I know, I know, d20s are super swingy. But if I can predict the chances of at least one d20 hitting the target number, does it really matter if I have the players roll say 2-5d20s and keep 1? Other than the sheer number of rolls of course.
Isn't that the point of tools like AnyDice anyway? To help predict the results of such interactions?
>>45494061
Learn how to construct a sensible sentence. What the fuck is that garbled rant supposed to be asking?
>>45494089
>>45494089
You sound like that one kid who couldn't read that always gets called up to read a passage to the class...in college.
Why don't you take the time you spent needlessly whining and answer the damn question instead?
>>45494061
For what purpose?
Why do you want to do a roll-and-keep system with d20s as opposed to something like d10s or d6s that are more commonly used for dice pool type setups?
>>45494146
Not him, but I have exceptional reading comprehension and failed to glean your meaning.
>>45494061
I suspect you are asking if it matters if the players roll several d20 and keep only one in order to achieve the target percentage chance you are after.
If we ignore the number of dice.
The answer is yes.
Elegance of the system is very important.
Rolling one die and using the result is always better than rolling several and keeping one, for that reason.
Can you do it? Sure.
However, "you can drive your car with your feet, that doesn't mean it's to be done."
>>45494146
Child, I am getting a Master's in teaching English, and I have no idea what you are asking. I regularly parcel both the slang of inner city teenagers and the verse of the Bard, and I cannot understand your question. You are a drunken word-shepherd letting his thoughts run wildly over the field, trampling the grass and ruin any coherence. Marshal your flock, motherfucker.
Are you asking if rolling more dice affects the outcome? Are you looking for a bell curve? Are you actually asking if being allowed to roll five dice and keep one will yield different results than rolling one die? You are incredibly vague. Be specific, and ask what you mean.
God I wish I could fail your bitch ass right now.
>>45494296
>*ruining*
Fucking autocorrect. Right in the middle of my rant, too.
>>45494309
It also got you on "parse".
>>45494323
Damn it all. That's ten points off. I'm getting a B on my own scale.
>>45494352
>>45494323
Learn to BS.
Perhaps anon also ships books on Ebonics and Shakespeare?
Yeah, that's the ticket.
I believe what you want to know is the odds of getting a number on multiple d20s?
Either way, do what you like. I think 5th edition d&d has a mechanic like this where sometimes you get to roll 2 and keep the higher number.
Theoretically you could house rule this in. Savage worlds also has an extra die, so you always roll 2 and keep the better result. If you're trying to increase success rates though you could just lower the target number.
(I *think* that's what he's asking about?)
>>45494061
>But if I can predict the chances of at least one d20 hitting the target number, does it really matter if I have the players roll say 2-5d20s and keep 1? Other than the sheer number of rolls of course.
So a dicepool system, with d20s?
That sounds awful.
>>45494473
Not dice pool exactly, it doesn't sound like he wants to add up successes just take the highest number. It's the uh.. Savage worlds die. What's it called.. Ah... Like, Ummmm.. Wild? Something?
>>45494215
If you use d6s or d10s, the granularity per die is pretty small. I feel like if you aren't adding stat and skill dice together to form a pool, the size of the die itself doesn't really matter.
If you have a different resolution mechanic and a way to keep track of the average chances of success, does it matter what size of die you use?
>>45494238
You're on a site that communicates mainly through grammatically incorrect "greentext" and "meme arrows". This is done on purpose and people understand what is trying to be conveyed just fine. Hell, most the of OPs on the catalog right now are horrors of modern language, but that's not triggering people's autism.
tl;dr If you are a veteran of 4chan and can read any of the slop that passes for language in this place and not the OP for some reason, I wouldn't say your reading comprehension is "excellent".
Getting back on topic: There are a number of systems that use dice pools to resolve challenges. Reducing things to their simplest doesn't work for every system. Likewise, simple can be elegant but that isn't always true. Elegance can also come from intuitiveness in a system's design.
