I'm just kicking around ideas.  Please dear God no essays on theory, the definition of "realism," existentialism, etc.  And I don't care about being innovative, at ALL.  I just want to throw around some mechanics and see what works for me.

That said, in a nutshell my goal is to come up with a combat system that has a realistic feel, grim and gritty, without being plodding or too complex.  Some complexity is okay, because I want combat to be quick and deadly, and thus relatively rare (i.e., it might take longer to run a round but there will be far fewer exchanges than a D&D combat, and combat will be rarer since players will tend to avoid the risk).

I'm thinking I'll drop this thing into a somewhat heavily modified WFRP2 setting, but that might change in future.  And I'm going to keep TRoS (and any other similarly historically and reality minded RPG brought to my attention) somewhere in the back of my mind at all times.

Goals:

Only Fools Rush In

The cat-and-mouse element of "duels."  I lost the GD-ed German longsword fencing DVD I bought years ago, but I remember the C&M bit, where "duelists" (i.e., 1 on 1 as opposed to full on group melee) circled, angled for position, blended from stance to stance looking for advantage, etc.  This is cool and allows a system to somewhat mitigate the deadliness of realism and the fun of unrealism; combatants aren't actually hacking at one another so there's nothing unrealistic about the protraction, but players get a bit of extra "beef" in combat too.  So, I want to do something in the mechanics to reflect this.

He Who Dares, Wins

That said, it seems pretty well established amongst ARMA types that the attacker has the advantage.  This makes sense, even if I can only back it up on an abstract level, or with analogies (football teams practically always take the kickoff when they win the toss, presumably for good reason - they get to control how the game starts, get the initiative vis-a-vis scoring, etc)

Armor

It's there for a reason, and it's much more important than in RPGs, in my experience.  An experienced fighter in good plate is a Man's Job, and I want that reflected.  On the other hand, weapons vs. armor was an arms race, so the balance is probably going to be very tricky.  I'd appreciate anything anyone can add to the info found in this thread (including calling BS on info therein):

http://www.myarmoury.com/talk/viewtopic.php?t=4427

So far this is my armor list:

Soft Leather
Hard Leather (Cuir Bolli)
Padded/Gambeson/Arming Jacket
Scale
Maille (might need to differentiate maille quality)(Hauberk, Byrnie, etc.)
Lamellar/Klibanion (Byzantine Lamellar, often worn over Mail)
Brigandine/Coat-of-Plates
Plate

I think I have the order of preference/efficacy right.  Anyone want to take a stab at quantifying the distances?  E.g., rate them on a scale of 1-10 (or whatever) vs. bludgeoning, piercing, cutting?

I think I like TRoS' stance that all attacks are "called shots," which dovetails nicely with my desire to have armor at the per-body-part level.

Basic vs. Advanced Combat

This is even less thought-out than the above, but I was thinking maybe of having Basic and Advanced combat.  Basic would serve in the pell-mell of melee (can't get too fancy or you'll hit your mates, no room to spin in a narrow dungeon corridor, etc.), and even in more focused "duels" for those who don't want to use Advanced, whereas Advanced combat would be for more dramatic or important engagements by those who want to use it.  This could be as simple as using optional stuff like maneuvers, stances, repartee, murder blows, etc., in Advanced but not Basic combat.

Think!

Obviously, a realistic approach favors skullduggery, planning, etc.  I don't think this needs mechanics, per se, but it's worth mentioning.

Bluff & Blind

There should be some element of bluff & blindness; fighters can fake weakness, and there's always an element of the unknown (is he bluffing or vulnerable?).

I'd like stances to be incorporated somehow.  The actual stances would be nice, but more abstract would be okay, too.

I'm thinking whatever stat I use for reflexes could determine initiative, without any random role added.  Any downsides to this?

