D&D 5E My nearly comprehensive critique of the August Packet

[Remathilis]

... Aside from that, how did you enjoy the play, Mrs. Lincoln?

 

[Kobold Stew]

My personal thinking is that if you want to reinforce stat biases, reinstate racial maxs and mins. Let the elf get a 22 Dex, but he can only have a 16 Con. The dwarf can get a 22 Con, but she's limited to a 16 Dex.

This would be a good way to go, but I suspect that players would revolt. Because once one starts down this path, it becomes necessary to limit the strength of small races. If we were to extend the table, I'd suggest something like this:

Humans: Max 20
Elves: 22 Dex, 16 Con
Dwarves: 22 Con, 16 Dex
Halflings: 22 Dex, 16 Str
Gnomes: 22 Int, 16 Str
Half-Elf: 22 Cha
Half-Orc: 22 Str

I'd actually be fine with that. (Though what if the cap were at 14 (so that it represented a hard limit for more players), rather than 16?) But the game seems determined on having small and medium-sized creatures interchangeable. That's never felt right to me.

 

[N'raac]

A racial adjustment has some impact if stats are rolled, in that a 3 - 18 does not allow a Dwarf to start with a 17 CHA, while he can have a 19 CON that a human cannot.

It also has a point buy impact - with +2 CON, that Dwarf can pay for a 14 and get a 16 - he's not likely to have an 8 CON. But if he has a -2 CHA, and it costs, say, 1 point to add a point from 8 to 13, 2 points for 14 and 15, 3 for 16 and 17, and 4 to buy a pre-adjustment 18, will the Dwarf spend 17 attribute points on an 18 CHA to have it drop to a 16, or 10 on a 16 CON that gets bumped to an 18?

I suggest that players looking for a high CON character will be more likely to select a Dwarf, and players seeking a high CHA character will be less likely to select a Dwarf, so the game will tend to feature more high CON dwarves than high CHA dwarves.

 

[Li Shenron]

If all of those enemies have treasure equivalent to the PCs, the PCs' wealth just quadrupled.

Yes, I think I already mentioned this too, that is something I had issues with, NPC needing to have magic equipment as well, PC looting them etc. In general, that NPC equipment itself can be kept in line with the wealth-by-level guidelines, so that when the PCs defeat NPCs they loot equipment but don't find chest treasure (or find less) and when they defeat monsters they find lair treasure. Or alternatively, you can just balance the amount of NPC fights and monsters fights. Mechanically, you still manage to follow the wealth-by-level. However, it restricts your freedom... because for example if you want to have a campaign where you only fight monsters, then you must put lair treasure or you must have the PC receive money from someone for killing monsters. It's not that there aren't options to look for, but still as a DM I am not complete free, because I just had to keep those wealth-by-level as much as possible (although I understand, that you were able to make it work even when deviating).

Conversely, you have an economy set up for goods and services, and magic items completely break that economy, as has been covered elsewhere, by being ludicrously expensive but implicitly not all that rare.

Yes indeed... I was fine at first with this economy assumption, but as I say, it is a defining feature of a fantasy setting. After a while, I wanted to run adventures in a world that works differently from this, and that's when the issues started.

This is why I wouldn't call these problems "mechanical", because it works as long as you stick with the assumed fantasy setting type. Maybe there's a better way to call them.

Why is poison not for good characters? How is it different than magic, or sneak attack, or whatever?

It's an old issue. There are at least 2 very different ways to think about poison in the game: there is the combat-oriented poison (e.g. coating a weapon with poison, a poison spell with immediate effects...) and there is the story-oriented poison (e.g. poisoning food, potions, and anything which has a long-lasting effect and often also a significant initial delay).

These 2 are completely different things, and yet both gamers and designers mix them up all the time!

Story-oriented poison is for me an essential element in fantasy storytelling. Poisoning people in this way is pretty much in all RL cultures seen as an evil and dishonorable act, at least because:

(a) it's a deceit: it's not a fair act of war or like drawing a weapon (which makes your intentions clear and leaves the target with an option of surrendering or running away)
(b) it causes unnecessary suffering: many poisons are slow, almost all poisons are non-instantaneous (at least not as swift as a weapon strike)
(c) it is a cowardly act, because the poisoner usually does not reveal itself (poison is also most often used by someone who wouldn't stand a chance in a fair fight)

But if you take a look at how poison is represented in D&D, it normally has NONE of these features. Most of the times it's just added damage, or an immediate penalty that actually shortens the battle. You could even argue that it is more mercyful because it ends the fight more swiftly.

