D&D Blog : Dice Tricks

[Kingreaper]

Anyway, back to the Dice Tricks themselves: I don't like the idea of having more than one or two extra dice added to a d20 roll. One extra is fine, even if it's constant (as long as it's only for certain classes/builds, so that math-averse people can avoid it) but when you start getting three or four? I don't like bucket-o-dice for hit rolls.

Damage is cool, you know it's going to do something, the bucket-o-dice tells you how big a thing. But having to sort through a bucket-o-dice to know you failed? Frustrating

*note: this is why I hate WoD systems, they manage to have bucket-o-dice for damage, and still make it common to roll your b-o-d and have no effect*

EDIT: [MENTION=957]BryonD[/MENTION]: I could continue debating with you about the quality of the writing, but the evidence suggests you want it to be bad writing. So I'm not going to bother cluttering up this page anymore.

 

[BryonD]

And I could just as fairly say that you are going to praise them no matter what they do. See no evil does more harm than holding things to standards.

I WANT it to be great writing.
The problem is that regardless of what I want, what it actually is leaves much to be desired.


This was ONE blog post. It has a serious flaw and I made a passing comment on it. I remain highly interested in seeing what they can do and I don't presume that this post is indicative. But my original complaint about this specific blog remains valid.

 

[Mark CMG]

I get the sense that they are planning to include dice tricks in 5E, which gives me the further hunch that the OGL is not going to be in use. And honestly, I don't expect even a revised GSL to be useful to more than a few publishers who want to write in very limited areas of support. I figured at some point they would introduce some element(s) that would suggest which way they would go regarding the "open" licensing. This feels like it to me. We'll see.

 

[mkill]

Funny. Before the 5E announcement I was fiddling with a little RPG project and of of the ideas was to replace skills with a generic "when X roll twice on ability check." For example, Rogue types get a "skill" to roll twice on dexterity checks for sneaking and hiding.

The big advantage of such a system is that you can keep the same DCs for skilled and unskilled. Both skilled and unskilled PC will have results in the range of 1+ability to 20+ability, just that the skilled guy has a much better chance of scoring in the higher range. It's the difference between beginner's luck and consistemcy through training.

Btw, you can get nice charts on this by entering "output 1@2d20" on anydice.com.


[MENTION=957]BryonD[/MENTION], Kingreaper: You both win an Internet. Now please move on.

 

[pauljathome]

I don't think that they couldn't write a description, but rather that they chose not to.
.

I really don't care about how they got their 4.5 number.

The main point (and this IS significant) is that 2D20 drop the lowest is NOT particularly similar to add n (regardless of the n).

1) The probability distributions are significantly different
2) The chances for criticals are significantly different
3) The chance of getting a result of 1-5 or 21-25 are REALLY, REALLY different (ie, they are 0 in one case but not the other).

Their analogy is fundamentally flawed.

The preamble to the poll is just wrong.

And THAT shows some combination of innumeracy (my personal opinion), really really bad writing, or dishonesty on their part.

 

[pauljathome]

Funny. Before the 5E announcement I was fiddling with a little RPG project and of of the ideas was to replace skills with a generic "when X roll twice on ability check." For example, Rogue types get a "skill" to roll twice on dexterity checks for sneaking and hiding.

The big advantage of such a system is that you can keep the same DCs for skilled and unskilled. Both skilled and unskilled PC will have results in the range of 1+ability to 20+ability, just that the skilled guy has a much better chance of scoring in the higher range. It's the difference between beginner's luck and consistemcy through training.

I like this a fair bit.

The problem is that people also get better with more skill. An expert can not only do simple tasks far more reliably but he can also do harder tasks.

An ideal solution would probably be something along the lines of combining multiple dice AND bonuses. So, a novice would be D20, a journeyman 2D20+5, an expert 3D20+10 etc.

But that is getting too complicated and would be very hard to get right. See 4th edition skill challenges for how difficult it is to get complicated math "right" (especially since "right" is so subjective).