Stuff I have to incorporate, i.e., factors:

Weapon reach/length
Damage Type (Cutting (Slashing, Chopping), Piercing (Thrusting, Stabbing), Bludgeon)
Room (space required to wield effectively; a better word than "room" is welcome)
Penetration
Weight
Balance

Attack type (thrust, slash, haymaker, murder stroke (pommel), etc.)
Attack Location (called shots)
Counterattack
Parry
Dodge
Riposte
Feint
(other fencing-type stuff I know nothing about; suggestions welcome)
Style (Sword & Shield, Sword & Buckler, Longsword Fencing, Etc.)

Damage
Fatigue
Fumbles (forced  (sweat in eyes or on weapon handle, slippery floor, etc.) or unforced (clumsiness, morale failure, etc.))

Okay, here's the first idea that popped into my head.  Use a bell curve roll (3d6 or 3d10), add modifiers, subtract defender's modified bell curve roll, result is the degree of success.  If negative, defender has seized initiative for next round (can add it to initiative, or to attack roll, or something - heh).  If positive, attacker has successfully hit defender - value is added to damage number for weapon, modifiers, then armor value is subtracted.  Degree of success might carry over to next round, somehow.

Any big stumbling blocks involved with something like this?  Doesn't seem terribly innovative so I'm sure there's some preexisting thinking from which I can benefit.

As for the cat-and-mouse thing, I was thinking that it could be pretty similar to the mechanic above, except benefits are transitory so Carpe Diem is encouraged.  E.g., combatants determine initiative, make "attack" and "defense" rolls, degree of success is added to attacker's strike next round, or lost if no attack is made.

Sorry this is all so vague, I've just started actually working on this, rather than thinking about working on it as I've been doing for a long time.  All help is greatly appreciated.

This is remarkably similar to one of my old combat systems.  Well, that's not that remarkable, really, as "realism" (by which I also meant gritty and hard-hitting) was my goal with it as well, and so was making combat rare by making it high-risk.

Now, I should probably mention that there's a reason it's one of my old combat systems.  In practice, it became very, very dry and tactical, quite a lot like a chess match.  This was bad because it wasn't gritty and hard-hitting (we were buffered against the violence by several layers of tactics), and also because combat became kindof divorced from what was going on at the moment (that is, when combat happened, we shifted gears from "stuff happening with the characters" to "tactical combat" mode).  How weird that, the more detailed and layered I made the system, the more abstracted our relationship, as players, to the violence and mise-en-scene was.  (Edgar Allan Poe once made an astounding point about how checkers requires more analytical prowess than chess, and that it's easy to mistake the complex for the profound.  I think he made this point somewhere in the "Purloined Letter" if my memory serves)

I realized that what I wanted was a system that would deliver the Violence and the Grittiness of a highly "realistic" combat system, with all the detailed injuries and blood loss and shock and damage to armor and weapons, reliably.  Most of the time, the highly detailed modeling got there eventually, but sometimes it went on for too long before anything 'happened," and up until that point it was a bland, flavorless tactical exercise.  I realized that I wanted to skip past all that ARMA and Shadowrun stuff and get straight to the arterial spray, compound fractures, and loss of life and limb, so all that could feed back directly into "what's happening," instead of a bunch of sound-and-fury rolls and chart look-ups that required several repeat cycles to get to the result that I wanted and expected. 

See, for some reason, I thought that "what I wanted to happen" had to be justified by a rational, realistic modeling process, with lots of dice and charts. It hadn't occurred to me that I could make the system such that it reliably generated the expected results without a bunch of unnecessary steps (ever had to ignore a roll's result because it "messed up" what was going on?  In such cases, the system has kicked something at us that doesn't conform to our genre expectations, our expectations of what's supposed to happen in-game.  For my money, a system that does that is a BAD SYSTEM).  So I cooked up a system where the terrible violence expected nearly always happens, with rules that focus only on that happening, not bothering to model things that ultimately don't make a big deal by themselves (oh, there was so much data that I was able to cull without feeling bad about it).

I'm not saying that's what you're doing; I'm just suggesting that you think about it.