That's the problem: poison in D&D is mechanically designed as combat-oriented poison, but it is still traditionally labelled "evil" as story-oriented poison.

I think story-oriented poison can be represented very well with a mechanic similar to 3ed diseases, even better if generalizing the time frame so that the DM can use hours instead of days when appropriate. Then I would be in favor of labelling that kind of poison with "evil" or at least "non-good".

Personally, I think Treasure Tables are not the best way to assign treasure to monsters. That includes the random distribution ones in the DMG and last page of the MM. Just as Wandering Monster checks are now being based off of population density in an area (making that area more dangerous to explore) Treasure might best be based off of creature history.

I also prefer to give treasure and random encounters a reason based on the setting.

However, I recognize that rolling random stuff can be a lot of fun too, and it has a certain "old-school" vibe, feels very D&D-traditional... I myself like sometimes playing a campaign with plenty of random stuff

Overall the important thing is that these tables are not perceived as mandatory. They are great for a certain gamestyle but not suitable for another.

This would be a good way to go, but I suspect that players would revolt. Because once one starts down this path, it becomes necessary to limit the strength of small races. If we were to extend the table, I'd suggest something like this:

Humans: Max 20
Elves: 22 Dex, 16 Con
Dwarves: 22 Con, 16 Dex
Halflings: 22 Dex, 16 Str
Gnomes: 22 Int, 16 Str
Half-Elf: 22 Cha
Half-Orc: 22 Str

I admit I like this idea a lot! But shouldn't it be 18?

If using 18 instead of 16, the max difference between 2 characters (using core races) in a specific stat is 4 which equates to +2. In bounded accuracy, +2 maybe it's quite a lot, but maybe it is still feasible.

The most important thing IMHO is that the lower cap does not completely invalidate the choice of class. For physical stats, I really think that a Fighter with max Str18 or a Rogue with max Dex18 is still a very good PC (and it if it wasn't, it would mean that it is a MUST for every Ftr to reach the cap in Str, and I would hate to play a game like that!).

However in 3ed unfortunately that was a huge limit for spellcasters. If your spellcasting stat was max 18, you were sure that never you could have cast 9th level spells. That's a big deal to me! I don't remember the current playtest packet, but before in 5e there is no such limitation. If e.g. capping a Mage Int at 18 does not limit the spells, but only applies to spells DC and other parameters, then it's OK. Otherwise, they have to make sure no race has a low-cap mental stat.

 

[pemerton]

I really think that a Fighter with max Str18 or a Rogue with max Dex18 is still a very good PC (and it if it wasn't, it would mean that it is a MUST for every Ftr to reach the cap in Str, and I would hate to play a game like that!).

I can't comment on the rogue, but I think there game will generate considerable mechanical pressure to take a fighter's STR to 20.

Consider a 10th level fighter. A quick scan through the current bestiary reveals 10th level creatures with the following ACs: 12, 13, 14, 14, 15, 15, 16. That's an average of a bit less than 14 - we'll call it 14 for simplicity's sake.

A 10th level fighter has +4 to hit from class. Let's add a +1 weapon - then a longsword is +5 to hit for 1d8+1 damage before STR.

With STR 18, vs AC 14, that means the fighter hits on 5+ for 1d8+5: average DPA is .8*9.5 = 7.6 before crits. Of those hits, 3 in 16 are crits doing an additional 4.5 on average, so overall average DPA is 8.44375.

With STR 20, vs AC 14, the fighter hits on 4+ for 1d8+6: average DPA is .85*10.5 = 8.925 before crits. Of those hits, 3 in 17 are crits doing an additional 4.5 on average, so overall average DPA is an ugly fraction that, for convenience, I'm rounding up ever-so-slightly to 9.72.​

That's a little over a 15% boost in DPA for pushing STR from 18 to 20. Allowing for the 0.8 hit rate, the 18 STR fighter gets just a touch more benefit from boosting STR to 20 than s/he would get from an item or feat granting +1.5 to damage. I think that would be a fairly attractive item.

