note that while Warhammer doesn't have the same overarching troop categories as Chainmail, I'm using typical archers with Weaponskill 2, Strength 2, and Toughness Grade B as a proxy for CM-style LF.
ReplyDeleteIn later editions, the hit tables would be changed resulting in a 25% chance of scoring a kill in such a melee between archers, though that is still closer to 1-in-6 than it is to the D&D-inspired 3-in-6. Better troops will also have advantages in killing power, generally being assigned 3 or even 4 Weaponskill, potentially a higher Strength and/or Initiative as well, and bonuses for the weapons being used. The benefits are much smaller than the difference between LF and AF in Chainmail, but simultaneously much larger than the difference between a Normal Man and a 1st-level Fighting Man in D&D. Later editions would further increase the complexity of equipment rules, but still never reaching the level of advantage seen in CM.
Also, just an oddity that's always bothered me, Warhammer treats halberds like they're some kind of middle ground between hand weapons and great weapons (i.e., zweihander swords, bardiches, sparth axes, and the like) rather than treating them as highly-developed and versatile spears. Warhammer also has the quirk of de-emphasizing armor compared to what we see in the works of Gygax, with plate & shield reducing kills by 50% and plate armor alone reducing them by 33%, whereas both CM and D&D and pretty consistent in that plate armor reduces casualties/damage taken by about two-thirds.
Now that I go through this, it also occurs to me that perhaps you should generally reduce the kills by about half if you decide to make any revisions to your Book of War, whether by altering the odds or by adding a 'To Kill' roll. I say this because you based it off of the attack matrices in D&D, but if you take D&D rules to their ultimate conclusion than you'd come up with the fact that on a successful hit, dealing 1d6 damage against an man with 1d6 hit points, you only have a 7-in-12 chance of killing that enemy. This drops to 5-in-12 if you assume that for the most part, anyone with only 1 or 2 hit points isn't tough enough to be a soldier and that any trained/regular troops will have at least 3 hit points per man. Either way, I think it's close enough to round it off to a 50% of surviving the hit.
Good thoughts, thanks for the insight on WarHammer. Thanks also for thinking about a possible BOW revision (which I'm always kind of noodling on).
DeleteThat said, keep in mind that one die-roll in BOW represents time for a sequence of 3 separate man-to-man scale attacks. I did a simulation to set that mechanic at a level that it does in fact match the statistics for expected casualties in D&D. (That was in fact the top prime directive of that whole project.)
Hm, that's true about the multiple rounds of combat - I think where the difference is, is that both CM and WH check morale after each round of melee. If one side is significantly superior to the other, it's not uncommon in either system to see a retreat after only one round of melee - and almost assured if a flanking charge is involved.
DeleteSorry for the repeat comments - I keep thinking of more things! It also occurs to me that if you use the rule that a Fighting Man receives one attack per level against Normal Men, then Armored Foot only need to be 2nd or 3rd level in order to achieve the power differential seen in Chainmail between them and Light Foot. I'll also note that in Moldvay D&D, the 'standard' noble in the monster list is assumed to be a 3rd level fighter, with a squire who is a 2nd level fighter and up to 10 retainers who are 1st level fighters (are these men Heavy Foot?)
ReplyDeleteVarious AD&D sources would also tend to assign fighter levels between 2nd and 4th to generic knights.
That's another intriguing observation. :-) Admittedly I tend to gloss over that rule because (a) I'm not find of the jump-discontinuity from <1HD to 1HD, and (b) I feel that it's "double-dipping" in D&D to get both more attacks and increased hit chances (as opposed to CM heroes who just get the former).
DeleteTrue that it is double-dipping, though at lower levels the impact is minimal - by the book, a fighter doesn't even get increased hit chances until 4th level.
DeleteAlso, I like how it jives with the anecdotes from old-timers that they never used Hero figures in mass combat; if the 19 Armored Foot are all 3rd level fighters anyhow, then it makes sense that having a 4th level fighter leading them wouldn't appreciably change their combat abilities (at least, no change above and beyond the standard bonus given for having a commander present).
It also clears up the questions a lot of people have with the Fantasy Combat Table lacking entries for fighters of less than 4th level by drawing a line in the sand, that if you're only 3rd level then you're just an Armored Footman and unable to participate in fantastic combat.
Well, on the other hand, the pre-D&D Guidon draft shows the initial d20 system was set up formulaically so that each fighter level would improved hits by one pip. My interpretation of the OD&D combat tables is that was just an excerpt of what they "should" have been to fit in limited page space (see also AD&D DMG and "5% Principle" article that explicated that later).
DeleteNow, it's really interesting that your experience is that old-timers say the didn't use heroes in mass combat -- my experience is just the opposite, that the earliest players kick back at me pretty viciously with "we didn't care about that" arguments when I point out (citing EGG) that it would be inappropriate to do so. (e.g., that its perfectly fine to use a Hero in mass combat and have him cutting down 80 men per turn) Maybe we visit different sites? (Eager to learn where!)
Honestly, I don't think I visit anyplace different from you. I unfortunately never kept track of specific threads or citations, but the impression I came away with from all the Chainmail threads on ODD74 and Dragonsfoot was that the fantasy supplement was more-or-less always used with Man-To-Man rules, despite the rules providing troop classes according to the mass combat rules.
DeleteThat's definitely what Gygax always said (and it makes sense), but the guys who played with him have a conniption whenever I point that out there.
DeleteIs it actually "the guys who played with him" or is it just one particular guy? Because I can think of one person on those boards who regularly has conniptions whenever anyone asks a question about rules that are "totally obvious in play" or otherwise implies that there might be a prescribed way to referee beyond "use your own judgment."
ReplyDeleteWell, you might have a stellar point there! :-) M might be frustrating/taking up more mental space he deserves there. But he seems to have affected a lot of other people's opinions, and I consider it the top controversial thing I've said on this blog in the past, so I've got some scar tissue on the issue.
Delete