D&D 5E Poll: What is a Level 1 PC?

[Cadence]

Actually, even in a straight 3d6 world, a stat array comprised completely of 10's and 11's is more common than one with any bonuses and penalities, so that would be closer to the "average" (specifically, the mode) than those individuals whose poor results in certain abilities happened to be balanced out by better than average results in others.

Ah, but doesn't a stat array of all 10's and 11's only occurs around 24 in 100,000 times -- exactly as often as a stat array of all 12's or higher (or equivalently all 9's and lower)? For what its worth, having a sorted stat array equal to the median of the order statistics (7, 9, 10, 11, 12, 14) occurs around three times more often, approximately 72 in 100,000 times.

In any event, following the logic of your 2nd and 3rd paragraphs, the 7, 9, 10, 11, 12, 14 array seemed like a more useful guideline to me for quick-statting a "typical" NPC.

 

[Obryn]

3.x answers this with the "standard array" which is 13, 12, 11, 10, 9, 8. It's what generic humanoids and NPCs use. "Elites" get a 25-point buy.

-O

 

[FireLance]

Ah, but doesn't a stat array of all 10's and 11's only occurs around 24 in 100,000 times -- exactly as often as a stat array of all 12's or higher (or equivalently all 9's and lower)? For what its worth, having a sorted stat array equal to the median of the order statistics (7, 9, 10, 11, 12, 14) occurs around three times more often, approximately 72 in 100,000 times.

Interesting ... the math seems to check out, but the result is non-intuitive. I wonder why this should be the case. Hmmm...

 

[hamstertamer]

Your average blacksmith is a genius and smarter than about 80% of the people around him? And he makes bows? Why? Traps? Why?

Your average blacksmith is a level 1 commoner with Blacksmithing. This guy is way, way beyond average. Heck, he's already got elite stats which takes him, by RAW, well above average. Remember, average people have 10's across the board.

And a blacksmith with a 16 Int and a 12 Strength and a Con PENALTY? Really? This is believable?

Who said anything about an average blacksmith?

I don't know what your talking about either. I don't usually use standard arrays for creating NPCs unless their is a reason to and I create my own standard arrays as I see fit. I create the characters I want to create based on what is needed or just interesting to me.

And a blacksmith with a 16 Int and a 12 Strength and a Con PENALTY? Really? This is believable?

I don't know what's not believable. I really don't understand what your getting at. He's the 'royal blacksmith 56 year human' in my campaign he can be as intelligent as I see fit and have weak constitution too. It's completely up to me. I create their backgrounds.

Remember, average people have 10's across the board.

No they don't. The average ability scores for an average person is 10. That means, it could be a 9 for int, could be a 11 for con maybe even an 8 for dex and a 13 for cha. You get the drift I think.

If you used straight 10's all the time, well then that was your personal prerogative. I think that's a strange way to DM but hey that's your thing.

I think the point that went over your head was that you can make a master blacksmith with very few NPC levels. You don't have to have a 10th or 15th level expert to make a one, in fact you shouldn't make one, but I definitely like that I can if I so desired to. If you wore making 20th level expert master blacksmiths I would probably laugh but maybe that's just your game logic so good luck with that.

 

[slobo777]

Interesting ... the math seems to check out, but the result is non-intuitive. I wonder why this should be the case. Hmmm...

I'll try: In craps, making a 10 the "hard way" 5+5 is only half as likely as rolling a 6+4, because there are more ways of making that. Remember once you sort the array, you have hidden the 6! (=120) ways of arranging the results. In an all 10's listing, there are still those combinations to consider, but there are likely to be repeats with exactly the same rolls to generate them (e.g. 3,3,4) - i.e. some of the 10's are "the same", and reduce the number of different combinations down from 120. This effect more than offsets the fact that 10 is a mode on 3d6.

As in the real world, a "precisely average in all ways" person is in fact exceptional.

 

[FireLance]

I'll try: In craps, making a 10 the "hard way" 5+5 is only half as likely as rolling a 6+4, because there are more ways of making that. Remember once you sort the array, you have hidden the 6! (=120) ways of arranging the results. In an all 10's listing, there are still those combinations to consider, but there are likely to be repeats with exactly the same rolls to generate them (e.g. 3,3,4) - i.e. some of the 10's are "the same", and reduce the number of different combinations down from 120. This effect more than offsets the fact that 10 is a mode on 3d6.

Unfortunately, I'm not familiar with craps, so that example went over my head.

Anyway, I wasn't looking at the all 10's case - I was considering all combinations of 10 and 11.

I am aware of the multiple permutations factor (and 6! = 720, actually ) but I am rather surprised that it was enough to overcome the fact that the probability of getting a 10 or 11 on 3d6 is higher than the probability of getting a 7, 9, 12 or 14.

The probability distribution of 3d6 generated ability score arrays must be fascinating (and not easily represented in 3D space! ).

 

[Bluenose]

KM - my problem isn't that the peasant gained levels by killing goblins, my problem is that he gained levels at all. Why? Why is he gaining levels. If he's killed enough goblins to gain 3 levels, he's been in literally dozens of life threatening fights. Why isn't he dead? It's ridiculous. If your farmers are fighting off the goblins time and again, why do they need the PC's?

3e linked skills to levels. Arguably, with Thieves in mind, so did AD&D/BD&D. Every edition of D&D linked level to combat ability. So when you're making NPC classes, and giving them levels, and keeping to the way D&D "always" did things, the result is that anyone unusually skilled in a particular skill has to be good in a fight.