The more important the crit is as a component of damage, the less that STR difference will matter (because the crit rate is STR-indpendent), and the bigger the damage dice the less the STR component matters. If I do my comparison with a greataxe fighter, for instance, it looks like this:

With STR 18, vs AC 14, that means the fighter hits on 5+ for 1d12+5: average DPA is .8*11.5 = 9.2 before crits. Of those hits, 3 in 16 are crits doing an additional 9 on average, so overall average DPA is 10.8875.

With STR 20, vs AC 14, the fighter hits on 4+ for 1d12+6: average DPA is .85*12.5 = 10.625 before crits. Of those hits, 3 in 17 are crits doing an additional 9 on average, so overall average DPA is an ugly fraction that, for convenience, I'm rounding down ever-so-slightly to 12.21.​

But even here, the increase in STR is granting a little over a 12% increase in expected DPA.

Now comparing with advantage on the attack:

With STR 18, vs AC 14, that means the longsword fighter hits 24 in 25 times for 1d8+5. Of those hits, 111 in 384 are crits doing an additional 4.5 on average, so overall average DPA is 10.42 (very slight rounding down).

With STR 20, vs AC 14, the longsword fighter hits 391 in 400 times for 1d8+6. Of those hits, 111 in 391 are crits doing an additional 4.5 on average, so overall average DPA is 11.54 (very slight rounding down).​

That's still better than a 10% increase in DPA. And for the greataxe fighters, it looks like this:

With STR 18, vs AC 14, that means the longsword fighter hits 24 in 25 times for 1d12+5. Of those hits, 111 in 384 are crits doing an additional 9 on average, so overall average DPA is 13.64 (very slight rounding down).

With STR 20, vs AC 14, the longsword fighter hits 391 in 400 times for 1d12+6. Of those hits, 111 in 391 are crits doing an additional 9 on average, so overall average DPA is 14.77 (very slight rounding down).​

That's a bit over an 8% increase in DPA. At this point, I'll concede that there's not much at stake between the two scores. But even here, it's still granting better than +1 to DPA, which is in turn better than a +1 damage bonus (because, given some attacks miss, that's less than a +1 expected DPA).

Now DPA is not the be all and end-all; but for a fighter it's not nothing, either. Of course defence matters too: the longer you can stand up, the more rounds you have to deliver your DPA (though killing the monsters quicker helps your friends with weaker defences). And non-warrior fighters have other tricks to deploy besides damage. But unless those tricks (like the gladiator's Dirty Trick) give you a reliable way of getting advantage on attacks (and of course the gladiator doesn't get the increased crit range, which is very strong with advantage and is helping to dilute the importance of STR in the above examples) I still think you're going to want to max out STR.

Making STR add to damage only would be a more radical way of significantly closing the gap. For the longsword fighters without advantage, the DPA boost would be a bit less than 9%, and for the greataxe fighters with advantage it would be less than 7%. At that point, boosting STR would still be a viable option for improving your fighter, but I think other options would start to look more viable too.

EDIT: A very basic comparison to the rogue:

A 15% boost in effectiveness is going from needing a 15+ to succeed on a check, to needing a 14+. (Success rate goes from 6/20 to 7/20.)

An 8% boost in effectivness is going from needing a 9+ to succeed on a check, to needing an 8+. (Success rate goes from 12/20 to 13/20.)

I think if, as the player of a rogue, most of the checks I faced were 15+ for success, I would want that +1 pretty badly. Whereas the boost from needing 9+ to needing 8+ I could more easily live without.

4e maths is based around 9+ as the basic threshold for success. If D&Dnext is carrying that logic into its maths, than rogues may not be under the same pressure to max DEX as fighters are for STR, especially if they can get advantage on checks fairly easily.

It's the unrelenting nature of the hit point attrition game, and the fact that it adds to both attack and damage, that makes STR so important for fighters.

 

[Li Shenron]

Very nice analysis [MENTION=42582]pemerton[/MENTION], but if a player is obsessed with performance/optimization to the point that he thinks he cannot play well enough with a Str18, maybe he should just pick a race that does not have a lower Strength cap.

 

[gyor]

Honesty I disagree with most of the ops thoughts and stat miniums too, no way they'll do that, the players freaked out over a -1 to ability depending on race.