D&D 5E D&D compared to Bespoke Genre TTRPGs

[AbdulAlhazred]

Could very well be.

I’d be curious to have some data and testimonials attached to that data (% of 5e GMs who are presently running SwC and why/when they started).

PBtA and the huge influx of indie games over the last decade (and their increasing acceptance) may have a hand to play as well.

Hard to say which is cause and which is effect, but I would think it is a kind of snowball thing where once a critical mass is reached things reinforce. I also wonder about the demographics. I mean, which styles of play are prevalent in different age groups, etc.?

 

[AbdulAlhazred]

And I'm going to stop you right there. There's a difference between a combatant and a front line combatant. The ability to e.g. fire a crossbow doesn't require that much.

Let me stop you there. In 5e a hill giant is not just slightly bigger than an ogre. They're 16' tall according to the Monster Manual, and huge creatures rather than large. They're actual giants (and it's one of the 5e fluff changes I fully support).

Fortunately for the giant it's a hill giant, not a plains giant. Which means it can get relatively close.

The villagers are vaguely organised and they live in the sort of country a giant is likely to live in. However what the giant finds first isn't a farmer but a herdsman with a flock of sheep or goats. Part of the point of having herdsmen is the early warning they provide. And the herdsman always carries two things. The first is a sling, primarily for driving off wolves. The second is a horn (made of horn) so they can call for help or alert the village.

At this point most giants will be looking to steal a sheep or a goat, which I agree they can do relatively safely. And naughty word happens. Losing a goat every week or so to a giant is a problem, but it's not village-threatening.

If the giant is particularly homicidal they get the shepherd/goat herd. But the horn has been sounded.

Already done unless the shepherd in particular was known for crying wolf. They heard that horn. And it got passed on. The warning's passing at almost the speed of sound.

"What weapons they can find" - your village is a bunch of utterly clueless twits who leave their weapons buried under the junk? When they live in giant country?

Did he teleport? The giant has barely killed the shepherd by the time people are grabbing weapons.

We're in hill country and the call has gone out for "giant". The weapons of choice are bows, slings, and javelins with hatchets as a last resort.

Of course not. They pepper that giant with sling-stones and a few arrows when it comes into sight. No ones grabbing staves or clubs. They don't want to get into range of that thing.

Almost all with ranged weapons. Slings are dirt cheap and in D&D count as simple weapons. If it's a particularly dangerous area most of them might have slings already on them.

If we assume half your "non-combatants" are Str 8 Dex 8 and have slings so can do damage that's still 30 slingstones that need 10 or more to hit (vs AC 11; +2 proficiency, -1 stat bonus); the non-combatants between them do an average of 28.5dpr at short range (accounting for crits) or 13.8 dpr at long range.

The giant lasts around 4 rounds against the non-combatant slingers if he closes to short range and 8 if he stays at long if they get to focus fire.

Or around half the "non-combatants" take the giant down and only lose half a dozen. It is, of course, faster if you get the actual fighters involved.

Possibly the foolish ones that don't use slings and that rush down the throat of the giant rather than trying to shoot it from far away. Yes, villages that are that badly prepared and live in giant country might get wiped out. Of course so do the giants.

No walls and organised militia needed. Just proficiency with simple weapons and not doing anything stupid.

Personally, I think @FrozenNorth's version of the story rings pretty true. You make some points, people might well be more careful. OTOH 16' is not that tall, most trees in a forested area are easily much taller than that. So I would think that engagement range is going to be SHORT, 10 meters or less typically. OTOH dogs are pretty darn good guards, so daytime alertness probably means there's a decent chance of detection with some warning. This is probably enough, since you 'play the odds' on security. The reasoning is, the Giant has to live its whole life, it isn't going to take a 10% chance of death just to do what? Wreck some houses for a lark?

