What Flavor Is Your Weapon?

[sixdemonbag]

A quick aside:

It's funny to me. The dropping an item when surprised "rule", the surprise "rules", the various "x in 6 chance rules", along with the "rules" for distracting monsters with food and treasure, etc. always read to me as Gary saying: "My players tried to pull off these shenanigans once, here's how I handled it. Just a head's up!" YMMV.

[waysoftheearth]

Jul 18, 2017 9:15:32 GMT -6 magremore said:

I’ll walk back my earlier statement a bit and admit there could be ambiguity if this rule is broken down sentence by sentence, and that ambiguity comes in, I think, because of the “some item” in the first sentence. But reading the two sentences of the rule together as a unit, as they’re meant to be taken, I think provides a clear procedure. 

I guess I was deliberating between:

A. There is a 25% chance that any character surprised by a monster will drop an item. If a character drops an item, roll again to determine what is dropped remembering that only those items held could be so dropped.

and

B. There is a 25% chance that any character surprised by a monster will drop something. Roll separately for the possibility of dropping each item, remembering that only those items held could be so dropped.


It seems the consensus here is A, so that's constructive information. The practical upshot of the A is that when the 25% chance comes up:

. A figure carrying a two-handed weapon will be disarmed.
. A figure carrying two weapons will not be (wholly) disarmed.
. A figure carrying sword and shield may be (50% chance) disarmed.


The question then becomes: What happens to a figure who is disarmed?
Presumably there should be some kind of opportunity cost, but what exactly is it?


Thoughts?


FWIW-- The Dalluhn ms has: "If a player must drop his bow for a hand weapon, the opponent will get a free chop, and the player will hit second in the melee round." which appears to imply the player misses his attack in the first round while switching to another weapon, and then is considered the defender (thus attacks second) in the second round.

(One might note that a round is, in this scenario, about as long as it takes a combatant to switch from a bow to a sidearm).

[magremore]

Jul 18, 2017 16:06:34 GMT -6 sixdemonbag said:

A quick aside:

It's funny to me. The dropping an item when surprised "rule", the surprise "rules", the various "x in 6 chance rules", along with the "rules" for distracting monsters with food and treasure, etc. always read to me as Gary saying: "My players tried to pull off these shenanigans once, here's how I handled it. Just a head's up!" YMMV.

And it seems like Gary might not have been too particular about the this rule as Holmes simplified it to a 1 in 6 chance of dropping everything in hand and Gary let it stand. 

[magremore]

The question then becomes: What happens to a figure who is disarmed?
Presumably there should be some kind of opportunity cost, but what exactly is it?

The surprising party having gotten their free move, the disarmed figure can now declare an action.

I guess that means a spell caster wouldn’t really be penalized, but otherwise the round is taken up by some action like usual, but the effect will necessarily vary between refs.

I’d have retrieving the dropped weapon take a round, but drawing a new one just lose initiative (even if normally using simultaneous initiative). A figure with something like shield or torch in off hand could use that as an improvised weapon (for me, that’s usually 1–3 damage) or punch/kick (–2 to hit vs armed opponents and 1–2 damage). Or the figure might just try to get behind bigger, tougher friends. Or run like hell.

FWIW-- The Dalluhn ms has: "If a player must drop his bow for a hand weapon, the opponent will get a free chop, and the player will hit second in the melee round." which appears to imply the player misses his attack in the first round while switching to another weapon, and then is considered the defender (thus attacks second) in the second round.

I’d interpret that the same way. In practice, in the context of a surprise situation, that looks a lot like the free move (which may or may not result in a free attack) + a loss of initiative if drawing a new weapon.

[hamurai]

I've never used this rule because of surprise.
In my opinion, the characters (at least the fighting type above first level) are competent enough to not drop their weapon due to a surprising encounter with an enemy - that's pretty much what they'd expect to happen in many situations.
If I'd ever make a general rule I'd say a chance of dropping something 1 in 20, maybe the Magic-User might have a 1 in 6 chance and only if the surprising party is pretty close already.

