HP

[PanamaLane]

So, yeah, I'm sure this has been a serious topic of debate, or at least has had to have been in the DM staff. I'd just like to point out among every PC I've talked to though, seems like everyone would favor a max HP situation as opposed to the random one in place.

I think it makes sense partly due to the 75% (or some such) rule regarding min HP, which is the system by which your min hp is decided as 75% of your starting HD.

The way the system works, if you have a low con, low hit dice character you end up -way- more likely to roll max HP and for your low HP rolls to not seriously effect your character. If you have a high con high hit dice character you end up much more likely to roll well below your max HP. and for your min rolls to be, well, devastating.

For example:
A 10 con cleric has a small range: 6-8hp (average 7hp)
A 14 con fighter has a much larger range:7-12hp (average 9hp)
You'll notice the average difference is only 2hp, which is just what the fighter should be gaining from his Con bonus. Furthermore your cleric has a 33% chance of rolling max hp compared to your fighter who has a 20% chance.

In regards to barbs v. fighters...the randomness plays into the fighter really, as a low con fighter has a much higher percentage chance of rolling max hp then a high con barb.

a 20 con Barb: 9-17hp (average 13hp, percentage chance max hp 12.5%)
a 10 con fighter 7-10hp (average 8hp, percentage chance max hp 25%)
So what you have in this instance is a fully toughened barb that is averaging only 4hp more (less then just his con bonus, not to mention his +2hd) then your weakling fighter and has half the chance at rolling max.

Seems a bit busted to me.

Additionally, I don't think people on the server really have a problem with your high HP people being your Barbs with high Con. I mean, that's the best part of having a high Con, high HD character in combat. Your low con, low HD characters should have, well, much less HP. The two don't need to be brought closer together, there should be serious differences.

Anyway, not complaining or anything, its just a number crunch I thought to point out. For what its worth, I think the vast majority of the server would be happier with a max hp system. Though, I'll admit it might make spawns tougher to asses, I think for the most part it should remain pretty simple. Hey, you could always just bump a spawns con up a point or two to even it out.

[ExileStrife]

Before I even begin a response to this, can you provide a source to this "75% rule"?  I am genuinely curious as to whether this is fact or just wild speculation.  Regardless though, I know that what I plan to say won't be affected by the answer.

[9lives]

I thought minimum HP was half your class's HP per level (d4/d6/d8/d10/d12), plus corresponding HP bonuses from CON or toughness.

[Pup]

I thought the same as 9lives.

[ExileStrife]

half per level?  what the blazes does that mean?

[PanamaLane]

I could be wrong about it. I remember Musachi of all people telling me that was the case and it seemed to fit the dice rolls. I'll look into it for you though. 9's thought seems to fit with numbers I've gotten as well.

To note though, even if it is half plus your con bonus, it still leaves the situation of larger ranges for higher HP characters and thus lower percentage chance at max. In fact, it might even make it worse. Though getting the Con bonus straight up certainly helps? I don't know?

Let me find the answer and crunch the numbers.

[AKMatt]

It is not 75%, because I have definitely rolled lower than that with two characters.  nwn.wikia.com shows the hp range per level as being 2-4 for d4 classes, 3-6 for d6 classes, 4-8 for d8 classes, and so on (half/level minimum).

[chatellerault]

I thought minimum HP was half your class's HP per level (d4/d6/d8/d10/d12), plus corresponding HP bonuses from CON or toughness.

This is my understanding as well.

[Egon the Monkey]

Riiight. I have no idea what this is going on about.

http://nwn.wikia.com/wiki/Hit_dice says minimum per level is 1+CON+toughness. Could someone please explain this all slowly for the clueless people like me who hate the HD system but probably dont' understand it?

I'd be all for any adjustment that reduces variation between identical builds, because with the lower level of the server, good/crap rolls are not going to average out as much. It's the only aspect of char creation that's random, it's not like you gamble on skillpoints or what BAB you get each level.

[PanamaLane]

"At higher character levels, the number of base hit points gained is equal to a roll of the hit die, but with the provision that rolls under half the maximum are re-rolled. Thus, if a level 3 PC gains a level of fighter, the PC will gain between 5 and 10 base hit points (inclusive)." -Con is then added on top.

So to make it clear, its half your hit dice plus your con bonus.

I think the problem still exists though, although slightly better then the scenario above.

In this case your 14 con fighter now rolls 7-12hp (20% chance max)
Your 14con wizard rolls 4-6hp (33% chance max)

Certainly its better!

Still though, a fighter rolls a 7, 20% of the time. A wizard rolls a 6, 33% of the time, so there is still a slight "problem" so to speak when your 14con fighters are somewhat likely to roll only 1hp more then your 14 con wizard. (granted the fighter also rolls a 12, 20% of the time, but yeah, you get it by now!)

Also, the wizard is averaging 5hp per level where your fighter is averaging 9hp, a difference of 4hp per lvl, which is well below the design i should think.

[ExileStrife]

I don't know what the point of arguing this "20% chance to get max compared to a 33% chance to get max" is.  The useful statistic here is the mean of HD distributed among the classes because it can easily be used to predict the HP of a class at any given level.