Instead though, WERE I A GIANT, I would be coming in like gang busters at 3 AM. By the time anyone realizes there's an attack and gets organized, you will be long gone. You CANNOT underestimate chaos in the real world. If you have ever been in a real disaster situation you will begin to understand. People DON'T just automatically self-organize, and there is a very large, surprisingly large, 'Entropy' to a situation. You wake up at 3 AM hearing screams and crashing sounds, dogs barking, etc. What do you do? Is it a giant? Is it an earthquake? A dragon? WTF? Sure, you store your sling and your spear near the bed, so you grab them and stick your head out the door. All you see is a bunch of people running left and right, and then something BIG smashes through the house next door, with a body dangling by one leg in its grip. Stunned you begin to fumble a sling stone into position and move away from the door to be clear to cast. Meanwhile the giant turns right and disappears into the darkness. 20 minutes later it turns out 3 people are dead, 2 houses are damaged, and the giant may or may not have been hit by a couple weapon attacks.

And it is hardly going to matter which system you 'run' this under. Some will mechanically do a bit better job than others, I guess. In AD&D you have maybe 60' visibility, and the giant has 15" movement. So the above is pretty natural, it got a surprise round or two based on your unpreparedness, so by the time you moved outside it got initiative and moved out of your vision range. In that game the villagers COULD hurt the giant, it would have about 35 hit points and AC4, so it would take a bunch of them to do much, but as an organized group with reasonable weapons they could indeed repel a Hill Giant attack. 2e giants have more hit points IIRC but the basic outline is similar. I don't know about 3e giants, never messed with them. 4e giant would be a level 1-5 solo. Surprise is less of a mechanical big deal here, but the whole thing is unlikely to be a straight up combat. I think the general result WOULD look roughly similar to 1e. 5e seems to say the giant is easier to hit but has a LOT more hit points. I think it still works out about the same.

One thing that you see here is that D&D doesn't really attempt to model things like the 'chaos of battle' in any realistic way, so mechanical analysis of this kind of situation doesn't really work. This is another illustration of my core point that RPGs cannot simulate reality, not with rules.

 

[AbdulAlhazred]

What I like it that it doesn't affect CR (it is basically Legendary resistance, but with the HP penalty built in). I have made a couple different versions and can't decided which I like best.

Yeah, back in 4e days we implemented a LOT of different variations on it. Anywhere from sacrificing an action (MM3 solos typically do this) to taking some damage, using up APs, and I think there was at least one monster that could take some damage, though I seem to recall it was a bit less direct than that. I like the simplicity of this approach though, HP are the vital currency of a fight, so using them as such makes sense. You could even use HP cost to grant extra action, etc.

 

[AbdulAlhazred]

I have - I generally stop using skills with those dms (if I even keep playing with them), but it's a thing that can happen. Which is more common is unknown (and unless you got surveys to cite, it's all anecdotes.)

I will note that I see it much less often recently, which I attribute to better dm advice on the interwebs.

I think this gets into the history of D&D. The original system did not envisage any sort of unified skill system, but obviously specific types of 'check' (BB/LG for example) were provided. In NO CASE was any explanation provided as to what success or failure entailed. This is also true of the Thief Skills introduced in Greyhawk. AD&D 1e doesn't clarify this. Some people did use various ability check procedures all through this era, but AD&D again doesn't document this, and doesn't clarify anything about success or failure, or even describe how checks work within the system, except for a few specific cases. Even then failure is never explained.

I'm not sure exactly what 3.x says. It obviously has a unified check mechanic and skill system. I just haven't actually read the rules much. I suspect various possibilities are entertained at different points, but the basic stock default that I have always seen in play is that a failure is a 'hard fail' and you cannot try again, except if you gain a level. This is obviously going to be modified for some types of situations (IE where the check leads to entirely new fiction naturally so that trying again doesn't even make sense).

4e is the first time D&D actually builds mechanics which don't treat skills as binary pass/fail on the specific action the check maps to, and then builds another process.