Your analogy isn't a particularly good one, either. I get that you like single roll systems, but other systems aren't like "driving with your feet" just because you don't like them. Besides, you hit the gas, brakes, and clutch almost more than you steer. You drive with your feet at least as much as your hands.
>>45494061
>>45494494
Wild die
>>45494296
Not that anon, but being vague and being unintelligible are two different things, Master of English.
>>45494515
I don't think op classic blundered the thread is still mostly on-topic.
>>45494500
Bad spelling and grammar are fine if people understand you
>>45494500
Nigga, what the actual fuck are you trying to do with this multi-d20 system? What is your goal? What features of the number distribution it provides do you like, how do you want it to affect your game?
>>45494500
>a way to keep track of the average chances of success
Presumably you'd be using an app for this to make it quick and easy?
It seems overcomplicated, also, most people don't have many d20s.
>>45494500
>You drive with your feet at least as much as your hands.
I was using a mildly relevant quote that it can be assumed he meant *steer* with his feet.
>Likewise, simple can be elegant but that isn't always true. Elegance can also come from intuitiveness in a system's design.
That's true.
However, nothing in OP's post is intuitive.
OP was obviously unclear in his post, as most agree in this thread.
Part of the lack of clarity comes from anon's use of language and part of it comes from anon not actually saying what they want or want to know.
>>45494553
You're full of shit. The average fa/tg/uy post is something I usually have to read twice to understand.
>>45494636
I think that might be a personal problem, anon.
>>45494547
Reversed Classic Blunder
Something like the wording of anon's post being discussed more than their topic.
>>45494553
>Bad spelling and grammar are fine if people understand you
No, not really.
Saying "feet mittens" is not just as good as knowing the word for socks.
Similarly, spelling it "feat mitons" is not as clear and good as saying "socks."
And yet, is still more clear than griblv afins.
Clarity of communication is not binary.
>>45494636
>The average fa/tg/uy post is something I usually have to read twice to understand.
see >>45494644
>>45494534
True, but being vague can easily cause you to be unintelligible. I see it all the time, especially in the freshman class in which I student teach. I have never experienced a pain quite like reading through a paper in which not a single pronoun has an antecedent.
Similarly, OP just throws out the idea of multiple dice without actually defining what it is he is looking for. Is he talking about dice pools or super-advantage? Does he want to know the degree to which it will skew results, or does he actually not realize how heavily weighted such a system would be? I don't understand what he thinks he's asking, and even the people who aren't mocking him are a bit confused.
>>45494146
You sound that one faggot OP who couldn't write and always got called out for getting BTFO... on 4chan
Why don't you take the time to navigate back to >>>/reddit/
>>45494722
Uh, yea, saying feet mittens is confusing but spelling it soks in the right context isn't as bad. Still bad but my point was as long as you can be understood that shit isn't as bad as whatever op just did
>>45494818
Foot mittens is a novel way of describing something in your own way. Saying, "soks," is just stupid.
>>45494818
>my point was as long as you can be understood that shit isn't as bad as whatever op just did
Ah, that was not clear in your post at my first reading of it. heh
>>45494878
>Foot mittens is a novel way of describing something in your own way. Saying, "soks," is just stupid.
Technically, both are novel and stupid.
>>45494624
Most of those people are either samefagging or just appreciate a good circlejerk of inane values that don't really count for shit here. There's only been 7 unique posters here for a good while now. Take that how you will.
As for your analogy, I understand the implication but was capitalizing on the fact that you weren't clear to make a point.
>Subject: Rolling multiple d20s
Presumably this thread topic will be about rolling multiple d20s.
>I know, I know, d20s are super swingy.
There's no context for this sentence. However, the results of d20 rolls have always been described on /tg/ as having too much "swing" due to the equal chance of hitting any number. I assume that is the proper context.
>But if I can predict the chances of at least one d20 hitting the target number, does it really matter if I have the players roll say 2-5d20s and keep 1?
Needs to be pared down, but the intent is pretty clear: "Does it matter if you roll X number of d20s as long as one of them hits the TN?"
>Isn't that the point of tools like AnyDice anyway? To help predict the results of such interactions?