For the sake of argument here, I'm going to use the extremes, barb HD and wizard HD.  Using this "minimum = half max HD" rule, we get ranges for barb and wizard to be 6-12 and 2-4, respectively.  Their average rolls (the value that's relevant in our current server with the hidden HP rolls) are 9 and 3, respectively once again.  The difference is 6, so we can expect barbs to have 6*level more HP than a wizard of equal level and that's how everything has been balanced with the careful collection of the DM team's experience in these matters.

With max rolls, this "difference" simply increases.  The expected hp roll for a wizard becomes 4, and for a barbarian it becomes 12, the difference now being 8.  The difference in the hidden-expected-vaules and the max-expected-vaules isn't much, but it's going to mean that equal level barbs will have an additional 2 more HP per level than wizards.  This is equivalent to suddenly giving every barbarian 2 free stacking Toughness feats.  I don't think our game is currently balanced for a change like this, and I'm not sure it's wanted as much as you make it out to be (though this thread could prove me otherwise).

I don't think it's worthwhile to consider the average consitution for these classes because regardless of the hidden or max HP roll situation because it affects the important difference in the same way.

[Thomas_Not_very_wise]

Mechanics....ew

Can we all pick Daisies?

[PanamaLane]

The difference is 6, so we can expect barbs to have 6*level more HP than a wizard of equal level and that's how everything has been balanced with the careful collection of the DM team's experience in these matters.

You've got a point certainly. But the percentage is where the argument is at because you can't expect the difference to be 6/lvl, because in this scenario your wizard rolls max hp 33% of the time, your barb only 12.5% of the time. Thus as the two progress in levels, its far more likely the wizard will get good rolls then the Barb.

Not to hammer on the percentages game to much, but its important to the picture I think. I was thinking of it this way.

A 14 Con Wizard has a 66% chance of rolling a 5 or a 6.
A 14 Con Fighter has a 60% chance of rolling a 7, 8, or 9.

At the greatest difference your talking about 4hp. At your closest you're talking about 1hp. So, all said and done, you have about a 60% chance of a 1-4hp difference. Seems fishy, right?. It totally is because I'm using fuzzy math. Still, it shows how important the percentages are to the argument.

And one more scenario:

a 10con rogue has a range of 3-6
a 10con barb has a range of 6-12

This would seem to be closer to how the game was designed (obviously). However, in this case the rogue is literally twice as likely to roll max as the Barb, right? Not to mention, a barb rolling a min 6 is just plain devastating. I know he has a 10con, but a 6hp Barb, yuck!

Finally, outside of the number game entirely, I think there is a real desire for this from the players anyway. I know whatever character I'm playing, I'd prefer to have the max available. I'd prefer for everyone else to have their max too. It just seems too darn important to leave up to a dice roll, -especially- for players who are designing their characters specifically for high HP. I mean honestly, can you imagine a 14con Barb getting the same hp at lvl up as a 10con cleric? I can, because it happens nearly 20% of the time, over 5 lvl ups that almost assures it happens at least once, and it just plain shouldn't go down like that.

For comparison sake:
A 16con lvl 8 Barb has at max 120hp. Not unreasonable I think.
A 16con lvl 8 Barb at current averages out at 96hp. Also not unreasonable.

The difference between the two, however, is often life and death. So why not give the tougher characters the benefit of the doubt? Especially since the way the system works currently, you could end up with lvl8 16con barbs with only 80hp and potentially less. I mean, it never happens, but that's because those characters die. That's the hard truth. Can you dig it?

[ExileStrife]

This "max HP roll percent" thing just doesn't work, Panama!

I will give you a game with two choices, with the goal to get the highest score over a large number of runs.  You have to pick and stick with one of two different distributions of values, each value having an equal percent chance to be chosen among the other values in that given distribution (namely, %-to-be-chosen = 1/#-of-values).  Pretend these are your choices:

A)  1 3 5
B)  1 2 3 4 5

With your argument, you would choose A) to play this game, since 5 has a 33% chance of being picked, compared to the 5 in B) which only has a 20% chance of being picked.  But that's just arbitrary and doesn't mean anything.  The only value you can use to make a proper choice in this game is the distribution's expected value, which happens here to be the mean (3 in both A) and B)).  The distributions are identical for our purposes and will yield the the same result in the long run.

I'm not trying to be mean, but I just can't follow anything else you are saying because of this fault.

[PanamaLane]

Word, we are working it out which is cool.

Using your game scenario, try it this way:

A:1 2 3
B:1 2 3 4 5

Now your goal is not to get the highest value, but instead to get the closest value to the total value possible over ten picks. Group A gives you an average of 20/30. B gives you 30/50. Group A gets a passing grade (barely), B failhards.

Where this makes the biggest difference game-wise is in your low HD high Con characters compared to your low Con high HD characters. Just to use the most severe example...A 20con wizard rolls 7-9. A 10con Barb rolls 6-12. In this scenario, over ten levels your wizard earns a score of 80/90. Your Barb earns a score of 90/120. One will get you through college, the other one won't. And no, there was no bias because the wizard was a white guy.

That or I'm completely wrong, which in a discussion over math, with the likes of Exile, I'm fully willing to admit.