Personally, I can't remember any situation when I dropped something because I was surprised.

[Scott Anderson]

I love the weapon drop rule in theory but disagree with it in practice for two reasons. Reason one is, during a one-minute round, there is plenty of time to pick up a weapon before the bad guys close on you and fight normally. Reason two is, it's not a fun rule at the table in my experience.

Sooooo, if it was up to me, I would skip that part of the programming.

[magremore]

Agreed. I have go-to ways of handling disarmed characters because those situations come up. Have never had a character drop something only because they’re surprised.

[waysoftheearth]

There are lots rules options out there, so my code allows you to configure a "rules engine" before you start the sim. You pick whatever combination of (supported) rules you want to test out, pick the opponents, their loadout, then go. It will start with support for a minimal core of OD&D rules.

That currently includes:
- AC/HD/hp/FC support
- A whole bunch of monster stats (but sims only valid for basic man-types at the mo)
- Loadout (including armor, shield, helm, one-handed and two-handed weapons, main-gauche)
- Fantastic/Normal attack determination
- OD&D surprise odds
- (no encounter distance determination; presumes 1" distance)
- OD&D dropping rule (with dynamic AC/armament modification, no re-arming yet; coming next)
- M2M rear attack rule (surprise attacks are presumed to be rear attacks)
- FAQ initiative rule (the M2M initiative rule is more work to implement)
- Normal attack (first column of Attack Matrix I only)
- (No support for helm/no-helm vs. M&T's 10% head hits yet)
- Normal damage
- OD&D kill rule (dead at 0 hp)
- 465 unit test assertions (all passing)

Execution of each rule during each battle sim creates one or more events, which are streamed to a battlelog (and optionally tee'd to a console) for analysis after the fact. In it's current (infantile) state a battle report can only output the "raw" even stream of one battle sim, like this:



The plan is to add some smarts to the battle report so it can aggregate stats from lots of battles and report back a summary.
So then, finally, we can see what impact the different loadouts really have on player "success" in these one on one battles...

[sixdemonbag]

Jul 20, 2017 8:20:19 GMT -6 waysoftheearth said:

The plan is to add some smarts to the battle report so it can aggregate stats from lots of battles and report back a summary.
So then, finally, we can see what impact the different loadouts really have on player "success" in these one on one battles...

I am beyond excited for this!! I'm humbled by the work you've put in.

I think this absolutely deserves it's own thread (and maybe a sticky) once you get it up and running to your satisfaction.

So cool!!! 

[Scott Anderson]

I don't agree with using the x in 6 drop rules for two reasons.

1. It's not fun in practice

2. In a one-minute combat round, I would say you have time to drop and recover an item. It seems odd to model this action particularly when everything else is abstracted or narrated.

If I were to run the sim, I would not include drops.

[darien]

In my OD&D games, all weapons do 1d6 of damage. Everything else is just fluff and flavor.

My current character is a Magic-User with a hunting knife (specifically one styled like the Buck 119 Special), anachronistic though that may be. Both my DM and I agree that a hunting knife counts as a dagger.

[Scott Anderson]

Looks like i double posted. Sorry

[waysoftheearth]

Jul 20, 2017 14:59:29 GMT -6 Scott Anderson said:

I don't agree with using the x in 6 drop rules for two reasons.

1. It's not fun in practice

2. In a one-minute combat round, I would say you have time to drop and recover an item. It seems odd to model this action particularly when everything else is abstracted or narrated.

If I were to run the sim, I would not include drops.

If I ever "finish it" (cough), you'll be able to toggle whatever options you like on/off.

Meanwhile, the point of this effort was to get a handle on how effective some different weapon options are. As I posted upthread:

Jul 19, 2017 3:46:12 GMT -6 waysoftheearth said:

. A figure carrying a two-handed weapon will be disarmed.
. A figure carrying two weapons will not be (wholly) disarmed.
. A figure carrying sword and shield may be (50% chance) disarmed.

