Reducing the HP Discrepancy

Currently the difference between a high and low HP character is so vast that some may die from a single attack from a lowly kobold while others can survive a dozen blows.  The problem is only exacerbated by the Constitution modifier being added each level.  A 10 Con wizard may have 50 HP by level 20 while the 18 Con fighter may have close to 200.  This makes adventure balance very fragile and combat very swingy.  Furthermore you have HD which have multiple meanings and seem to be a somewhat confusing healing mechanic.

I suggest finding a compromise between 4e HP and the current 5e HP iteration.  Base HP for a level 0 creature (either a level 0 PC or low level monsters such goblins and kobolds) would be 1/2 their Constitution score.  High HP classes would gain 5 HP per level, medium HP classes would gain 4 HP per level, and low HP classes would gain 3 HP per level.  A level 20 fighter with 18 con would have 109 HP.  The level 20 wizard with 10 con would have 65.  The fighter would have 14 HP at level 1 while the wizard would have 8. The constitution modifier would no longer play a roll in determining HP.  

This method of HP solves a few problems.  It eliminates the swingy nature of level 1 combat by making combatants on average able to take 2-3 hits before going down.  This method also keeps classes HP more bounded which makes monster design easier. You no longer run into the problem where the monster will auto kill Tim the wizard but barely scratch Bortho the fighter.  Lastly it removes luck from permanently cripiling or enhancing your character beyond the expected values for a character of their level.

Now I get that some people like rolling for HP.  HD healing help provide the best of both worlds.  They allow you to roll for HP over and over.  Far more than just once per level.  The issue with HD right now is they represen a poor stamina/endurance mechanic because they run out far too quickly.  Perhaps PCs should receive bonus HD at level 1 equal to their Constitution modifier.  This would help keep Constitution important even though it no longer increases Max HP after character creation.  HD would stay at their current value for the classes or be consolidated to High: d10 Medum: d8 and Low: d6.

My 5e Homebrew Material

The Warblade: A Mythic Fighter

The Hero: A Modular Class

I like the general idea.  What I'd like to see is something very simple like:
1st lvl = Constitution Score
Every Other lvl = +1 Hit Die (or take the average+1, if you don't want to roll - 3 on a d4, 4 on a d6, 5 on a d8, 6 on a d10)

Use that nice and simple system right there and design the game around those hp levels.

That's still a large difference, but, it addresses the low hp problem at first level, and I don't think D&D will ever loose rolling for hp, so thus you have the second part.  A first level fighter might have 15hp, while that same 20th level fighter might have 129.  Really, it's very similar to yours, but a tiny bit more simple without the halving of the constitution score, and with the tradition of using hit dice for those that want to roll or the "take the average" approach.
I like the general idea.  What I'd like to see is something very simple like:
1st lvl = Constitution Score
Every Other lvl = +1 Hit Die (or take the average+1, if you don't want to roll - 3 on a d4, 4 on a d6, 5 on a d8, 6 on a d10)

Use that nice and simple system right there and design the game around those hp levels.

That's still a large diference, but, it addresses the low hp problem at first level, and I don't think D&D will ever loose rolling for hp, so thus you have the second part.  A first level fighter might have 15hp, while that same 20th level fighter might have 129.  Really, it's very similar to yours, but a tiny bit more simple without the halving of the constitution score, and with the tradition of using hit dice for those that want to roll or the "take the average" approach.



During the entirety of 4th edition, rolling for HP didnt exist, not even as an option. D&D already has lost rolled HP for an entire edition.

 
...whatever
...that is quickly being swept under the rug and being replaced by a new edition... ;)

Still, what 4E had was the average of the old hit die + 1, IIRC.
The problem with rolled HP is twofold.
1) it can permanently effect your character if either very negative or very positive ways, both of which ruin balance especially with bounded accuracy and HP being king.
2) It keeps high and low HP classes too far apart. If someone has a d12 and someone else has a d4 the variance is just too great. The 3/4/5 approach makes monster design easy. If we assume a PC can take 3-4 hits before dropping then monsters can scale up at about 1 point of damage per level to remain relevant. Someone who has 3x the HP of another player can just laugh off anything dangerous to the low HP class, but anything dangerous to the high HP one will be an 1 hit KO.