So, basically, historically, the rules were that you hard fail, that a check represents the specific physical action invoked in the fiction, and that a failure thus MUST logically equate to that action not having its desired effect. I believe, reading between the lines of thief abilities explanations in the DMG, that failures were intended to be 'default soft', so you could just try again (which is why lock picking takes 1-10 minutes for example, the time is a penalty for failing in effect). This pretty much explains why we mostly see binary pass/fail today as the 'default mode' of the game, and concrete association of the mechanical check to a single physical action, as the 'default process model'.

I think it is fair to say that most educated GMs today know that there are other ways, that 5e even discusses (though barely/optionally supports) a couple of these, etc. Given how clumsy that support is though, my experience is it isn't really used. Instead GMs sort of 'wing it' and apply a '2e-like sensibility' which basically says "fudge things until it comes out right" so if you failed your stealth check during the heist, then they're not likely to instantly ring the big master alarm, instead this type of GM will likely do something less drastic. OTOH you cannot count on that, and I have no idea how prevalent this type of DM is vs any other type 'in the wild'. Also it is surely true that there are 'violently traditionalist' GMs (at least online) who insist that AD&D 1e was the right way and presumable DO raise the master alarm when you fail a check.

 

[AbdulAlhazred]

Interesting, lets analyze this from an 'indie game' perspective:

So this thread, and more specifically @Hussar 's story about the 3 DM's he played under, inspired me to go back and re-read the DMG section for 5e on Ability checks... and I have to say I was surprised at just how much it did to set out the expected practices of 5e... which contrary to many people's assumptions don't appear to be... do whatever you want (though like any game the DM is free to ignore advice, examples, practices, etc.). I decided to pull excerpts (with commentary) from the DMG below to highlight what I am talking about...

Using Ability Scores...

When to roll
1. When a player wants to do something it's often appropriate to let the attempt succeed without a roll

2. Only call for a roll if there is a meaningful consequence for failure

3. When deciding whether to use a roll ask the following questions
a. Is a task so easy and free of conflict and stress that there should be no chance of failure
b. Is a task so inaapropriate or impossible that it can't work

4. If the answer to both of these questions is no then some type of roll is appropriate


So above we have the process by which a DM should decide to say yes to an action... say no to an action or roll the dice.

OK, so 4e, for example, says basically "Say yes, or roll the dice." However, I think it is fair to assume that there is generally a "no, you can't just find a laser in the Duke's bathroom" sort of principle at work, though it is unstated. So a GM can say "nah, that's just not appropriate to the fiction, at all." or "nah, that breaks genre conventions." 4e does have a mild 'rule 0' in that sense, this is a GM call, though there are places where the text talks about the table discussing things. Other games may vary, so maybe in some cases rejection of an action declaration would effectively take table pressure or something, making it 'bad faith play'. I would judge that 5e is compatible at this level with what I would consider good practice.

Ability Check: A test to see whether a character succeeds at a task that he or she has decided to attempt

The DMG defines what a check represents... specifically the chance to succeed or fail at a single task. Which in turn implies that any action being rolled for should be framed as a single discrete task that has success state and a failure state.

Here we start to run into an issue for me. While this sort of framing of checks may be acceptable in some cases, if it is taken as an absolute requirement of the system, then it basically broke pretty much every way to do anything but binary hard pass/fail. Obviously said pass/fail could still entail a fail state that is effectively 'partial' (IE the guard doesn't ring the master alarm when the thief fails the stealth check right off).

Multiple Ability Checks

• If the only real cost is time yes (a character spending ten times the normal amount of time auto-succeeds at the task)
• If not the circumstances or approach must be changed to attempt again (with a harder DC at the DM's discretion)

Just a restatement of 'say yes', plus a milder restatement of hard fail that gives some wiggle room.

Contests
-Use a contest if a character attempts something that either directly foils or is directly opposed by another creature's efforts

• Instead of a DC ability checks are compared to each other
• DM picks the ability that each creature must use

Above we are given the practices for using multiple ability checks and contests as opposed to a regular ability check. Of particular interest is the fact that if players can come up with a new approach or change their circumstances they can roll again. Personally I don't think many DM's enact this specific practice

Contests, AKA opposed checks, don't really change the model, they are simply an alternative DC-setting mechanism. Not bad, but not changing much. The shading on the hard fail is hard to evaluate, some GMs might use it to provide soft fail, effectively, and that might open up some cans of worms, but that isn't system's problem.