Yes, AnyDice can be used to predict the results of multiple d20s. It can help when you're trying to measure the exact percentage of success for multiple dice.
Those were my first impressions of the OP and my last English class was a decade and a half ago.
>>45494146
He don't think it be like it is. But it do.
>>45494296
>>45494749
Get in there Masters student. >>45494647
The world needs you.
>>45494296
Oh look, another valuable liberal arts major! Let me guess...minor in Women's Studies?
>>45494296
Is this like one of those memes where the guy talks about all his military accomplishments for several paragraphs? Have I been pasta'd?
>>45494061
Anon, here I got some specific types of rolling from d20 variantes that I'm aware of.
http://anydice.com/program/7acd
I must say from my experience that the Mutants & Masterminds (Mastermind's Manual) solution of rolling 3d20 and pick the middle one, while a bit slow at first, works WONDERS.
Hope that can help you in someway.
>>45495267
This...is actually pretty great. Thanks, anon.
>>45495023
>Does it matter if you roll X number of d20s as long as one of them hits the TN?
That's where the vagueness comes in. Is he asking if it will affect the swing? Yes. Is it a good fix for the swing? No. Willits Q results? Yes. Do any systems allow it? No. Does he want to know how exactly it will affect results? Well, just about everything attempted will succeed. Or is he looking at some sort of dice pool mechanic? I can't really answer in much detail because he's not very specific in his questioning. I mean, this question is literally, "will drastically changing the car mechanic matter?" I assumed he wanted more than, "Yes, idiot," but maybe I was mistaken.
>>45495510
And this is the last time I use voice text on Android. This is just... awful.
>>45495023
Guy you responded to here.
I think OP was still less than clear.
>"Does it matter if you roll X number of d20s as long as one of them hits the TN?"
Even your translation is less clear than >>45495267
>the Mutants & Masterminds (Mastermind's Manual) solution of rolling 3d20 and pick the middle one, while a bit slow at first, works WONDERS.
OP's intent could have been worse, was less than clear.
That it is possible to work out what OP *might* have meant does not mean anon was clear.
Pretending otherwise is silly and likely just being argumentative.
>>45495620
>Pretending otherwise is silly and likely just being argumentative.
You must have missed the majority of the thread then. This whole thread is more argumentative than asking for clarification or being helpful. What more can you expect from neo-/tg/?
>>45495747
>This whole thread is more argumentative than asking for clarification or being helpful. What more can you expect from neo-/tg/?
Fair enough, but my point stands.
>>45495620
>Even your translation is less clear than...
I agree that the OP is vague and can be reworded, but your example isn't a good analogy.
That anon is explaining how a tested and published system works. OP is asking if a specific form of d20 dice pool exists and if not, why. Even if you separate the fact that one is a statement and the other a question- they are conveying two different things.
>>45494061
OP here. I'm trying to make a game where players are given a certain amount of d20s per round to spend on combat actions. You take one result to keep. Players can add d20s to certain actions to extend them or increase the effect or save them to spend on reactions. As you can imagine, d20s are the primary resolution mechanic. Damage rolls are actually part of the d20 result. Even if you roll multiple d20s, you're only picking out one as your result. Sometimes you might pick a number that is actually lower than the best result purely because it provides more damage.
Again, sorry for the commotion and I'll answer any questions if you have them.
>>45495852
>OP is asking if a specific form of d20 dice pool exists and if not, why.
Citation needed.
But seriously, OP seems to have responded to that anon favorably:
>>45495410
>This...is actually pretty great. Thanks, anon.
Making anon's interpretation of OP's question probably a correct one.
And my point was that "rolling a selection of d20 and picking the middle one" has a clearer intent than "predicting the target number, rolling a selection of d20 and keeping 1." even though they could mean roughly the same thing.
>>45496309
So, a player can roll for multiple actions per round, adding one or more optional d20's to a roll to increase the odds of getting the ideal roll, then the player selects the roll they want?
And the player has an economy of d20 they can use or save for later rolls?
Sounds like a decent mechanic.