During the entirety of 4th edition, rolling for HP didnt exist, not even as an option. D&D already has lost rolled HP for an entire edition.

 



Yeah rolling for HP should be one of those concepts that is left in a fire to die, as should all "Screwed for life" mechanics.

There's nothing exciting about having a 3rd level fighter who rolls a couple 1s for HP and has less HP than the party rogue. There's no reason for the game to have that possibility at all.

In 1st and 2nd edition we always use to have the DM AND the player roll the dice for HP at a new level and take the best. You didn't get a 1 very often, but when you did clearly the universe wanted you to.

In 3rd we rolled, but took 1/2 maximum as the "floor". (So a fighters d10 roll would have a range of 5, 5, 5, 5, 5, 6, 7, 8, 9, 10).

And in 4th we never had to roll.

Interestingy, all of those were equally satisfactory.
I've never found the difference between low hp characters and high hp characters to be a problem. It's quite true that a low level wizard can be one shot, but that's part of their tradeoff, and it encourages a player to play carefully. Traditionally, a wizard type character has been weaker in the beginning, but grows to become a powerhouse later levels, while the high hp characters trend in the other direction. Again, I don't see it as a problem, but a challenge that a player has to overcome. I would never want to see hp "balanced" to a degree where you may as well not have any difference between the classes

Set amounts of hp gain per level have always seemed far less interesting to me. It takes alot of variance out of the game which some people find enjoyable, it removes a moment of tension as you wait to discover your fate on the dice roll, and it removes one thing that can distinguish your character from every other character of it's type at the table. We've all heard stories of the ridiculously high hp fighter who's player always rolled near max and who could live through almost anything. There are also stories of the fighter who rolled low too often, but who's player fought to overcome this disadvantage through smart play, careful item and feat choices.

In short, instead of looking for more hp to help a character survive, I'd rather see that player look for options to keep his character alive. In my experience, smart play, smart item choices, and smart spell usage can keep any character alive. In additon, players should take some responsibility for their play. A glass cannon wizard shouldn't be toe-to-toe with that monster who's dishing tons of damage per turn. If he is, he frankly deserves to die.


Traditionally, a wizard type character has been weaker in the beginning, but grows to become a powerhouse later levels, while the high hp characters trend in the other direction.


That is a problem, since "Get Mauled by Housecats Now, Destroy the Universe Later" is not balanced.
Fourth edition did loose the rolling for hit points deal. Even if the numbers came from the old system, rolling didn't exist. I might be a little biased since I prefer 3.5 over 4th, but I prefer rolling for hit points. In the end it should average out except in rare cases. The inherent problem is obviously that the wizard will fall significantly behind the fighter, where if you don't roll this is less of an issue.


If game balance is the desired result, then you shouldn't roll for hitpoints. However, 4th edition solved the game balance issue perfectly...and is being tossed away after a few short years. All in all I would prefer one approach and end the modularity stuff. As written D&D Next allows the game master to choose if you roll for hit points or take the middle +1 approach. Since there is a topic claiming D&D Next should take the consistent approach, it appears the modularity design idea has failed horribly as it hasn't been paid attention to and the assumption is the defualt is the way the game is to be played. The optional additional ways to play the game seems to not exist.
I actually really like the modular approach. I think admitting that every DM runs their game in a different way, and one size doesn't fit all is a good idea. I've been running long enough that I'm not affraid of tinkering with rules sets if I positvely hate certain rules, but the modular approach makes that more "ok" to people who are affraid they're going to break something if they tinker. Even something like rolling hit points or not, or rolling stats or not can get divisive, so I'd rather just see a multitude of approaches, and let the DM customize their game.
I have no problem with HP dicrepancy. To me a fighter should have double a wizard's HP.

As for rolled HP. When I did rolled HP. I reroll all HD every level up.

Orzel, Halfelven son of Zel, Mystic Ranger, Bane to Dragons, Death to Undeath, Killer of Abyssals, King of the Wilds. Constitution Based Class for Next!

I have no problem with HP discrepancy. To me a fighter should have double a wizard's HP.