Now, other than that, I have to ask if adding modifiers to a bell curve is a good idea.  Wouldn't that tend to result in awfully binary results?  I mean, the dice will tend to come up near the median, and then it wouldn't take too much of a positive or negative modifier to knock it into the extreme ranges, so you'd have to make the modifiers very small, but then that seems like a lot of fuss about nothing much.
Unless maybe the modifiers are based on proportion, instead of being additive?

-Marshall

Hi!
  OK, I think you do need to explain what you mean by realism. But, here are things I have seen in tactical games that are not realistic:
1) HP - I know its supposed to be an abstraction, etc. blah blah, its unrealistic. I recommend a DMG track
2) AC makes you harder to hit - This seems counter intuitive to me. In reality if you wear heavy armor, you will move slower and be easier to hit, but it is likely that such a hit will not be as life threatening...
3) Multiple rounds of combat - The reality is any weapon (including the lowly dagger) is dangerous enough to take you out of a fight. Pretty much the first guy who lands a solid blow wins, right?
4) Death Spirals - Death Spirals are realistic. They may not be fun, but them's the breaks.
5) Separate hit and damage rolls - Rolling a critical hit and then doing 2 points of damage both sucks and blows. Its not realistic that you hit them in a spectacular way in a critical location, and then somehow didn't do any meaningful damage.
6) Knocking people out - Its easier to knock someone out then kill them. And once they are down, you can kill them at your leisure if you are that bloodthirsty. Any game that makes it harder to knock someone out than it is to kill them is unrealistic
7) Wrestling - Why is wrestling so hard in these games. And why is it usually the worst choice? If a guy has a sword or knife, its better to be grappling him than to be 3 feet away. I mean it is even better to be outside of running/attack distance, but if your choices are 3 feet or grappling, you know what you would do...
8) Bleeding - Many times blood loss will kill you long before pierced intestine will. Now that's realistic.
9) Infection - I get an infection every time I get out of surgery. Something tells me a barbarians sword is less clean...
  Well, you get the idea. I think you should play 5 or 10 rounds of practice with the math you prefer and see if its really fun. Run the numbers a little, then do a one-on-one combat with someone whose opinion you trust. See if it is fun. I started with a more realistic combat and the fights were over too fast. SO, I decided to dial it down...
  You might like it, not saying you won't. Just definitely run the numbers and then play straight battles (nothing else, no RP) at least like 5 or 10 of them. And listen to what people are saying when they critique them, don't argue away their view point, k?

I'm not saying that's what you're doing; I'm just suggesting that you think about it.

1) HP - I know its supposed to be an abstraction, etc. blah blah, its unrealistic. I recommend a DMG track

2) AC makes you harder to hit - This seems counter intuitive to me. In reality if you wear heavy armor, you will move slower and be easier to hit, but it is likely that such a hit will not be as life threatening...

3) Multiple rounds of combat - The reality is any weapon (including the lowly dagger) is dangerous enough to take you out of a fight. Pretty much the first guy who lands a solid blow wins, right?

4) Death Spirals - Death Spirals are realistic. They may not be fun, but them's the breaks.

5) Separate hit and damage rolls - Rolling a critical hit and then doing 2 points of damage both sucks and blows. Its not realistic that you hit them in a spectacular way in a critical location, and then somehow didn't do any meaningful damage.

6) Knocking people out - Its easier to knock someone out then kill them. And once they are down, you can kill them at your leisure if you are that bloodthirsty. Any game that makes it harder to knock someone out than it is to kill them is unrealistic

7) Wrestling - Why is wrestling so hard in these games. And why is it usually the worst choice? If a guy has a sword or knife, its better to be grappling him than to be 3 feet away. I mean it is even better to be outside of running/attack distance, but if your choices are 3 feet or grappling, you know what you would do...

8) Bleeding - Many times blood loss will kill you long before pierced intestine will. Now that's realistic.

9) Infection - I get an infection every time I get out of surgery. Something tells me a barbarians sword is less clean...