Difficulty Class
- Think of how difficult a task is and then pick the associated DC from the typical DC's table

- Most people can accomplish a task of DC 5 (Very Easy) with little chance of failure, if a roll is deemed necessary a task usually will not fall into this category. Unless there are unusual circumstances this should be an auto-success

-If not ask yourself if a task's difficulty is easy moderate or hard...If the only DC's you ever use are 10/15/20 your game will be fine

-A DC 10 task is accomplished 50% of the time w/attribute 10 and +0 prof bonus

-You can use a higher DC than 20 but caution and level consideration is advised

So the first thing here is it really brings home how easy it is to determine success rates for an ability check in D&D 5e due to nearly every +1 easily translating into 5% and using DC's that increase in 5 point intervals. But more importantly there are practices laid out above for determining the DC and while one can go beyond them in both the range of the DC and rolling for the DC 5 the game takes time to caution against it except in special circumstances.

Just as an observation: This seems like a poorly thought out process to me. The DC table doesn't account for the (usually level associated) capabilities of the PCs. Thus it encourages GMs to simply call things DC 25 or DC 30 simply because they 'sound hard' and that bones lower level PCs. The rules NEED to discuss relative difficulty. This is not even a foreign concept to 5e, as it has CR for monsters. The lack of this discussion is a MAJOR problem for 5e's ability check system! It means, by the book at least, low level PCs should shun all skill checks, and high level ones should waltz past most of them. It doesn't even make sense.

Applying Adv/Disadv
-Advantage when: Circumstances provide an edge, An aspect of the environment contribues to success, player shows creativityor cunning in attempting or describing a task, previous actions improve success.
-Disadv when: Circumstances hinder success, an aspect of the environment hinders success, some aspect of a plan or descriotion makes success less likely.

I like this mechanic overall, but it should be the ONLY modifier mechanic IMHO. Anyway, this is just me, as a mechanic it doesn't really impact overall system characteristics.

Resolution & Consequences
-As DM you determine the consequences of attack rolls, ability checks and saving throws

-In most cases doing so is straightforward: When an attack hits, it causes damage...when a creature fails a saving throw, the creature suffers a harmful effect...when an ability check equals or exceeds the DC, the check succeeds.

- As a DM you have a variety of flourishes and approaches you can take when adjudicating success and failure to make things a little less black and white
a. Success at a Cost: When a character fails a roll by 1 or 2 you can allow the character to succeed at a cost. When you introduce costs such as these try to make them obstacles and setbacks that change the nature of the adventuring situation. In exchange for success, characters must consider new ways of facing he challenge
b. Degrees of Failuer: Sometimes a failed ability check has different consequences depending on the degree of failure

-Critical Success and Failure : Increase the impact of success or failre on a 1 or 20

In the context of the default hard fail where a skill check is absolutely bound to a single fictional action something like a more robust system for deciding and arbitrating the degree and consequences of failure and success would be very valuable, but what is presented is both optional and not even very well written.

Practices for determining advantage and disadvantage as well as practices for determining consequences of rolls with some optional ways to expand on it.


OPTIONAL RULES

Variant: Automatic Success
-A character automatically succeeds on a check with a DC less than or equal to the relevant ability score -5
- If a characters proficiency bonus applies to a check (through skills or tools) they automatically succedd at a DC 10 or less, at 11th level they auto-succeed on DC 15 or less

Also I included this rule because its optional but it does seem to address the general competency complaint some have about the d20...at least when it comes to tasks that should be relatively easy for heroic characters.

One thing noticeably lacking is a closed resolution system for multiple tasks... On the one hand I don't know if it's necessarily needed... On the other hand I think for those that want it... that the complex trap rules in Xanathar's would be a good basis for something similar to the 4e skill challenges... though I also think the skill challenges from 4e could easily be adapted to 5e.

