Some old-school gamers were complaining that the hit points in 5e were inflated, and I afree.  The kobolds, for instance, pose little threat to the fighter and cleric with their low attack bonus and piddling damage.  This does not seem like a low-level adventure.  On the other hand, it sucks to create a character and have him dead in the first five minutes due to a lucky crit.  Perhaps a compromise is necessary.  I would say that characters get two attacks to start out with to balance against the inflated HP totals.  Thus, a single attack is is unlikely to kill a PC outright, but they are still in considerable danger in combat.

The "Math" in 5e is pretty much broken, and it's liable to stay that way. They say that they aren't much concerned with math formulas right now because "math is the easy part", but since they categorically reject theorycraft and mathhammering, there's no evidence that they'll ever be able to do that "easy part". And let's be honest, a lot of design goals cannot be reconciled and to figure that out it would be really helpful to check the math before putting them into the system.

As an example, let's consider their idea about automatic success thresholds of DC == Attribute. If you stopped to consider that from a mathematical standpoint for even a moment, you'd notice that that increases the threshold of auto-success by 2 for every increase of the attribute by 2. Now you have two choices for rolled success:

• +1 per 2 points. The Automatic Success threshold rapidly covers the entire die roll, and high level characters cannot be given a challenge that they can fail that they won't almost always fail.
  
• +2 per 2 points. The difference between an 8 stat and an 18 stat is +10 to the die roll, and the random number generator is broken at first level.

The problems are obvious. Even a casual glance at the theorycraft behind it shows that no matter what you set the rolled modifiers to, the end result can't be good. And yet, they wrote several articles about how cool it was and debuted it in their closed playtest where... it flopped horribly. As indeed, it couldn't possibly do anything else.

And now what they've done is scale back the automatic success thresholds so that they only kick in on DCs that are considerably lower than your stat. A Rogue needs to have a Dex of more than 26 for it to even affect his Lockpicking checks. According to the current design documents, such a Dex score isn't even possible for non-Divine Beings, so it's effectively been removed from the rules completely. Not in a way that would entail admitting that mistakes were made or canceling the mechanic outright - but in every conceivable other way it's gone.

And basically all of the ideas they've floated so far are like that. Consider "bounded accuracy": it is a sham. First of all, they aren't really doing it:

Quote from: mearlsThe math will be baked into the class and race. Since those are the only things that are 100% required for the game, between the two of them they contain all the math that we assume.

But they still have equipment upgrades, which are of course neither Class nor Race. The difference between Chainmail and Shield (AC 17) and +5 Adamantine and +5 Shield (AC 30) is 13 points. The Minotaur who hits your first level ass half the time can only hit the high level warrior (or at least, character using a high level warrior's equipment) on a natural 20 because a natural 20 always hits - if it weren't for the auto-hit rule they'd still be 4 points short. And for Dex Fighters, the difference is even more extreme because in addition to picking up 13 AC for using high level equipment they also pick up whatever the difference is between a high level character's Dex modifier and a starting character's Dex modifier.

But let's say they decide to cut equipment way back and actually deliver on the bounded accuracy such that:

Quote from: mearlsthe monster design is aiming to keep hordes of orcs/goblins/etc a viable threat at high levels. So, at level 1 it might be 18 rats, but at level 10 it might be 18 orcs.

So you're fighting a bunch of Orcs at 10th level and they still have the same to-hit on you that they did at 1st level. And your hit points are... well your hit points are a lot, because even if you chew them up thrice as fast at 10th level as you did at 1st level, you're still looking at about 63 times as many Orc attacks from that group as you were exposed to at 1st level. If you actually line fit those supposed design goals into a coherent picture, the 10th level Fighter is being asked to come into battle with over five hundred hit points. That is what "bounded accuracy" would actually require.

Now, obviously they are going to shrink back from that. Because 10th level Fighters having half a thousand hit points to keep track of in order to be able to grind Orcs is painfully silly even for a computer game. But without that, what is there? They clearly haven't thought of what handling all power growth through hit point inflation means, and that's terrifying. There obviously isn't going to be a simple math fix, because right now they have people making design guidelines who cannot intuitively feel that doubling a number six times in a row is going to make that number large and unwieldy.