However, the fighter shouldn't have five or six times the wizard's hp, on top of an armor class that gets him hit about a a fifth as often.

I suspect/hope they will provide several options, or encourage you to find you own HP comfortability zone.

Con score at 1st level, 1 (or 2, or 3) HD at first level, and everything in between (cool, there are so many ways to go).
I have no problem with HP discrepancy. To me a fighter should have double a wizard's HP.

However, the fighter shouldn't have five or six times the wizard's hp, on top of an armor class that gets him hit about a a fifth as often.

It's a good thing, then, that he won't. The only way for a fighter to have five or six times the wizard's HP is if the fighter puts every resource possible into increasing HP and the wizard takes every option possible for reducing it. If that happens, then yeah, I think it's okay for the fighter to take five times as many hits. But that's not what happens, and if it does happen it's something that the wizard player very specifically went out of their way to make happen.

I can understand (and even personally belive) the notion that it's not great for there to be a portion of the game where certain characters are likely to be killed in a single hit from even a minor threat. That I understand.

I would understand compelling arguments that for various reasons 3/4/5 is a better ranges of "hits to kills" than, say, 2/4/6. I think it's very good to challenge existing setups, rather than just blindly swallowing them because that's how we used to do it.

What's less compelling to me is that the notion that two extreme characters having the property that one takes four times as many hits to go down is intrinsically an issue. 3.5 had a bargeload of mathematical dubiousness. It's be possible to rattle off the mathematical dubiousnesses for days. One thing that I didn't have major problems with, though, was HP discrepancy. (Distinct from the related but wholly separable issue of characters getting killed in a single hit.)

Essentially, what I still don't understand is why a character being able to survive three times as many hits as another character is an issue. I agree that it sounds like it might be one; anything that's of the form "several times more" might raise alarm bells. But what actually breaks down when it's the case? The guy who devoted all of his resources to having a lot of HP can take a lot of hits. That seems like the kind of thing that should happen.
Dwarves invented beer so they could toast to their axes. Dwarves invented axes to kill people and take their beer. Swanmay Syndrome: Despite the percentages given in the Monster Manual, in reality 100% of groups of swans contain a Swanmay, because otherwise the DM would not have put any swans in the game.
Rolling or taking a static value should be a tableside choice, made by the group.  Will rolling for points be potentially more unbalanced, leaving some characters one-shotted while others are rolling (HAH!) in hit points?  Sure, but that is the risk they took by deciding to roll, and presumably they are ok with that.  The fact that I prefer static boosts shouldn't dissuade anyone from rolling if that is what they prefer.

Now, with regard to overall HP calculation, I think HP should start much higher and be raised far slower.  Probably CON score + HD at 1st level (or something like that), and then just +HD at each level.  Fighters start higher than wizards, and gain faster, but not overwhelmingly so.

By the way, that "start higher, gain slower" idea is a paradigm I'd like to see applied to a lot of mechanics right now.  Casting (start with more spells known/castable and gain fewer over time), Combat Superiority, etc.
Essentials zigged, when I wanted to continue zagging. Roll dice, not cars.
OP:  Your system makes CON next to meaningless.  A CON of 18, at level 20, gets you 5 HP over a CON of 8.  For a fighter, that's less than 5% more HP, when he's spent every one of this character resources to achieve the highest HP possible.  You'd be a fool to have a CON higher than 12, and a fool to have one higher than 8 if you were starting above first level.  At least in 4e it was full CON and you got more surges, and even then you didn't build CON for the durability (at least past 13).  I'm a 4e fan, and even I thought going back to "all attributes get you something" was a good idea (except intelligence, which sadly is still useless unless you're a wizard) .

I also don't think it's fair to compare CON 18 fighter to CON 10 wizard.  Very few if any fighters will have 18 CONs, very few if any wizards will have 10 CONs, and those who do deserve the fruits of those decisions.  If we compare a CON 12 wizard with a CON 16 fighter, the gap is more like 2-3x rather than 4x as you discuss (interestingly, the fact that you round up if you're not rolling makes a huge difference in reducing the disparity, that might need to get looked at).  Is 3x too much?  In most of the games I've played, the fighter is getting attacked more than three times more often than the wizard (probably more than 6x, if you want to take the higher AC into account).  And even that's allowing for a huge disparity in relative resource investment, it drops to less than 2x if both have 14 CONs.  

