Wielding a weapon should increase your AC

[AmerginLiath]

Being armed already contributes to defenses: it makes you less likely to be targeted (especially for melee attacks) compared to those without armed capabilities. That’s why the original Armor Class system in the naval war games* that Gygax and others played didn’t include weaponry in that calculation, a ship’s weapons would be a deterrence for attack rather than adding to a calculation of whether a given attack hit and did damage.

———
*The briefest of history lessons for those who don’t know: we have Armor Class, and originally AC that ascended, in D&D, because it was taken from literal classifications of naval armor in war games — Gygax and Arneson worked on “Don’t Give Up the Ship!” before “Dungeons and Dragons” — noting a ship with first-class armor, second-class armor, etc. Dexterity bonuses come into play as a modification of the ship-speed rules (which could lead to glancing blows), but the basic armor rules right down to name always acted as if the character was a ship or fort at humanoid-scale, rolling to see if the latest volley did underlaying structural damage. As a guy who’s worked on historic ships before and examined armored gun decks, this has always amused me.

 

[akr71]

I’m no expert, but it does seem that 2 people fighting with swords would very often use their sword to deflect the other sword.

I think that's where your Dexterity Modifier to AC comes into play.

 

[Tony Vargas]

Many games have rules that punish you attacking an armed opponent when you are unarmed

Sure, including other editions of D&D. In 3e (without the right feat, of course), it provoked an AoO. In 1e, though it was obscure (and I hope I'm remembering it correctly, not some variant), making a pummeling/grappling/overbearing attack against an armed opponent game them a chance to hit you, that, if it hit, ruined your attempt. Unarmed combat like that also progressed in segments, rather than rounds, so it could end very badly for you, very quickly.

but what if we turn it about for d&d?
I definately can't say i've had the experience, but intuition tells me that it would be much harder to attack someone with a knife, let alone a flail or a long sword.
I feel that for melee at least, this should grant a bonus.

It's harder to attack someone armed with a decent weapon, even if you're armed, yourself - they can parry (and make counter-attacks, etc, etc...). It'd make oodles of sense for proficiency with your weapon to apply to AC to some extent in some way - it might make the most sense for it to replace your AC, when the result is higher.

For instance, AC could be your 'in the round' defense, vs all attacks from all quarters. When your engaged in melee, you can calculate a Parry AC based on your skill with your weapon and apply it to one opponent that round, that enemy's melee attacks go vs that defense, all other attacks against your AC. If you are entitled to multiple attacks, you might get to use that defense against more enemies. With a shield, you could defend against ranged attackers the same way.

 

[toucanbuzz]

Sounds a bit like what the Hackmaster rules do for combat, which are insanely detailed. If you get hit in melee, you can roll a defensive die to parry, results vary based on the weapons. You get more realism, but at a severe cost of slowing down combats. More time spent in combat, less time spent doing other roleplay.

 

[AmerginLiath]

The simplicity of AC is part of the beauty of D&D. At the time that the game was taking off, other nascent RPGs all seemed to have complex tables and series of rolls upon rolls for contested hits. Meanwhile, D&D assigned classes for types of defense that covered different sorts of protection (from armor to hide to magic to speed) and could act as a number to roll against on an adjusted d20. Seeing the Combat Matrix evolve to THAC0 to the d20 DC with only minimal changes over the course of forty years is a testament to figuring out a simple system that basically got it right enough the first time (“Did my attack do well enough to damage an opponent with those defenses? Let’s compare my roll to the target number.”).

 

[Tony Vargas]

The simplicity of AC is part of the beauty of D&D. At the time that the game was taking off, other nascent RPGs all seemed to have complex tables and series of rolls upon rolls for contested hits. Meanwhile, D&D assigned classes for types of defense that covered different sorts of protection (from armor to hide to magic to speed) and could act as a number to roll against on an adjusted d20.