And listen to what people are saying when they critique them, don't argue away their view point, k?

Marshall wrote:

Now, other than that, I have to ask if adding modifiers to a bell curve is a good idea.  Wouldn't that tend to result in awfully binary results?  I mean, the dice will tend to come up near the median, and then it wouldn't take too much of a positive or negative modifier to knock it into the extreme ranges, so you'd have to make the modifiers very small, but then that seems like a lot of fuss about nothing much.
Unless maybe the modifiers are based on proportion, instead of being additive?

Anyone know of a freely available skills web I could use?  One of those things that shows skill overlap (train in longbow and become naturally more proficient with the short bow, train in tracking and get a boost to hunting, etc.)?

More generally, it'd be nice to see a thoughtful breakdown of 14-15th century skills since I doubt I'll be needing to come up with anything revolutionary or reinvent the wheel.

David Artman came up with a targeting grid in a First Thoughts thread a few months ago that you might wanna check out.  I can't rmember the title, though.  If a search doesn't turn it up, you could PM him.

Your armor types sound good to me (though I know nothing about lamellar), except that there's no historical record of coeur boulli being used for armor (at least not that I can find on the internet).  SCA folks can vouch for the fact that the stuff works, and it seems the means were available back then -- so hard leather armor could have been worn.  Was it?  My best guess is no.  Maybe everyone who could have afforded it could also afford mail?  I dunno.

Here's my game's armor table:

[table][tr][td]Armor[/td][td]penalty[/td][td]Dice[/td][td]Difficulty[/td][td][/td][td][/td][td][/td][/tr][tr][td][/td][td][/td][td][/td][td]Edged[/td][td]Weight[/td][td]Pierce[/td][td]XBow[/td][/tr][tr][td]Soft Leather[/td][td]  none[/td][td]1[/td][td]4+[/td][td]5+[/td][td]4+[/td][td]6+[/td][/tr][tr][td]Padded[/td][td]  none[/td][td]1[/td][td]3+[/td][td]5+[/td][td]4+[/td][td]6+[/td][/tr][tr][td]Mail[/td][td]-1 agility[/td][td]3[/td][td]4+[/td][td]5+[/td][td]4+[/td][td]6+[/td][/tr][tr][td]Legionnaire[/td][td]-1 agility[/td][td]3[/td][td]4+[/td][td]5+[/td][td]4+[/td][td]5+[/td][/tr][tr][td]Plate[/td][td]-2 agility[/td][td]4[/td][td]2+[/td][td]5+[/td][td]2+[/td][td]4+[/td][/tr][/table]

I want a free skill web too!  Someone hook us up!

There was a report on coeur bouilli in a Royal Armouries Year book and a chap making a helm out of it.
Personally it would be easier to get than mail, but less likely to survive.
I'll see if I can find the report.

JW

Thom Richardson, Royal Armouries keeper of Armour, quotes that many inventories of the late 13th and earlt 14thcenturiesw list its use and that a mid 14th century upper arm defence survives in the British museum. There is also a surviving piece of CB horse armour in the Royal Armouries itself.
JW

David wrote:
David Artman came up with a targeting grid in a First Thoughts thread a few months ago that you might wanna check out.  I can't rmember the title, though.  If a search doesn't turn it up, you could PM him.

Here's my game's armor table:

Hereward wrote:
Thom Richardson, Royal Armouries keeper of Armour, quotes that many inventories of the late 13th and earlt 14thcenturiesw list its use and that a mid 14th century upper arm defence survives in the British museum. There is also a surviving piece of CB horse armour in the Royal Armouries itself.
JW