Lesp has a point that the disparity isn't really an issue directly.  It becomes an issue when monster damage - in order to challenge someone at the high end - becomes so high that the person at the low end is at too great a risk of being one shotted.  I'm not a fan of risks of being one-shotted, but to my mind that has less to do with the fighter/wizard spread than the fact that the wizard is just too low.  I don't mind if the fighter has 2x as many HP as the wizard for a given investment in resources, so long as the wizard can consistently survive being hit once.

My preference?  10  + level * (3+CON mod for the low end like wizards, 4+CON mod for the middle types like clerics/rogues, and 5+CON mod for the high types like fighters).  You get real, meaningful benefit out of CON, not too much variation in equivalent-investment HP but not too little either, and starting HP are well beyond the range of a d12 so there's no one shot kills.  Monster damage should then be adjusted to take about 1.5 hits to knock out a low HP class, 2 for a middle HP class, and 2.5 for a high HP class.  In other words, average about 5+3*level, with a range of about +/- (4+1/2 level).  Thus, even two weak hits will fell a wizard but a rogue/cleric could survive and a fighter could take a third, one strong hit could not fell a wizard but could compensate for a weak follow up to fell a rogue/cleric, two strong hits could fell a fighter but it would take three average hits or four weak ones.  Damage rolls therefore matter, nobody is too fragile ever to take risks, and nobody is too sturdy ever not to.  Low hits per 0 but always more than one means a full HP character can take risks, but a character that's been hit even once needs to start playing smart and can feel the danger (although a fighter who's been hit weakly can still keep going unphased).  You can bump yourself up or down a level by investing in a 16 or only a 12 CON, but even a fighter with an 18 CON can still only survive 3 average hits, 2 strong ones, and four week ones.  He's not some unstoppable juggernaut.  A wizard only needs to invest in a 12 CON to keep up with the 1.5 hit rule and stay out of the risk of being one shotted, but with a 14 he can survive a second weak hit by level 10 and if he decides to run marathons in his spare time for a 16 CON he's as tough as the average rogue/cleric.
How about hitpoints are equal to CON+level.  Nice and bounded there.
OP:  Your system makes CON next to meaningless.  A CON of 18, at level 20, gets you 5 HP over a CON of 8.  For a fighter, that's less than 5% more HP, when he's spent every one of this character resources to achieve the highest HP possible.  You'd be a fool to have a CON higher than 12, and a fool to have one higher than 8 if you were starting above first level.  At least in 4e it was full CON and you got more surges, and even then you didn't build CON for the durability (at least past 13).  I'm a 4e fan, and even I thought going back to "all attributes get you something" was a good idea (except intelligence, which sadly is still useless unless you're a wizard) .



It isn't strictly true to say that CON becomes next to meaningless: it still affects skill checks and still boosts hit dice healing to a signifcant degree and I'm hopeful that they will introduce something akin to skill powers/traits that can be applied to higher scores.  4e understood the concept of using the healing surge pool as a pool of additional hit points that can only be utilised to a limited degree in combat.  This is a more balanced approach than front loading hit points so that fighters are at a much reduced risk in combat.

I'd also suggest applying the CON bonus to any additional hit dice obtained from the Toughness feat.

in the first game of Next my group played, the wizard rolled for stats and ended up with a max of 2 hp because of low Con...

I rolled for stats and got a Con 18 

Ironically, i was the one that got one-shotted in the first round of the first combat we ever played with the system and the whole thing pretty much soured me on the idea of "old school" hp. No wizard should start the game with an hp of 2 and no one with a Con of 18 should go to negatives in one hit. 
I also don't think it's fair to compare CON 18 fighter to CON 10 wizard.  Very few if any fighters will have 18 CONs, very few if any wizards will have 10 CONs, and those who do deserve the fruits of those decisions.

You make it sound as though a stat value of 10 is low, when it is actually the average around which the entire system is built!  