At the height of the fad, 1e AD&D asked you to calculate your AC, rear AC, and shieldless AC, and various AC bonuses or penalties might or might not apply in various circumstances - including parrying, which added an amazing +1 AC, FWIW. 3.x got a lot of flak for separate AC, Touch AC, & modifiers flying everywhere, but, if anything, it was significantly simplified from AD&D. (Arguably, THAC0, for all the flack it got, was a slight simplification from 1e's two facing pages of attack matrices with their series of natural 20s.) 4e further simplified not only AC, but consolidated Touch AC and REF saves into a single REF defense, along with shifting FORT & WILL to attacker-rolls resolution. 5e's handling of AC is still simple enough, but it returns many attacks to defender-rolls 'saves' and has six different categories of those compared with 3e's three categories, and, in addition to a relatively small number of bonuses/penalties, has advantage/disadvantage and randomized bonuses.

5e's prettymuch on a bounce as far as complexity of resolution is concerned. It wouldn't hurt to follow that trajectory and add a little more...

One neat aspect could be that when you're 1st level, your proficiency probably won't give you a better 'parrying AC' than your armor, so it needn't even be considered until later, when players have more bandwidth for a little depth to the combat system. :shrug:

Shields, OTOH, especially larger ones, could do with a very good AC vs ranged attacks from early on. Ranged needs to be taken down a peg, anyway, and enemies closing from behind a shield wall could help.

 

[jgsugden]

What about monks? Are they considered armed?

I'd think the answer to that would be self evident, but yes. Same with creatures with natural weapons, brawling feats, and bad BO.

 

[AmerginLiath]

At the height of the fad, 1e AD&D asked you to calculate your AC, rear AC, and shieldless AC, and various AC bonuses or penalties might or might not apply in various circumstances - including parrying, which added an amazing +1 AC, FWIW. 3.x got a lot of flak for separate AC, Touch AC, & modifiers flying everywhere, but, if anything, it was significantly simplified from AD&D. (Arguably, THAC0, for all the flack it got, was a slight simplification from 1e's two facing pages of attack matrices with their series of natural 20s.) 4e further simplified not only AC, but consolidated Touch AC and REF saves into a single REF defense, along with shifting FORT & WILL to attacker-rolls resolution. 5e's handling of AC is still simple enough, but it returns many attacks to defender-rolls 'saves' and has six different categories of those compared with 3e's three categories, and, in addition to a relatively small number of bonuses/penalties, has advantage/disadvantage and randomized bonuses.

5e's prettymuch on a bounce as far as complexity of resolution is concerned. It wouldn't hurt to follow that trajectory and add a little more...

One neat aspect could be that when you're 1st level, your proficiency probably won't give you a better 'parrying AC' than your armor, so it needn't even be considered until later, when players have more bandwidth for a little depth to the combat system. :shrug:

Shields, OTOH, especially larger ones, could do with a very good AC vs ranged attacks from early on. Ranged needs to be taken down a peg, anyway, and enemies closing from behind a shield wall could help.

Having multiple ACs is still simply a matter of knowing which target number to roll against in that case — it’s not a matter of rolling contested checks to parry or consulting multiple tables like other systems use. I personally prefer limiting the number of ACs that a character has, but having a few established categories that can be pre-calculated and the player and DM can both know is still a basic method compared to other systems (whether those in other games or tweaks suggested by houserules).

 

[Tony Vargas]

Having multiple ACs is still simply a matter of knowing which target number to roll against in that case ..

Which is as far as I suggested taking it. You'll note I said "calculate a 'Parry AC'," not "roll a parry attempt to set an AC."

rolling contested checks to parry.

While that could certainly work and isn't exactly un-due complexity, I personally don't care for using a check to set the DC of another check. Gets too swingy, IMHO.

 

[ad_hoc]

Exactly the rule from 3.x and pathfinder, if I recall correctly. I was surprised not to see it in 5th, but I guess the greatly reduced ubiquity of feats in 5th would make it very rare for non-monks to overcome the danger of attacking armed opponents with unarmed strikes.

Different design goals