-Frank

I was actually just thinking about character lethality at low levels.  Kobolds rolling 1d8-2 is a joke when characters have a hearty +Con to HP.  Even orcs rolling 1d8+1 is fairly weak, especially when they have a meager +3 on attack rolls.  Has anyone hammered out the exact math for 18 orcs yet?

Frank,
Please tell me you've been submitting comments about their math. Hopefully, at least one of them is reading it.

I'm still of the opinion that all you really need to do to fix hit points in D&D is to give everyone a Hit Dice for zero level.

As for extra attacks I think allowing attack bonus to be divided between targets works as long as you're not looking at +1 / level bonuses.  I also like allowing AC to be increased by dropping points from your base attack bonus.

I just don't like it when you have to have a feat to do them.

Quote from: danbuter;549345Frank,
Please tell me you've been submitting comments about their math. Hopefully, at least one of them is reading it.

Unfortunately, they've already said that they intend to filter out any submitted comments about "the math" on the grounds that they intend to fix it later. Unfortunately, as previously noted most of their ideas are in the "you can't fix the math later because you have mutually exclusive design goals" area. And by telling the more mathematically inclined testers "Nyah nyah nyah! I can't hear you!" they have certainly not encouraged me to try to break through to them.

Quote from: BTI was actually just thinking about character lethality at low levels. Kobolds rolling 1d8-2 is a joke when characters have a hearty +Con to HP. Even orcs rolling 1d8+1 is fairly weak, especially when they have a meager +3 on attack rolls. Has anyone hammered out the exact math for 18 orcs yet?

The mathhammer on the current ruleset is not particularly encouraging. PCs get their full Con score in hit points at level 1, but their Con bonus only comes in on level-up if it's more than their rolled hit points for the level (a Fighter with Con 14 gets 2-10 hit points a level). Aside from the question of "what happens if my Con score changes when I'm 8th level, do I seriously have to remember what my literal rolls were for each level?", this brings in a system where players get a big grip of hit points at first level and a comparatively much smaller amount of hit points at later levels - which is basically the opposite of what Bounded Accuracy actually implies. Meanwhile, the monsters actually follow the Bounded Accuracy script pretty well - the Gnoll has 11 hit points, the Ogre has 66 hit points, and the Troll has 132 hit points and regains 10 hit points a turn whether he has been hit by fire or not.

As printed, it takes an average of 8 Orc Attacks (or 5 Orc charges) to drop the sample first level Fighter.  The Fighter then gains the ability to soak one extra hit from an Orc (or 2/3 of a hit from an Orc Charge) per level. The first level Fighter drops an Orc 64% of the time when he swings that greataxe. At level 2 he picks up the ability to make two extra attacks per day, and at level 3 he gets an extra attack once per turn when he drops an Orc and his chance of dropping an Orc on an attack goes up to 68%.

The first level Fighter is an even match for about 3 or 4 Orcs (depending upon assumptions about charges). The 3rd level Fighter is an even match for like 5 or 6. Having the Fighter be able to take on 18 Orcs at level 10 doesn't seem super likely, because if Bounded Accuracy actually happened the Orcs will literally drop the Fighter to zero hit points simply by each of them getting a single charge attack on the Fighter. Even if our Fighter character can drop 9 of them a round or something silly, just the act of every one of them moving up and attacking him, he'll drop by the time every Orc has acted once.

Now, we know Bounded Accuracy won't happen, because they already gave the show away with armor upgrades. It seems highly likely that in reality the Orcs will hit on natural 20s only and the Fighter will grind through them at whatever rate and it will simply be a whole lot of die rolls before he inevitably kills them all.

I think we're starting to see Mearls realize how silly a game he has committed himself to by him admitting that 18 Orcs should be making group attack rolls. That in turn implies that they won't be able to simply grind the fighter to dust in one round by having them all attack, which in turn implies either a much faster (and accelerating) hit point increase (as fast or faster than the listed monsters), or dropping the Bounded Accuracy pretense.