So the drop rule differentiates these weapon options by the chances of being disarmed. I'm interested to see what difference it makes, just for the sake of knowing. That's not to say I'm insisting everyone must use the drop rule in at the game table; it just means we'll all get a better understanding of how that rule affects combat, and what we're doing when we choose to exclude it.

re: time to re-arm: I'm not in the one-minute combat round camp, so I don't share the concern about how long it seems to take a disarmed figure to re-arm. At the end of the day, the duration of a combat round doesn't impact the outcome of the sim, so there's no need to dive down into that one.

Anyways, tonight's effort was to start counting wins/losses across x many battles.

Don't read anything into these numbers but, below are some preliminary outputs to give the general idea where it could go. Here it's only reporting the "kill events", but of course it could report every detail: surprises, number of drops, gained/lost initiative, number of attacks, hits/misses, damage caused, etc.


The first one shows that bandits and orcs (statistically equivalent, for the purpose of the sim) were on equal terms. The second one shows that a verteran with the same loadout as an orc has an edge; the veteran's +1 pip of HD and FC equates to something like an 8% advantage. The third one puts the veteran in plate armor, and you can see what happens.

Okay, so next I'm gonna focus on putting in sixdemonbag 's weapon rules and see what we get...

[waysoftheearth]

I've now run several millions of simulated combats thru my sim and collected a bunch of data.

For this thread, I think it's worthwhile starting with the following as a "baseline":

TABLE A


Table A is a summary of 3.2 million simulated bouts between individual orcs and bandits. These belligerents were chosen because they are both 1 HD man-types with FC of "1 man". In other words: they should have equivalent combat performance. Therefore, any difference in performance observed may be attributable to the equipment loadouts they are given.

In Table A the orc is always equipped with a weapon and no shield. Meanwhile, the bandit is tested against the orc in 100,000 bouts for each of four different loadouts, and the bandit's "win rate" observed. The whole exercise is repeated for each of the eight possible combinations of: four armor types with/without the (U&WA) drop rule in play.

At the right-hand side of Table A I've calculated the mean number of wins for the bandit across all four armor types, for each of his loadouts. Since the orc is equipped with a weapon and no shield, we expect the bandit to get 50% wins when he has exactly the same loadout. In fact, we observe 49.948% and 49.946% with/without the drop rule, respectively, so that's pretty close.

From there we can compare the mean performance of the bandit's other loadouts against the supposed "fair" loudout (the 50/50 win rate). In other words: what advantage do the other loadouts deliver? I've calculated these advantages (in terms of win rates) in the far right column. I think what it's telling us there is:

Without the Drop Rule
* The first three loadouts (weapon+nothing, weapon+shield, weapon+main-gauche) perform equally.
* Carrying a shield delivers about a 4% advantage.

With the Drop Rule
* The main-gauche loadout now delivers about a 1% advantage.
* Carrying a shield still delivers about a 4% advantage.

These numbers seems reasonable to me; the main-gauche loadout is advantageous because the bandit cannot be disarmed (and therefore cannot lose a round having to re-arm). But this still isn't as good as the shield loadout, which reduces the chance of being disarmed and offers better AC... more on this below.

Okay, so then we can give our orc army shields and repeat the above exercise:

TABLE B


This time the bandit's weapon+shield loadout represents the "fair" case where we expect him to win 50% of bouts.  So... skipping straight to the far-right column, we can see pretty much as expected:

Without the Drop Rule
* The first three loadouts perform equally again, all carrying just shy of a 4% disadvantage vs against the shield loadout.

With the Drop Rule
* The main-gauche loadout once again has slightly less than a 1% advantage compared to the first two loadouts, which are again about 4% worse off than the shield option.

Okay, so by the book shields look good. If we play the drop rule, then the main-gauche option may be useful, but it's not as good as a proper shield.