I will agree that CON behaves very poorly as a negative modifier to HP, though.  You can't ignore it, like you can with the STR or DEX.  They really should dismantle the universal ability modifier, go back to the more nuanced system that they had in AD&D, and let CON contribute to HP at something like CON/4 per level:

CON 1-4: +1 HP per hit die
CON 5-8: +2 HP per hit die
etc.

The metagame is not the game.

You make it sound as though a stat value of 10 is low, when it is actually the average around which the entire system is built! 



That's not actually true.  The system is built around 4d6 drop the lowest (or point buy generating equivalent), not 3d6.  That means that the average PC stat (before racial modifiers) is 12.24.


Sorry, I'm just being pedantic.
The difference between madness and genius is determined only by degrees of success.
I for one like +Hp per level from Con. It makes con an important stat. 4E still kept Con important because in 4E you could use Con to attack your enemies.

The problem is that 18 Con gives you +4 Hp per level when a wizard only averages 2.5 HP per level. A Wizard with 18 Con nearly triple his hitpoints a cleric or fighter nearly double thier hitpoints with 18 con. But if the base Hp was Higher say if a wizard averaged 6 hitpoints per level. Then the con bonus is not that huge. But all other damage in the game needs to be higher to compensate for characters having higher HP.

So far in 5E Con does not look like a very imporatant stat, and Dex looks way too good.
That's not actually true.  The system is built around 4d6 drop the lowest (or point buy generating equivalent), not 3d6.  That means that the average PC stat (before racial modifiers) is 12.24.

Even if the rules suggest playing with above-average characters, the system itself is designed around 10 being average.

That might mean something more, once the game is nearly completion and they focus more on balance.  It's entirely possible that they'll do a math pass and design the game to be played with above-average stats, as was the case in 4E, but there is yet no indication that such will be the case.

The metagame is not the game.

Saelorn

I fail to see the relevance of average NPC stats to PC hitpoint disparities.  The only things that are relevant to to hitpoint disparities are PC hitpoints (and by extension PC CON), and NPC damage.  Unless NPC damage is determined by CON, NPC average CON is irrelevant.  We could arbitrarily say that human average is CON 250, but PCs still average 12.24 or use the array where median is 12.5 and monster damage still averages whatever it's supposed to average, and it would have absolutely no effect whatever on wizard/fighter HP disparities.  Average NPC CON is modestly relevant to encounter length, and by extension the appropriate average PC HP, but not the standard deviation from that average and not even much to the average given how much other noise there is in that equation (monster HD size, accuracy, damage...).
I'd like to see characters start with hit points equal to their Consiittution score plus a bonus from their class. After that, I'd have both hp and damage scale very slowly. The hp of chracters should scale enough that they become measurably more powerful, but not so much that they can laugh at vials of alchemist's fire being thrown on them, or survive being dunked in lava or falling off a 100 ft. cliff. In past editions, what does it matter if you fall from any height once you're high level? It's "just" 10d6 damage! Likewise, a round of immersion in lava is a mere 10d6 damage also, and high level characters could potentially survive several rounds of swimming in the stuff, just as they can survive being hacked dozens upon dozens of times with an enemy's axe, while a 1st level character could die from stubbing his toe. Ridiculous.

And before anyone comes at me with the "but HP are an abstraction!" retort, don't bother. I've heard all the excuses before, and it's still absurd. You might be able to say that hits in combat don't really mean the character was hit, it's just "fatigue" and "morale", but how do you justify a character falling off a cliff and hitting the ground at terminal velocity, but not taking enough damage to die? Sure, people in real life have survived falling from great heights, but such luck is so extremely rare that it's considered to be a miracle by many when it happens.

Besides, if hp are increased for 1st level characters but scale much more slowly, people won't need to jump through hoops and make abstraction arguments to defend it, because they'll be much more believable. If a 1st level fighter had, say, 20 hp, and that same fighter had like 40 hp at 20th level, the difference would be enough to matter but not so vast that it stretches things beyond belief. 