-Frank

Quote from: danbuter;549345Frank,
Please tell me you've been submitting comments about their math. Hopefully, at least one of them is reading it.

If he doesn't, I will.  I'll attribute it properly, mind you, but I will get it to them.

Quote from: danbuter;549345Frank,
Please tell me you've been submitting comments about their math. Hopefully, at least one of them is reading it.

That, at this stage, the designers still seem oblivious to so many things that should be basic and fundamental does not bode well for the design or give me optimism that they are listening or, if they are, that they really understand what people are telling them.  It can be difficult to understand styles of play different from one's own and to appreciate their concerns, so the problem may simply be that several important styles of play are simply not represented on the design team.  Or maybe the problem is simply so far up into the idea that D&D is an abstract game and don't grasp the idea that the results should reflect things happening in the game world that make sense to the characters that they don't grasp or care about the problems caused by mechanics that don't make much sense beyond producing some sort of abstract game outcome.

Quote from: FrankTrollman;549348Unfortunately, they've already said that they intend to filter out any submitted comments about "the math" on the grounds that they intend to fix it later.

WTF

The maths is the core of the game engine, if that doesn't work, you can't change it later because then you need to test everything again...

Quote from: Ladybird;549361The maths is the core of the game engine, if that doesn't work, you can't change it later because then you need to test everything again...

Many years ago, friend and I attended a science fiction convention that had a presentation for a new game called "Shadowrun".  At first, a lot of what we heard sounded interesting and we were excited about it, and then Tom Dowd started talking about the system and moving dice around between pools and said, "This is what we call 'buckets of dice' game," at which point our enthusiasm evaporated.  We considered system a tool to enable role-playing that should be as unobtrusive as possible, not the point of play that we wanted forced into the foreground.  From then on, "This is what we call 'buckets of dice' game," has been shorthand among the people from my group that were there for something that sounds great but gets ruined by a revelation or something in the execution that's a big turn-off.  

What Tom Dowd made clear during that presentation was that part of the point of the system was to have players physically playing with their dice like toys.  Not role-playing their characters but playing with the system like a toy, in much the way that finger painting is less about producing good art and more about the tactile experience of playing with paint with your fingers.  And what's become clear over time (and it's especially true of many of the Forge games) is that somewhere along the line, perhaps as a backlash to 1980s games embraced complexity to improve realism, playing a character has become an incidental, if not ignored, concern and the point for many people has become manipulating numbers, descriptors, and rules in an almost abstract way that need bear little resemblance to what the characters are doing or what's going on in the setting.

If WotC wants me to buy their new edition of D&D, they need to grasp that I'm not interested in finger painting.

Quote from: FrankTrollman;549348Unfortunately, they've already said that they intend to filter out any submitted comments about "the math" on the grounds that they intend to fix it later. Unfortunately, as previously noted most of their ideas are in the "you can't fix the math later because you have mutually exclusive design goals" area. And by telling the more mathematically inclined testers "Nyah nyah nyah! I can't hear you!" they have certainly not encouraged me to try to break through to them.

At times I wonder if they even have anybody on staff which bothers checking simple stuff, like writing a simple computer program to see whether it is doing what they intended.  (Such as combat slugfests).

QuoteUnfortunately, they've already said that they intend to filter out any submitted comments about "the math" on the grounds that they intend to fix it later. Unfortunately, as previously noted most of their ideas are in the "you can't fix the math later because you have mutually exclusive design goals" area. And by telling the more mathematically inclined testers "Nyah nyah nyah! I can't hear you!" they have certainly not encouraged me to try to break through to them.

I understand maybe half of what you're saying but I do know you know your math and something like this does need to be addressed.