So returning at last to sixdemonbag 's original proposal:

Jul 5, 2017 9:59:36 GMT -6 sixdemonbag said:

here are some house rules that might prove useful, all being equal in terms of DPR, just choose based on personal preference:

Two-handers reroll 1's for damage and two-weapon fighting earn a +1 to-hit. This makes all three main loadouts mathematically equal (sword and board, dual wield, and two-handed). An empty hand provides no mechanical benefit aside from carrying torches, potions, magic items, scrolls, grabbing, punching, etc. and nor should it, IMO.

How do sixdemonbag's equal DPRs translate into combat wins between bandits and orcs (supposedly "typical" 1 HD sorts)? Let's see...

TABLE C


So this time the two-handed weapon loadout rerolls any damage die of 1 (any number of times), and the dual-weapon loadout has a +1 advantage to hit.

Looking again to the far right column, we can see the two affected loadouts delivery similar advantage--somewhat under 4%--without the drop rule. The shield continues to deliver the 4% we saw previously. What's potentially interesting is that the "seemingly obvious" equivalence between the +1 AC advantage of the shield and the +1 to hit advantage of the main-gauche is perhaps not quite so obvious in the numbers. Or perhaps it's all down to random variation?

With the drop rule in play, the main-gauche option cannot be disarmed, and shield option has only a 50% chance of being disarmed. So, if the dual-weapon +1 to hit advantage cancels the shield's +1 AC advantage, then we should expect the main-gauche to gain a bit more ground than the shield due to its never being disarmed. However, the figures in Table C do not illustrate this nuance; possibly this may be due to random variance in the numbers, an error on my part, or something I haven't thought of yet? 


This next one is perhaps more interesting:

TABLE D


Here, the orcs get shields and continue to do battle under sixdemonbag's house rules.  As explained above, the "fair" case is now the mutual weapon+shield loadout, where we expect (and observe) close to a 50% win rate for the bandits.

The 0.6% observed advantage of the main-gauche loadout over the shield without the drop rule concerned me enough that I ran this scenario over. But I observed a similar outcome several times. So (unless I've screwed up somewhere) I think the main gauche loadout genuinely has another edge over the shield loadout, and I think it comes down to this:

Every figure wins unilateral surprise 8 times in every 36 bouts (that's 22% of battles), and the combat algorithm assumes that surprise attacks are also rear attacks. Rear attacks discount the shield's AC advantage. So, if you have the shield loadout, you will loose its +1 AC advantage in the first round 22% of the time. If you have choose the main-gauche loadout, you never lose its +1 to hit advantage (ignoring the case of losing one or other weapon is dropped, because this is the "without drop" case. Besides which, this can happen to a shield just as frequently in the "with drop" case). So the main-gauche loadout has an asymmetric advantage over the shield, which appears to be worth about 0.5%.

It's also interesting to observe in the above table that the two-handed weapon loadout appears to trend downward from ~51% versus unarmored orcs to ~49% versus plate armored orcs. Possibly, we're observing the damage advantage of the two-handed loadout decreasing as the target gets harder to hit. To run with that one for a moment, it's further possible to speculate this effect may become more significant in longer duration combats involving belligerents with more hp.


So, to finally wrap up, I think these combat simulations suggest that the interplay between loadout, AC, the drop rule, and surprise attacks is more subtle than it may at first seem. I don't think we can say that "all three main loadouts mathematically equal", but nor are they going to be vastly different without larger adjustments.

[hamurai]

Great job, very informative! Thank you!

[waysoftheearth]

For completeness, these data compare each of the 16 possible combinations of weapon loadouts, with and without the drop rule:



Note that the all important % advantage calculation is now the ratio of the observed kill ratio to the expected kill ratio if there is no advantage (i.e., 50,000 kills or exactly 50% kills).


The final "% advantage" figures can be aggregated into two simpler tables, like this:





Each cell of these two summaries comprises 400,000 bouts (100,000 for each of four possible armor types). So each of the two summaries span 400,000 x16 = 6.4 million bouts. The two summaries cover 12.8 million simulated, one on one bouts. (No real orcs were harmed in the making of these tables).