Yeah, so, basically we see the root of the reason people may be talking about 'the heist fails at the first check failure'. Every check in 5e is an action bound to a fiction. The GM picks the type of check, and the DC, and does so AFTER the player declares the action. Consequences are never discussed at all, though there is a hint that a 'soft fail' could be possible (but often this won't even make sense fictionally anyway). GMs might optionally (or simply as a matter of principle) provide failure that allows for a continuation, but Fail Forward per se is not even discussed, at least in this summary. So, a 'by the book' 5e GM is certainly within the rules text when they would say "OK, make a DC20 Stealth check, oh you failed, the alarm rings!" There isn't any negotiation, no player-side management of risk, resources to pay down failures, or structure to explicitly produce other kinds of results besides a fictionally negative outcome of potentially unlimited consequence.

I don't feel that anything here would make me re-evaluate my existing opinion of 5e's check system/core mechanic. It is OK as far as it goes. It is simply woefully inadequate. I would note that we would have to review how the GM is instructed, and how explicitly, in terms of what kinds of consequences to invoke and to what ends in order to complete the picture (IE Dungeon World relies heavily on its agendas and principles to contextualize its mechanics). However I'm pretty sure that there is a strong message in 5e there to the effect that the GM is in charge and should "do what they think is right."

 

[doctorbadwolf]

I have - I generally stop using skills with those dms (if I even keep playing with them), but it's a thing that can happen. Which is more common is unknown (and unless you got surveys to cite, it's all anecdotes.)

I will note that I see it much less often recently, which I attribute to better dm advice on the interwebs.

Personally, I think @FrozenNorth's version of the story rings pretty true. You make some points, people might well be more careful. OTOH 16' is not that tall, most trees in a forested area are easily much taller than that. So I would think that engagement range is going to be SHORT, 10 meters or less typically. OTOH dogs are pretty darn good guards, so daytime alertness probably means there's a decent chance of detection with some warning. This is probably enough, since you 'play the odds' on security. The reasoning is, the Giant has to live its whole life, it isn't going to take a 10% chance of death just to do what? Wreck some houses for a lark?

Instead though, WERE I A GIANT, I would be coming in like gang busters at 3 AM. By the time anyone realizes there's an attack and gets organized, you will be long gone. You CANNOT underestimate chaos in the real world. If you have ever been in a real disaster situation you will begin to understand. People DON'T just automatically self-organize, and there is a very large, surprisingly large, 'Entropy' to a situation. You wake up at 3 AM hearing screams and crashing sounds, dogs barking, etc. What do you do? Is it a giant? Is it an earthquake? A dragon? WTF? Sure, you store your sling and your spear near the bed, so you grab them and stick your head out the door. All you see is a bunch of people running left and right, and then something BIG smashes through the house next door, with a body dangling by one leg in its grip. Stunned you begin to fumble a sling stone into position and move away from the door to be clear to cast. Meanwhile the giant turns right and disappears into the darkness. 20 minutes later it turns out 3 people are dead, 2 houses are damaged, and the giant may or may not have been hit by a couple weapon attacks.

And it is hardly going to matter which system you 'run' this under. Some will mechanically do a bit better job than others, I guess. In AD&D you have maybe 60' visibility, and the giant has 15" movement. So the above is pretty natural, it got a surprise round or two based on your unpreparedness, so by the time you moved outside it got initiative and moved out of your vision range. In that game the villagers COULD hurt the giant, it would have about 35 hit points and AC4, so it would take a bunch of them to do much, but as an organized group with reasonable weapons they could indeed repel a Hill Giant attack. 2e giants have more hit points IIRC but the basic outline is similar. I don't know about 3e giants, never messed with them. 4e giant would be a level 1-5 solo. Surprise is less of a mechanical big deal here, but the whole thing is unlikely to be a straight up combat. I think the general result WOULD look roughly similar to 1e. 5e seems to say the giant is easier to hit but has a LOT more hit points. I think it still works out about the same.