Quote from: John Morrow;549365Many years ago, friend and I attended a science fiction convention that had a presentation for a new game called "Shadowrun".  At first, a lot of what we heard sounded interesting and we were excited about it, and then Tom Dowd started talking about the system and moving dice around between pools and said, "This is what we called 'buckets of dice' game," at which point our enthusiasm evaporated.  We considered system a tool to enable role-playing that should be as unobtrusive as possible, not the point of play that we wanted forced into the foreground.  From then on, "This is what we called 'buckets of dice' game," has been shorthand among the people from my group that were there for something that sounds great but gets ruined by a revelation or something in the execution that's a big turn-off.  

What Tom Dowd made clear during that presentation was that part of the point of the system was to have players physically playing with their dice like toys.  Not role-playing their characters but playing with the system like a toy, in much the way that finger painting is less about producing good art and more about the tactile experience of playing with paint with your fingers.  And what's become clear over time (and it's especially true of many of the Forge games) is that somewhere along the line, perhaps as a backlash to 1980s games embraced complexity to improve realism, playing a character has become an incidental, if not ignored, concern and the point for many people has become manipulating numbers, descriptors, and rules in an almost abstract way that need bear little resemblance to what the characters are doing or what's going on in the setting.

If WotC wants me to buy their new edition of D&D, they need to grasp that I'm not interested in finger painting.

That's very well said, and I agree with the sentiment.

I'm not interested in finger painting either.

Quote from: John Morrow;549365Many years ago, friend and I attended a science fiction convention that had a presentation for a new game called "Shadowrun".  At first, a lot of what we heard sounded interesting and we were excited about it, and then Tom Dowd started talking about the system and moving dice around between pools and said, "This is what we called 'buckets of dice' game," at which point our enthusiasm evaporated.  We considered system a tool to enable role-playing that should be as unobtrusive as possible, not the point of play that we wanted forced into the foreground.  From then on, "This is what we called 'buckets of dice' game," has been shorthand among the people from my group that were there for something that sounds great but gets ruined by a revelation or something in the execution that's a big turn-off.  

What Tom Dowd made clear during that presentation was that part of the point of the system was to have players physically playing with their dice like toys.  Not role-playing their characters but playing with the system like a toy, in much the way that finger painting is less about producing good art and more about the tactile experience of playing with paint with your fingers.  And what's become clear over time (and it's especially true of many of the Forge games) is that somewhere along the line, perhaps as a backlash to 1980s games embraced complexity to improve realism, playing a character has become an incidental, if not ignored, concern and the point for many people has become manipulating numbers, descriptors, and rules in an almost abstract way that need bear little resemblance to what the characters are doing or what's going on in the setting.

If WotC wants me to buy their new edition of D&D, they need to grasp that I'm not interested in finger painting.

Nail, meet hammer.  QFTMFT!

Quote from: John Morrow;549365Many years ago, friend and I attended a science fiction convention that had a presentation for a new game called "Shadowrun". At first, a lot of what we heard sounded interesting and we were excited about it, and then Tom Dowd started talking about the system and moving dice around between pools and said, "This is what we called 'buckets of dice' game," at which point our enthusiasm evaporated. We considered system a tool to enable role-playing that should be as unobtrusive as possible, not the point of play that we wanted forced into the foreground. From then on, "This is what we called 'buckets of dice' game," has been shorthand among the people from my group that were there for something that sounds great but gets ruined by a revelation or something in the execution that's a big turn-off.
 
What Tom Dowd made clear during that presentation was that part of the point of the system was to have players physically playing with their dice like toys. Not role-playing their characters but playing with the system like a toy, in much the way that finger painting is less about producing good art and more about the tactile experience of playing with paint with your fingers. And what's become clear over time (and it's especially true of many of the Forge games) is that somewhere along the line, perhaps as a backlash to 1980s games embraced complexity to improve realism, playing a character has become an incidental, if not ignored, concern and the point for many people has become manipulating numbers, descriptors, and rules in an almost abstract way that need bear little resemblance to what the characters are doing or what's going on in the setting.
 
If WotC wants me to buy their new edition of D&D, they need to grasp that I'm not interested in finger painting.

But I like finger painting! I do agree with what you're actually saying though.