A couple of general observations:
* The green cells have identical belligerents, so should produce 50% bandit kills and 0.0% advantage. That these cells vary by around +-0.1% implies "a degree" of confidence in these figures.

* We should expect the result of A vs B to be equal and opposite to the result of B vs A in each case. That almost all these cell pairs are exactly equal and opposite again implies that the whole thing is, at least, not hopelessly broken.


You can download the raw data here.

[Deleted]

Jul 18, 2017 16:06:34 GMT -6 sixdemonbag said:

A quick aside:

It's funny to me. The dropping an item when surprised "rule", the surprise "rules", the various "x in 6 chance rules", along with the "rules" for distracting monsters with food and treasure, etc. always read to me as Gary saying: "My players tried to pull off these shenanigans once, here's how I handled it. Just a head's up!" YMMV.

That's because that is literally what they are. And yes, I do mean LITERALLY.

[sixdemonbag]

Jul 25, 2017 13:21:15 GMT -6 @gronanofsimmerya said:

Jul 18, 2017 16:06:34 GMT -6 sixdemonbag said:

A quick aside:

It's funny to me. The dropping an item when surprised "rule", the surprise "rules", the various "x in 6 chance rules", along with the "rules" for distracting monsters with food and treasure, etc. always read to me as Gary saying: "My players tried to pull off these shenanigans once, here's how I handled it. Just a head's up!" YMMV.

That's because that is literally what they are. And yes, I do mean LITERALLY.

Makes perfect sense and adds so much charm to the game. Thanks for the insight.

[Deleted]

Er...I'm late to the party, but I like to give weapons "bonus effects" (other than damage) that hopefully cause players to think strategically about them. Here's my rough draft...it hasn't been fully playtested yet because in my last game, zero combat occurred. The players negotiated their way out of combat encounters, and when it was clear that it was time to fight, they ran like the blazes, jumping out of second floor windows and the like.

Weapon	Bonus Effect
Axe, Battle	Two handed; Negates shield bonus; Can be used to destroy wooden doors
Axe, Hand	Can be used as both a melee and ranged weapon
Club	Costs nothing
Dagger	Allows two attacks per round against a grappled target
Flail	Negates shield bonus
Mace	2 in 6 chance to stun target
Spear	Has reach; Halts closing opponents on successful attack
Staff	Two-handed; Has reach; 2 in 6 chance of knocking target prone
Sword, Long	1 in 6 chance of making a second attack
Sword, Short	1 in 6 chance of hamstringing target (-3 to speed)
Sword, Great	Two-handed; 2 in 6 chance of parrying an attack
Warhammer	Reduces AC by 2;
Bow, Long	Allows 2 attacks per round (Range 70ft.)
Bow, Short	Allows 2 attacks per round (Range 50ft.)
Crossbow, Heavy	Reduces AC by 2; Always causes maximum damage; Fires once every 3 rounds
Crossbow, Light	Reduces AC by 2; Fires once every 2 rounds

I am very-much enamored by the "all weapons do 1D6 damage" ethic of OD&D, but I'd also like to give my players incentive for choosing one weapon loadout over another.

[Starbeard]

Regarding dropped items due to surprise, I've never decided what I think about it. The thought of losing a surprise check already keeps players awake at night, and adding a 1:4 chance of dropping your equipment seems like it just adds insult to injury. Plus, if everything else in combat is already abstracted, then you could argue that the abstract "surprise round" already takes care of the possibility that the PCs have dropped something and lost time picking it up.

The times I have used the rule, I've usually interpreted it as a stroke of bad luck that may or may not include literally dropping the weapon. If the encounter distance is close enough, then maybe the item was grappled or knocked out of hand by the opponent; or more usually, it means the character spends a round faffing about with his gear, feeling stupid that he had let his guard down and sheathed his weapon/slung his shield and now the enemy is pressing on him. Either way, it essentially means an extra round of fussing about before you can use the item in combat. But again, the more I think about it, the more I feel like that result should already be subsumed in the general surprise check.