Jay,
  I think that the gaff to creating a realistic system that is playable is to combine as much related information as possible. A good example of this is how handling is managed in Car Wars. Its a very simple mechanic, but when you play, you really feel like their is a physics engine behind it, but it is not hard to use and no one has ever made a computer program for it...
  The reality is, there is not a lot of moving parts to track when it comes to incapacitating someone in a fight. Pretty much, the factors are: blood loss, pain, mortal wounds, right?
  So maybe Blood loss has a time to live stat. and you just use the lowest time to live of all the blood loss wounds you have...
  And pain and mortal wounds (stab to heart, etc) can use a damage track ala CP2020 or Exalted. Then every weapon just has to have a damage stat and a blood loss stat.
  I don't think there is much advantage to keeping track of stats on such a skale that there is a meaningful difference between a 36" blade and a 38" blade or between a 8 pound sword and a 9 pound sword, do you?
  You almost have to look at it as a cost benefit analysis. How much effort will it be for the player to track that and how much extra fun is generated by the extra work?
  RPGs are a social experience, I think adding a computer play aid will detract from that, as there will be more time spent with that aid then there will be time spent with the other players, no?
  And if you want to see a medium crunch system, check mine out...
  Just my two cents, good luck man.

Dave,

You make excellent points.  I think you've got the priorities right; fun is always top priority.  On the other hand, I'd like to have enough granularity available so that if player wants the detail (his idea of fun), it'll be there.  The thing I keep thinking about realism is, "why did they do that"?  E.g., a great example is the armor thing; at some point it became "the thing to do" to suit up in plate, drop the shield, and use both hands for weapons.  They obviously had good reason.  I want to set the rules up so that this choice makes sense.  Not so much to track every inch of a blade as another point on a reach stat, just to make reality reflected in the mechanics.  Obviously there's a bridge too far somewhere, but I want to get the broad strokes.  So players will be rewarded for saving up for plate and a good sword, rather than running around in an arming jacket with a dagger.

I think a lot of complexity can be handled before combat even begins, leaving players to just track a few conditional modifiers, fatigue, bleeding, etc., so they spend more time during character creation, and checking the numbers before a session, and less time fumbling about in combat.

And I'm thinking about the software just in terms of dice rolling (and maybe crit tables), which is a time-saver not a time-consumer if we're talking about expanding bell curves (e.g., 3d20 can take a while to add up, but only takes a moment with something like DiceLab).  But I'd never make it opaque, or required - if players insist on rolling the dice and doing the math, who am I to argue?

I was also thinking yesterday about incorporating some version of a "take 10," where GMs can use an average number instead of rolling and use it for mooks, melees and whatnot.

In terms of "fun vs granularity" trade-offs, this thread about my combat system might provide some good food for thought.

Jay,
  The issue with Take 10 is this: It was designed to take a mechanic that can be tedious (re-rolling a skill check on something that is not critical if it fails) and uses it in a way that it is not intended.
  From what you have said, the number rolled is going to be critical, and if you Take 10, then you allow the damage to be totally predictable and/or preventable.
  So, what will happen is for guards that are a moderate threat Take 10 will get used (and the DMG will be totally predictable), and if it is not strong a threat, Take 10 will never be used (because they have no chance to hit unless its a lucky roll).

  I think we are touching on a key point and I want to draw a little more attention to it. You want to make a "good mechanic" and I agree with your aims. But, you are already considering die roller software for the GM and another mechanic to make it easier to handle more combatants (even though they are weak).
  This tells me that, even in your own mind, your mechanic is too complex. Don't get me wrong, I have made the same mistake myself, I am not accusing you of doing anything I haven't done. But, I am saying that maybe a certain level of abstraction and/or rigidity might help you enforce the things that are important to you.
  Look at it this way, why don't all RP'ers play freeform RPGs. I mean if they did they can play anything they want, right? The reality is that some rigidity helps enforce genre continuity, spark creativity and get players to agree on the focus/style of play. For instance, you can't use ditv to do a dungeon crawl. But most people would say ditv is a good game.
  So, my advice is to look at the stuff that happens in combat and do two things:
1) If it is not fun or not important it, bundle it into an mechanic that is fun/important.
2) If it there is overlap between two elements, then combine them. In other words, if two weapon stats or mechanical elements fulfill the same basic elemental function, then they should be combined.
  Examples of things that might be combined: Piercing and bleeding might be the same mechanics/stats. Initiative, Reach and Speed might be the same stat/mechanic. See what I mean?
  Good luck man!