Damage also needs to scale much more slowly. One thing I think 4e did right was that epic level (21st-29th) spells were only about twice as powerful as heroic level (1st-10th) spells in damage. Fireball, a 5th level spell, did 3d6 + Int mod. damage. Meteor Swarm, a 29th level spell, did 8d6 + Int mod. I'd like to see 5e follow suit. I don't want to see high level spellcasters throwing around empowered chain lightnings for 30d6 damage again, nor do I want to see 20th level characters with hundreds upon hundreds of hit points.
I like the discrepency, its been intentional since 1e, and I dont want it to change.
"The secret we should never let the gamemasters know is that they don't need any rules." Gygax
I'd like to see characters start with hit points equal to their Consiittution score plus a bonus from their class. After that, I'd have both hp and damage scale very slowly. The hp of chracters should scale enough that they become measurably more powerful, but not so much that they can laugh at vials of alchemist's fire being thrown on them, or survive being dunked in lava or falling off a 100 ft. cliff. In past editions, what does it matter if you fall from any height once you're high level? It's "just" 10d6 damage! Likewise, a round of immersion in lava is a mere 10d6 damage also, and high level characters could potentially survive several rounds of swimming in the stuff, just as they can survive being hacked dozens upon dozens of times with an enemy's axe, while a 1st level character could die from stubbing his toe. Ridiculous.

And before anyone comes at me with the "but HP are an abstraction!" retort, don't bother. I've heard all the excuses before, and it's still absurd. You might be able to say that hits in combat don't really mean the character was hit, it's just "fatigue" and "morale", but how do you justify a character falling off a cliff and hitting the ground at terminal velocity, but not taking enough damage to die? Sure, people in real life have survived falling from great heights, but such luck is so extremely rare that it's considered to be a miracle by many when it happens.

Besides, if hp are increased for 1st level characters but scale much more slowly, people won't need to jump through hoops and make abstraction arguments to defend it, because they'll be much more believable. If a 1st level fighter had, say, 20 hp, and that same fighter had like 40 hp at 20th level, the difference would be enough to matter but not so vast that it stretches things beyond belief. 

Damage also needs to scale much more slowly. One thing I think 4e did right was that epic level (21st-29th) spells were only about twice as powerful as heroic level (1st-10th) spells in damage. Fireball, a 5th level spell, did 3d6 + Int mod. damage. Meteor Swarm, a 29th level spell, did 8d6 + Int mod. I'd like to see 5e follow suit. I don't want to see high level spellcasters throwing around empowered chain lightnings for 30d6 damage again, nor do I want to see 20th level characters with hundreds upon hundreds of hit points.



First off, I would argue that the problem there is that falling damage is linear and lava damage is only 10d6, but that's a bit beside the point.  While I agree that high level heroes have somewhat ridiculous numbers of HP (and always have), there are a lot of difficulties with doing things the way you suggest.

1) Sure it feels wierd to survive 100' falls, but the truth is it's no wierder than surviving getting hit by a storm giant with a two handed warhammer.  I know, "a hit doesn't mean you got bludgeoned in the face, it's a glancing blow," but really?  It is totally impossible, in any circumstances, for a high level PC to get bludgeoned in the face?  Even when he's totally unaware the blow is coming (I hesitate to go as far as unconscious, but even the coup de gras rules haven't always saved us from this)?  Do you have any idea how much momentum would be imparted by such an event even if the blow glanced off at 1 degree?  HP are wierd everywhere, and fixing it for some things only makes it wierder for the rest of them.  If high level heroes aren't unstoppable juggernauts of HP that can jump off 100' cliffs with impunity, why can they fight warhammer wielding storm giants?  If you want to go all the way to "it didn't really hit, but I used up some of my mysterious luck reserves to make it miss," then why does it use up more luck reserves if I coat the hammer in poison (let's make it invisible poison, so we can't even say it's a morale issue)?  I'd sooner just accept that high level PCs are no longer human(oid), but magical demigod/legends.

2) When you throw in bounded accuracy, HP is one of the few things you get when you level.  If HP gains are small, damage gains must also be small or else combats will come down to initiative rolls.  If HP, damage, and attack bonuses aren't growing much, 10th level heroes just aren't that much better than 1st level heroes, and that's problematic from a high fantasy standpoint.  Of course, if you want low fantasy, by all means.  But to me D&D is a high fantasy game.