 

[Bluenose]

Unfortunately, I'm not familiar with craps, so that example went over my head.

Anyway, I wasn't looking at the all 10's case - I was considering all combinations of 10 and 11.

I am aware of the multiple permutations factor (and 6! = 720, actually ) but I am rather surprised that it was enough to overcome the fact that the probability of getting a 10 or 11 on 3d6 is higher than the probability of getting a 7, 9, 12 or 14.

The probability distribution of 3d6 generated ability score arrays must be fascinating (and not easily represented in 3D space! ).

Well, the probability that you'll roll a 10 or 11 on 3d6 is 54/216 (1/4). The probability that you'll do that 6 times in a row is (1/4)^6. That's pretty low.

 

[delericho]

A 4th-level commoner would probably beat a 1st-level fighter if you're using WBL. Using the standard 25 point buy (which is ridiculously weak but whatevs)...

Two things:

1) 25-points is the standard value for PCs. NPCs use a 15-point spread, with the 'standard array' being either 11/11/11/10/10/10 or 13/12/11/10/9/8.

2) Yes, if you use the WBL to kit the Commoner out for adventuring, he would indeed beat the Fighter. The problem with that logic is that the Commoner shouldn't be equipped for adventuring because he's not an adventurer. Most of the equipment he gets by virtue of WBL is tied up in his farm equipment, his crop, the pigs he keeps to slaughter in the winter, or whatever. A higher-level Commoner (farmer) has greater wealth by virtue of running a bigger and more successful farm.

Your average blacksmith is a genius and smarter than about 80% of the people around him? And he makes bows? Why? Traps? Why?

And a blacksmith with a 16 Int and a 12 Strength and a Con PENALTY? Really? This is believable?

Nitpick - the example given wasn't an "average blacksmith". Perhaps more to the point, though - his stats have been adjusted by virtue of aging effects, hence the higher-than-expected Int and lower-than-expected Con.

That's not the example blacksmith I'd build, and neither is it really a good example of a "royal blacksmith" according to the demographics tables of the DMG (which are insane, as I've said previously). But I don't think that those stats in and of themselves are too outrageous.

Your average blacksmith is a level 1 commoner with Blacksmithing. This guy is way, way beyond average. Heck, he's already got elite stats which takes him, by RAW, well above average. Remember, average people have 10's across the board.

I'm going to disagree here, as well. The DMG demographics tables would seem to give even the smallest villages a smattering of low-level Experts, and the blacksmith is one of the most likely professions to be represented. So, a 1st level Expert. And, likewise, I disagree with the "10's across the board" - as someone who works with fire and metal all day, the blacksmith is very likely to have both Str and Con slightly above the average.

My take on a typical blacksmith:

Duncan, the blacksmith
1st level Expert
Str 12, Dex 10, Con 12, Int 10, Wis 10, Cha 9
Skills: Appraise +4, Craft (smith*), Gather Information +3, Handle Animal +3, Knowledge (local) +4, Listen +4, Spot +4
Feats: Martial Weapon Proficiency (hammer), Skill Focus (craft-smith), Weapon Prof (simple), Armour Prof (light)
Hit Points: 7 **
AC: 12 (treat smith's apron as leather armour)
Fort +1, Ref +0, Will +2
Attack: Hammer +1 melee, 1d6+1 damage

* I'm assuming this is one skill. If general blacksmithing, armoursmith, and weaponsmith are separate skills, he should take each of these individually, in preference to Appraise and Knowledge.
** I've given him max for 1st level, because that's what I did with my other sample NPCs. If not maxing out that first die (which, to be fair, is probably right) drop this to 5.

I think that's a reasonable smith. It's obviously not the only possible way to build that character (and it's still way more effort than it is really worth), but it does the job and feels about right. YMMV, of course.

 

[delericho]

Something that occured to me while statting up the blacksmith:

I had a bunch of extra skills available with no clear use for them, and so I assigned them to Gather Info and Knowledge(local), on the grounds that the blacksmith is likely to have more interaction with more people in the community (and with travellers) than almost anyone else other than the innkeeper.

And that is potentially interesting information - it means that if the party want to consult with people about local events (and not just when finding a specialist to shoe their horse, or whatever), it gives a guide as to who they should speak to. If the village were properly statted up, that genuinely could give rise to a deeper sense of versimilitude than you might otherwise get.

But...

That only works if the DM basically has the whole village statted out, and so knows that the blacksmith is knowledgeable about such things, that the parson's wife is a notorious gossip, and so on.

And the problem is that, even at 2 minutes per set of stats, a village of 200 people will require 6 and a half hours to stat up fully - and virtually none of the data thus generated will be of any use.

(And you can't really automate that process meaningfully - it's only helpful if it throws up ideas and knowledge for the DM to use in his game to generate that versimilitude. Buying a book of stats, or having a piece of software do it won't give the DM that same intimate knowledge.)

Just an idle thought...

Edit: I suppose perhaps you could have a handful of pre-generated skeletons of NPC stats, perhaps with everything assigned except one 'wild-card' skill, plus a few random tables giving names and positions with the village. And so, when the PCs ask about strange goings-on by the bridge, the DM rolls on his 'villagers' table to get "Alice, daughter of a local farmer", who he decides is a "Commoner 1", grabs one of his skeleton NPCs and assigns "Listen +4" as her 'wild card' skill. I suppose that might give the illusion of versimilitude (the illusion of the illusion of truth? ) without an outrageous amount of work. Not least since the DM wouldn't need a different table for every village - he could just reuse the same table, and rename "Alice" once her entry has been used...