Most of my roleplaying have happened in Eon, a swedish game with a very detailed 'realistic' system. Fights are never brutish because administration is so slow. Sure it sometimes helps since I creates cool shit like an arrow takes an eye out and sticks and you'll take additional bleeding if you pull it out + getting an infection (different strengths depending on how well kept the arrow was and who's treating the wound) and having to choose between putting on a bandage that might stop the bleeding or surely stop it by burning the wound (with the risk of the patient dying from loss of blood plasma). Usually though, despite all different rules for realism and shit the characters usually just stand in place waving their pointy sticks hoping luck and their superior skills will prevale making super boring.

Something that's important about realism: you can't slash through plate armor. Nope. Crossbows are hardly worth shit either, when the pope whined about crossbows it was back in the days when knights had chain armor.

If you want players to see combat as something scary it should be fast. Like if you charge someone with a halbeard readied they should probably get sliced up in the first round. I had an idea for a super fast combat system: each weapon gets a value which represents how likely it is to inflict a wound. Seeing all weapons Can kill we're only interested in the likelyness of it doing so. So long weapons get a better chance. If you are unarmed and someone attacks you with a dagger, it's not like you're gonna let them stab you. You're gonna keep your distance and your eyes on them so if they do a bad move you'll move into a clinch and grapple them. So give grapple a chance as well, just much lover than weapons. That was you can't rush someone with a spear and start wrestling but if they're focusing on someone else you can.

Now you have something simple like: Skill + roll + weapon value vs skill + roll + weapon value. The difference dictates how bad the wound is. Use a table or something to determine where and how the wound manifests or just say something that makes sense. Like if someone with a sword hit someone 3 it's a pretty bad wound probably meaning bleeding that takes the person out of combat in X rounds or lowers the persons rolls -1 per round (cumulative) since they get dizzy from bloodloss. Also you don't heal 1 HP per week. A snapped bone will trouble you for the rest of your life more or less. A clean kill (say 5 difference or something) doesn't have to be something like cutting someones stomach open, it might as well be hacking an arm off or anything that would make that person lose conciousness or the ability to fight.Add modifiers for closed spaced ect.

A few details regarding plate armor:

As Krippler pointed out, swords are very ineffective against it.  They glance right off the armor's angles and curves (which is the reason those angles and curves are there).  What you need against plate is a piercing weapon that you can get into one of the weaker points (anywhere the surface is concave).  Thrusting with your longsword can work if you've got the skill and the presence of mind, and a spear is good, but your best bet is an estoc or a poniard.

However, axes can mess up plate armor somethin' good.  This is because an axe isn't really about the edge; it's about the weight and the leverage.  Its working principle is closer to the hammer than the sword.  Even if the axe's blade doesn't penetrate, there's still a large amount of connecting force.  Remember, a good longsword weighs under three pounds; the head of a battle axe is going to be at least that much, plus mounted on the end of a shaft for extra leverage (then there's the dreaded poleaxe, wielded by the Anglo-Saxon house-karls, with a spear-like shaft).  The problem with axes is that they leave you vulnerable, due to their weight and also due to the fact that they rarely cut all the way through something:  you usually have to dig the blade back out.

And one more thing:  plate armor is heavy!  I mean HEAVY!  A fully suited-up knight is generally wearing his own weight in total equipment (armor, shield, and weapon); that's why they rode big, big horses.  When a knight gets knocked from his horse and has to go at it on foot, he throws down that shield, holds the sword with both hands, and he still moves pretty darn slowly.  If he misses, he's vulnerable for quite some time before he can get back to a good stance.  My money's on the guy with the arming jacket and something pointy.