A much easier answer that is probably "good enough" for D&D is to just use bonuses and a modification of the Take 10 rule. The better you are, the more circumstances that you can take 10.

 

[Kingreaper]

I like this a fair bit.

The problem is that people also get better with more skill. An expert can not only do simple tasks far more reliably but he can also do harder tasks.

My favourite system for this (that wouldn't work for D&D at all) is a system that had every skill be a die-type, from d4->d12.

You roll that die, and all smaller die, and take the highest.

So if you're on d12, you've only got a 1/6 chance of getting 11 or 12, but you're nigh-certain to get at least a 4

 

[kitsune9]

I like some dice tricks myself, but I think it should be with action points primarily.

 

[I'm A Banana]

One big meh.

I'm not really against 'em, but they don't attract me in and of themselves any more than d20+mods.

In fact, the slightly turn me off, just because d20+mods is pretty easy, straightforward, and easy to customize. Dice tricks might be a little nifty, but they're not exactly easy to figure the odds for.

So, they're not a great replacement for core rules. They might exist as an option, or in specific spells or effects, but the game shouldn't expect or demand their use from the get-go. They're superfluous.

 

[triqui]

I think rolling 2d20s instead of one is a great way of simulating a luck power of some sort. You can't rely on it, but sometimes it makes thing work better.

It is not a good way of replacing a static bonus. If I'm trying to roll a 24 with my measly +2 bonus, I would much rather have a +5 to the roll instead of two chances. Maybe in the cases of "20 always hits" it works out somewhat, but I suspect in most cases it doesn't.

Part of the point is to make the system work without needing to escalate to DC that can't be touched at all. A DC 24 is completelly untouchable for anybody without a +4 or better bonus, and this come to a point where, lvl 10 and beyond, only specialist can even try.

There are allways 3 kind of people in a given roll. The one who masters it, the one who is just decent, and the one who is horrible at. With static bonuses, any DC that is just half-threatening to the master, is completelly out of the possibilities to the bad guy.

For example, let's take Perception, using something like Pathfinder as an example. A level 10 guy with 10 "ranks" in the skill, Class bonus, wisdom 20 and maybe a +2 racial/focus/feat bonus will have +21 to perception. A guy with 7 ranks and wisdom 14 will have +9. A guy with 0 ranks and wisdom 10 will have 0.


Against a DC 21 task, the guy without ranks can't even try. Ever. The specialist guy will never fail, becouse with a 1, he will get 22.

Against DC 25 or so, even the decent guy have very small chances to suceed (16+), while the specialist is almost a guaranteed win (4+)

Against DC that become a challenge to the high specialist, for example DC 30 (you need a 9+), the guy with a decent level in the skill can't even try.

So it get's to a point that, when the DM puts a challenge that is a threat to the guys specialist in that roll, everyone else is completelly useless. For example, perception. When the party get ambushed, either the guy without perception as a class skill will ALLWAYS be ambushed, or the guy with keen senses will NEVER be ambushed (or both at the same time)

Compare it to a system where the bonus are much shallow, and the speciallist gets extra rolls.

A system might have DC 15, but the specialist can roll 3d20 (skill training+racial focus) and take the best and also add +4 to the roll from the ability. That means he has 87,5% of success. The Decent guy might have 2d20 (skill training) but just +1 from the ability (so he has roughly 60%). But the no-training guy can roll, and have a 30% chance.

This smooth out the chances for everybody, allowing everybody to play the game in any moment. A fighter with Cha 10 and no ranks in diplomacy can *try* to talk a NPC, and still have a chance to suceed. With flat bonus, any situation with a DC that allow the fighter to roll, it's an auto-success for the group bard. Any situation with a DC high enough to make the Bard have a chance to fail, it means the fighter HAS to be silent, becouse it is IMPOSSIBLE for him to succeed that roll