One thing that you see here is that D&D doesn't really attempt to model things like the 'chaos of battle' in any realistic way, so mechanical analysis of this kind of situation doesn't really work. This is another illustration of my core point that RPGs cannot simulate reality, not with rules.

It’s also just odd to me that Hill Giant became the assumed giant in the example, and it went from me talking about towns guard to a small village with no guard, but some other notes:

Hill Giants are the least scary giant. They’re just big and hungry. Scare them and they run.

But the real giants? They make and use real weapons, and armor, and shields, and have natural magical abilities, and are smart. But, the towns guard should still be able to drive them off.

4e and 5e both have ways to allow you to only be able to help fend off big threats at low level, and then come back several levels later and take that threat down.

IMO, 5e’s solution is cleaner. You just use the same statblock, and the townsguard can still help by peppering it with Missiles but can’t kill it unless it just stands there, but now you can fight it and win.

The two big things I didn’t like in 4e were the number scaling and stacking bonuses. In 5e I can make a new statblock for the same critter, but I don’t have to in order for the fiction to make sense.

 

[AbdulAlhazred]

It’s also just odd to me that Hill Giant became the assumed giant in the example, and it went from me talking about towns guard to a small village with no guard, but some other notes:

Hill Giants are the least scary giant. They’re just big and hungry. Scare them and they run.

But the real giants? They make and use real weapons, and armor, and shields, and have natural magical abilities, and are smart. But, the towns guard should still be able to drive them off.

4e and 5e both have ways to allow you to only be able to help fend off big threats at low level, and then come back several levels later and take that threat down.

IMO, 5e’s solution is cleaner. You just use the same statblock, and the townsguard can still help by peppering it with Missiles but can’t kill it unless it just stands there, but now you can fight it and win.

The two big things I didn’t like in 4e were the number scaling and stacking bonuses. In 5e I can make a new statblock for the same critter, but I don’t have to in order for the fiction to make sense.

Honestly, I am not super hostile to 5e's approach either. I mean, sure, if it works, that's fine. OTOH I'm interested in the game, so whichever is better in that perspective is my only concern. I lost my thought that anything should represent anything in a fixed way mechanically a long time back. As a GM I am unbound by such notions and free to experiment. To a large degree I don't see core elements of 5e at this level as problematic really. Certain details, like the sheer ugly bad game design of saves as a mechanic, that's different, but it doesn't really impact the overall crude scaling properties of the engine, which are mostly OK. Even the ability check system is not totally out in left field, though I would tweak a few things.

 

[Imaro]

Just as an observation: This seems like a poorly thought out process to me. The DC table doesn't account for the (usually level associated) capabilities of the PCs. Thus it encourages GMs to simply call things DC 25 or DC 30 simply because they 'sound hard' and that bones lower level PCs. The rules NEED to discuss relative difficulty. This is not even a foreign concept to 5e, as it has CR for monsters. The lack of this discussion is a MAJOR problem for 5e's ability check system! It means, by the book at least, low level PCs should shun all skill checks, and high level ones should waltz past most of them. It doesn't even make sense.

I'm a little confused by this statement... Why would the DC's have to account for the level associated capabilities of PC's? Let's walk through your reply... First the book actively discourages DM's from using DC 25 and DC 30 except in extremely unusual and extremely difficult circumstances and instead advocates for a norm of DC's that range between 10 to 20... so I'm confused that you are interpreting it that way. From a reading there is no relative difficulty, it is instead an objective difficulty (though I and others have used it in a relative sense and posted to that effect on these boards but that seems to be the realm of homebrew)... Are CR's relative? I thought they were objective in that CR 10 is CR 10 regardless of the levels of the PC.... or am I misunderstanding your complaint?

Maybe a better way for me to approach this is to ask... what exactly needs to be discussed? Why would low level PC's shun all skill checks and high level character waltz past most of them... there has to be something I'm not grasping in your reply.