Fighters went to the trouble of plate because it was useful; while the guy in the arming jacket is trying to grapple and find the vanishingly small weak spots, the guy in the plate has his choice of where to put his sword into Mr. Arming Jacket.

AFAIK guns killed plate armor.  Until then it was getting more and more sophisticated.

One thing I was wondering though, was how much of the evolution of hand weaponry was part of the arms race vs. armor, and how much of it was due to other factors.  For instance, in the Middle Ages the kit of choice seemed to be maille (probably with padded armor beneath), helm, shield, and sword; by the 15th c (or whenever, I'm not that interested in the dates except as they impact the discussion) the kit of choice seemed to be fairly complex plate armor (or maybe plate over maille), with both hands using a weapon or weapons (zweihander, pick & mace, two picks, polearm, whatever).  The question is, if the Middle Ages types could've had penetrating weapons like picks, would they have preferred them over their swords?  The likely answer seems to be no, because picks & maces aren't exactly highly innovative weapons, so they probably had access to them and preferred their swords.  But, I'm still curious what the ARMA types would say.  I guess I should go ask them.  :)

Back to mechanics, I was just thinking about the bell curve, its width, etc., and it occurred to me that skill has more effects than the obvious; if you have two equally skilled Masters facing off, variables like weapon factors (weight, reach, design, balance, sharpness) become much more salient.  Give one a dagger and the other a longsword, and 9 times out of 10 the former is shishkebab.  This isn't as true of equally skilled Novices, because they don't know what they're doing with their respective weapons, relative to the Masters; the Novice with the dagger has a relatively much better chance against the Novice with the longsword than the Master with the dagger has against the Master with the longsword.

I was wondering if anyone has any suggestions as to how to handle this.  The only thing I came up with was to have some kind of wielder trait limit the weapon factor.  For example (numbers from rear for illustrative purposes), if a weapon has a range bonus of 10, the wielder must have a skill of ten to get the full range bonus.  If a guy with a score of 5 in the same skill he only gets a range bonus of 5 with the same weapon.

Hi!
  I think you are over thinking it. The answer is probably in the scaling. If the dagger gives you a bonus of 2 and a novice has a skill lvl of 1 and a master has a skill level of 10 it will be more useful to a novice, no?

dindenver wrote:
Hi!
  I think you are over thinking it.

The answer is probably in the scaling. If the dagger gives you a bonus of 2 and a novice has a skill lvl of 1 and a master has a skill level of 10 it will be more useful to a novice, no?

Jay,
  Sounds like you are on to something. I think it all comes to scaling the various factors so it matches your model. In this case, if you use 3d20 for the randomizer, than that means that skill levels probably need to go to 100 (maybe 200 or more). In this case, high skill will trump good rolls. Thats what you are saying, right?
  So, I would recommend taking the various factors that you want to model, listing them out and then factor them according to their impact on an individual roll. for instance:
Attacker's Natural ability (5%)
Attacker Skill (20%)
Weapon quality (5%)
Defender's Natural ability (5%)
Defender's skill (5%)
Defender's action taken (Attacking, taking cover, etc) (10%)
Armor quality (20%)
Terrain (Trees, cover, etc) (10%)
Conditions (foggy, dark, etc) (10%)
Random Occurrences (10%)

  So, assuming these numbers are correct (and I am sure you will want to tweak them). If 3d20 is your randomizer than, Stats should be scaled 1-30, Skills should be 0-120, Weapon accuracy/DMG should scale 0-30, Armor should scale 0-120, Terrain should provide bonuses or penalties from -30 to 30, as well as conditional modifiers. As you can see, I would recommend decreasing the die size to keep the scaling within a tighter numeric range. Unless having discreet modifiers is your highest priority.

  As to starting out overwrought and then scaling it back, please reconsider. The issue with that is you get unusual artifacts when you use this process (I know it has happened to me). Either way, good luck man!