[rossik]

Jul 30, 2017 23:26:43 GMT -6 @gti said:

Er...I'm late to the party, but I like to give weapons "bonus effects" (other than damage) that hopefully cause players to think strategically about them. Here's my rough draft...it hasn't been fully playtested yet because in my last game, zero combat occurred. The players negotiated their way out of combat encounters, and when it was clear that it was time to fight, they ran like the blazes, jumping out of second floor windows and the like.

Weapon	Bonus Effect
Axe, Battle	Two handed; Negates shield bonus; Can be used to destroy wooden doors
Axe, Hand	Can be used as both a melee and ranged weapon
Club	Costs nothing
Dagger	Allows two attacks per round against a grappled target
Flail	Negates shield bonus
Mace	2 in 6 chance to stun target
Spear	Has reach; Halts closing opponents on successful attack
Staff	Two-handed; Has reach; 2 in 6 chance of knocking target prone
Sword, Long	1 in 6 chance of making a second attack
Sword, Short	1 in 6 chance of hamstringing target (-3 to speed)
Sword, Great	Two-handed; 2 in 6 chance of parrying an attack
Warhammer	Reduces AC by 2;
Bow, Long	Allows 2 attacks per round (Range 70ft.)
Bow, Short	Allows 2 attacks per round (Range 50ft.)
Crossbow, Heavy	Reduces AC by 2; Always causes maximum damage; Fires once every 3 rounds
Crossbow, Light	Reduces AC by 2; Fires once every 2 rounds

I am very-much enamored by the "all weapons do 1D6 damage" ethic of OD&D, but I'd also like to give my players incentive for choosing one weapon loadout over another.

thats a fine list

[Deleted]

Aug 1, 2017 17:08:30 GMT -6 rossik said:

thats a fine list

Thanks. I just kind of grabbed it from S&W Whitebox and dropped the entry for Polearms. I'm still tinkering with it, particularly the entries for swords.

[hamurai]

While I do enjoy reading over that list I must say that, personally, I like it simple in my games.

Also, just from reading over it, I'd say maces, long and especially great swords are too powerful. Why would anyone wield axes or flails? Negation of shield bonus means AC is reduced by 1, and only if the target is using a shield. A warhammer gives a flat 2 AC reduction against all targets. Great swords parry every 3rd attack, which probably makes them *the* defensive weapon.
Some weapons reduce AC by 2 - I'm guessing that's an armor piercing quality? Then what if these weapons are used against unarmored targets - do they still reduce AC by 2?

Just my thoughts

[sixdemonbag]

Aug 6, 2017 14:34:22 GMT -6 hamurai said:

While I do enjoy reading over that list I must say that, personally, I like it simple in my games.

Also, just from reading over it, I'd say maces, long and especially great swords are too powerful. Why would anyone wield axes or flails? Negation of shield bonus means AC is reduced by 1, and only if the target is using a shield. A warhammer gives a flat 2 AC reduction against all targets. Great swords parry every 3rd attack, which probably makes them *the* defensive weapon.
Some weapons reduce AC by 2 - I'm guessing that's an armor piercing quality? Then what if these weapons are used against unarmored targets - do they still reduce AC by 2?

Just my thoughts

And this is exactly the kind of thing I was trying to avoid 

[sixdemonbag]

Jul 23, 2017 6:48:20 GMT -6 waysoftheearth said:

So, to finally wrap up, I think these combat simulations suggest that the interplay between loadout, AC, the drop rule, and surprise attacks is more subtle than it may at first seem. I don't think we can say that "all three main loadouts mathematically equal", but nor are they going to be vastly different without larger adjustments.

I think this is the valid conclusion as well. I will go back to the OP and point this out.

"all three main loadouts mathematically equal" is not the correct phrase to use it seems. I need to think of something better to replace it. Hmm...

[sixdemonbag]

Added a note to my OP. Thanks everyone. More suggestions are always welcome!