 

[Imaro]

Here we start to run into an issue for me. While this sort of framing of checks may be acceptable in some cases, if it is taken as an absolute requirement of the system, then it basically broke pretty much every way to do anything but binary hard pass/fail. Obviously said pass/fail could still entail a fail state that is effectively 'partial' (IE the guard doesn't ring the master alarm when the thief fails the stealth check right off).

I'm not grasping how a skill system that is used to resolve at the task level with a fail state and success state... means it has to be a hard binary pass/fail. I think the default of 5e is that but there is an option for success on a failure setting the precedent and underpinnings for a DM to homebrew out a more extensive system if they want it. IMO, having the task itself defined with a success state and fail state does not in turn mean there can't be gradations of both... only that those first two should be defined as part of the task. In fact I would go so far as to think that in order for there to be gradations one has to know what both success and failure mean...

 

[FrozenNorth]

And I'm going to stop you right there. There's a difference between a combatant and a front line combatant. The ability to e.g. fire a crossbow doesn't require that much.

At the very least it requires a crossbow. At 25 gp each, this is something that NO ONE in our small village of 10 families would have.

Let me stop you there. In 5e a hill giant is not just slightly bigger than an ogre. They're 16' tall according to the Monster Manual, and huge creatures rather than large. They're actual giants (and it's one of the 5e fluff changes I fully support).

The D&D SRD had them at 10.5’. Forgotten Realms wiki has them at 15’, so I woon’t debate the point.

If the giant is particularly homicidal they get the shepherd/goat herd. But the horn has been sounded.

Village of about 10 families, so there are maybe one or two herds of goats around the village.

Entirely possible that the giant’s approach on the village doesn’t cross the goatherd (in which case, see the surprise scenario), but let’s suppose the giant comes across the goatherd.

Goatherd blows his horn, gets eaten.

We're in hill country and the call has gone out for "giant". The weapons of choice are bows, slings, and javelins with hatchets as a last resort.

The call didn’t go out “Giant!”. A horn was blown, the goatherd was eaten. The villagers don’t currently know if the threat is a wolf or a giant, but they know an unspecified threat has attacked the goatherd.

They were in the fields. They return to their homes to grab weapons. Maybe they have some slings in addition to bows, but they probably don’t have javelins.

If we assume half your "non-combatants" are Str 8 Dex 8 and have slings so can do damage that's still 30 slingstones that need 10 or more to hit (vs AC 11; +2 proficiency, -1 stat bonus); the non-combatants between them do an average of 28.5dpr at short range (accounting for crits) or 13.8 dpr at long range.

Most farmers and homesteaders don’t send their noncombatants out to fight because they don’t want them to die, in addition to the fact that they are well, non combatants.

Also, I would expect that most able folk would have some sort of weapon, but I wouldn’t expect the non-combatants to.

No walls and organised militia needed. Just proficiency with simple weapons and not doing anything stupid.

It isn’t stupid. You are just operating with perfect knowledge of what is happening. You know that the goatherd blew his horn because his flock was attacked by a giant, and not because it was attacked by a wolf, or because the goatherd saw a squad of knights of evil Lord Redrum bearing done on the village. The villagers don’t have that knowledge, so they have to balance the risk of overreacting to a small challenge (their flock gets eaten because they called out the whole village and set out in force) with incorrectly approaching a larger challenge (hiding would be a better approach to a squad of mounted knights).

So what would be the villagers next step? Thanks to the valiant goatherd, the village is in a better scenario than the surprise scenario I described in my previous post. They have advance warning that something is happening, and have time to go home and grab weapons.

So the first group to get ready (maybe 10 or so) head to where the flock generally grazes. At this point, they are still not sure what the threat is, so if they act quickly they believe they might still save their flock. They come across the giant coming in the other direction.

The group has two options. They can attack now, or go back and get reinforcements. If they attack, they will probably win (unless their nerve breaks and they flee), but quite a few of them will die.

If they run back to the village, the giant, with his greater movement, will probably kill a couple of the retreat. They can join up with the other combatants (who are finally ready), and their odds of defeating the giant is pretty good, though quite a few villagers still get killed.