A few ponts in no real order.
Plate armour weighs around 70-90lbs all in, plus its spread over the body and made to measure, so it is not as bad as people make out though of course its not a easy to move in as lighter armours it doesn't make you clumsy. Also the horse they rode are not big, they are solid and have short backs but they are not huge. So basically you are looking at someone wearing about the same wieght of kit as a modern soldier does with ammo, weapon and body armour

Armour of this type is expensive, it made to fit and requires good metal and special skills to make it so it is not common. So most opponents will not be wearing it.
But when fighting a mixture of armour types poll axes and war hammers, axes etc are good because they work against al the armours, the best agasint plate is that you bend the laminations and then you can't move.

So why swords? prestige, its an expensive piece of kit and to have one set you apart and as already said, it can be sued against different types of armour.

Also fighting styles deal armour in some ways more than weapons, so if faced with an armoured opponent, you get in close, use hals sord to help thrust to gaps and/or to get in grapple and chuck them to the gound and kill them there.
Best
JW

Jay,
  As you can see, this is a controversial topic. I don't think anyone can agree what realistic is... One person has said that armor is the best survival tactic, another has a guy with a dagger taking the armored guy out...
  So, in the end, two things are true:
1) You are actually modeling what you think is realistic
2) No matter how meticulous your research and modeling, someone will call it bunk.
Try and bear that in mind when coming up with your cool mechanic. Good luck man!

I'd like to make a non-armour related point, thinking instead about the other two goals you stated (quote edited for brevity)...

JQP wrote:
Goals:
Only Fools Rush In
The cat-and-mouse element of "duels."  <snip>

He Who Dares, Wins
That said, it seems pretty well established amongst ARMA types that the attacker has the advantage. <snip>

I don't know if anyone mentioned the combat system in Swashbuckler? Its got some very good ideas which make a good bridge between "reality" and cinematic/descriptive fights. It also requires the use of tactics in maneuvcers and you also design the PCs fighting styles. I know there is a PDF of some house rules which include more Medieval weapons.

Definately worth a look as it had good ideas and has been the inspiration for Lightsabre fighting and space fighter combat.

Best
JW

Perhaps you could work on a system of limiting factors:
Every weapon has a set of negatives and positives, but you must always pick the lowest of the positive value or the persons skill. The negatives still remain, so people will mainly use knives and swords rather than bother with handling the problems of an axe or a spear.

I also like the idea of having weapon qualities that can be bonuses or penalties depending on the use, such as weight effecting the speed of a swing but also allowing more damage. I'm not quite sure how to combine this with the above, but perhaps the "power speed and grace" system could be applied to weapons also, as well as armour. In this way you have a layering of factors that allows someone to blame their tools legitimately!

It also fits my experience of tennis of all things, where using a better bat suddenly allowed me to do much better. There is also an analogy with communication, in that your ability to express your intention will be set by the bottlenecks of expressiveness in the system.

But that doesn't mean that a useless sword makes you hopeless, as you could have a system of adaption, where like in the wild west people would learn the quirks of their guns and so know that they always fire slightly left.

A quirk learning system almost ends up like a magic item identifying system, with people studying the weapon in order to get the most of it. How to mechanically work this? A "quirks" score, that you can adapt to by training, reducing it's modifier for you at least. But the trick is that adaption to one sword puts you at a disadvantage with others, so you could have a set of allowed adaptions, say 3-5, that would act like proficiencies and specialisations, and using a weapon that was not one of them would put you at a disadvantage without training.

So to recap I suggest 3 layers of skill ratings, personal, weapon and armour. This is in contrast with the modifier system that most games run on, except that each piece of relevant equipment has a quirks rating, which is reduced by familiarity. This can be made easier I think by pre-calculating the reduction to weapon quirk when your character attunes themself to it. The adjusted quirk score could be shown with the original in brackets next to it, and the total number of familiar items shown by either a master list or by remembering to remove one when you add another.
I think probably I wouldn't go any further than that, but you could add environmental conditions (like light intensity) to the mix, with the idea that fog is the great leveller, except for those with the ability to adapt.