3) If CON doesn't add a lot to HP, it becomes a not very useful stat (I know there are a few occassions when you'll need to roll a check, but not that many compared to other attributes and honestly a +1 or 2 doesn't help that much on those rare occassions).  If 18 con is only getting me 8 more HP than 10 CON, even at level 10 when monsters are doing at least 10 points of damage (conservative even in a system where they don't grow much), it's just not worth an investment.  On the other hand, if CON does add a lot to HP but HP isn't growing much, then the disparities between high CON and low CON PCs get bloody ridiculous (to take the extreme example, suppose you get 10 HP + CON mod HP/ level: someone with 18 CON will have 5x the HP of someone with 10 CON by level 10).  The way to fix this is to gain n+CON mod HP/lvl, but then you're gaining HP pretty fast.

I fail to see the relevance of average NPC stats to PC hitpoint disparities.  The only things that are relevant to to hitpoint disparities are PC hitpoints (and by extension PC CON), and NPC damage.

The question is about system design, and there has yet to be any indication that NPCs will use rules different rules than PCs.  The system is designed around average CON of 10 or 11, with only the suggested stat-generation methods for PCs deviating from that at all.

CON 10 is average.  Sure, most PCs will be above average, but that just means that an above-average stat like a 12 is going to be more common for a PC to have; the average is still 10, and below-average stats are still entirely probable.

Back to the original point, the wizard with 10 CON isn't throwing a dump stat in there, and shouldn't expect to suffer the consequences as such.  CON 10 is average and should feel perfectly adequate.  CON 8 is very slightly below average, and should not mean that a character is instantly doomed.

The metagame is not the game.

What do you mean there hasn't been any indication NPCs will use rules different than PCs?  The rules that PCs use have been clearly indicated: roll 4d6 and drop the lowest, or use an array, either way getting an average of over 12.  Either NPCs will use different rules (as clearly indicated from the fact that their average is concededly not 12) or they will use the same rules, in which case their average is also over 12.  Unless you're suggesting that the PC stat generation rules will be changed to some heretofore unmentioned generation choice that makes the average 10, in which case I find it highly unlikely that very many fighters are going to have CONs of 18. 

CON 10 is average for NPC humans.  CON 18 or something is probably average for NPC giants.  CON 4 is probably average for NPC housecats.  Why should the first be more relevant than the others to PCs, whose average is 12?  And if we define "dump stat" as below average for PCs (which I contend is a perfectly reasonable definition of "dump stat," indeed a more reasonable one than "below average for some group other than the one to which PCs belong and against which any reasonable system designer would be balancing monster damage values"), then 10 is a dump stat.  Unless you're contending that the monster damage numbers are designed to be appropriate for targets with CONs of 10, in which case I suggest they revise them because very few of their targets will in fact have CONs of 10. 

In point of fact, there's no particular reason to assume that fighters will have, on average, more CON than wizards in the first place.  CON is in many ways MORE valuable to the wizard, as it adds proportionately more to his HP. So why assume that wizards will put one of their worst attributes into it, while fighters will put their their best?  If we assume they do not have any control over where they put their worst stats (e.g. rolling in order), then it makes even less sense to assume wizards will systematically roll worse on their second set of 4d6 than fighters. 

I happen to agree that wizards don't have enough hitpoints.  But I don't think that has anything to do with the fact that fighters have too many, or too many more.  And I don't think inflating that disparity by assuming wildly different CONs is fair or helpful.  If we make the disparity appropriate for fighters with CONs of 18 and wizards with CONs of 10, then when both have CONs of 14 the disparity is too small.  Since it is far more likely that both will have CONs of 14, or perhaps that the Wizard will have a CON of 12 to the fighter's 14, the disparity should be built around that assumption.  If you simultaneously wish to limit the disparity between wizards and fighters and CONs of 10 and CONs of 18, that's a whole other question from limiting the disparity between CON 10 wizards and CON 18 fighters, with a whole other solution.  Don't conflate the two together.
The whole language begins to break down when you consider the difference betwen a human (PC) and human (NPC) to be on the same level as the difference between a giant and a house-cat.

The metagame is not the game.