I am irrationally fond of the d20 and have been playing gurps recently, so this appeals to me.
>>45496436
Correct. I want an action economy based on spending d20s. Most actions have a mandatory die cost though.
For example: A basic attack costs 2d20s and you'd pick the one you want to keep. You can double the cost to 4d20 to make it a full attack, not only increasing your chance of hitting your target number but giving you more damage options to pick from on those dice that are considered successes. If you do that, however, you won't have any dice to spend to react to your opponents.
What I wanted to say is that people usually don't like anything that isn't a predictable bell curve where the results of rolls all congregate towards a mean. The math between target number and number of dice rolled is easily performed with the help of roll20 though, which should theoretically make it easier when deciding attribute values and the like.
I hope that explains a little of what I wanted to do a bit more succinctly.
>>45496690
*AnyDice, not roll20. Dunno why I typed that.
>>45495843
So does my point. When OP responds to clarify, everyone fucks off because they're only interested in shitposting or boasting about their Liberal Arts degrees.
>>45497342
It's not like you're adding anything to the conversation.
>>45494061
What the fuck are you on about OP?
Are you asking if you should let your players roll 5d20 and pick from the highest when doing skill checks/attack rolls?
Of course this is going to massively fuck everything up, and vastly overpower the PC's.
Just, keep whatever system you're using the same, and if you plan on make any alterations, make them minimal.
>>45498074
>>45496309
>>45496690
At least read the thread.
Ok, so, I'm not OP, but I do have a basic understanding of reading. What he's asking is essentially this:
If I roll 1d20, the chance of getting above 5 is 3/4. If I roll 2d20, the chance of neither getting above 10 is 1/4, so the chance of one getting above is 3/4. Thus, rolling 1d20 against DC16 is the same as taking the highest of 2d20 against DC11.
Using devices like Anydice, it's possible to get these numbers to line up. So, would this be a legitimate thing to do?
The answer is yes, but it takes a lot of fiddling to get the numbers right.
>>45498345
OP here. Yes that is basically what I was trying to say. I explain what I'm trying to accomplish more comprehensively here >>45496309 and here >>45496690
My point is that AnyDice SHOULD make fiddling with numbers easier since it shows exact percentages. There are only two variables here: Target Number (Character Attribute/Modifiers) and # of dice rolled (type/length of actions being taken). Since AnyDice can show me the results of any roll along both axes, I can use the results to determine what stats characters should be expected to have at minimum/maximum and to adjust the die cost of certain actions to be more in line with percentages I want.
>>45494636
That sounds like a you problem.
>>45494061
>I know, I know, d20s are super swingy.
Poppycock. If just all steps away from 10 or under (or 11 or over) -- where both systems have a 50/50 chance of success -- your probability of success on a d20 will never be as many as 5 percentiles away from your probability of success on 3d6. This means that +1 bonus on 3d6 is worth +2 on a d20 (and if you're rolling straight against an attribute score to see if you succeed, a 12 or under on 3d6 is roughly equivalent to a 14 or under on a d20).
>>45498833
So you want to take your AD results and work backwards to make sure the numbers work when using d20s? I don't see anything wrong with that. Just make sure the chances of success are reasonable.
>>45496690
That's an interesting way to do it. It gives a decent bonus to your ability to hit things, but you have to make sacrifices to hit hard.
That said, rolling four dice to hit is super OP. That's like an 87% chance to hit.
>>45502154
OP here. I assume the Target Number you set for success was at least an 8 or so for an 87% chance to hit with 4d20? That is an awfully high chance to succeed.
However I don't have the numbers decided just yet. I don't know where exactly character attributes will be at just yet, so I don't know if I will settle on 4d20 being the total number of dice a Full Attack provides. I don't even know if the 4d20 will be final number used for a Full Attack, yet.
This thread was created mostly to see if such a thing is feasible as long as I keep a close eye on percentages with AnyDice while I work out how many dice should be able to be used at maximum and where attributes should be approximately.
Right now I'm looking at 3-4 dice per roll at maximum and maybe 1-8 or so for attributes. Needs more testing, I just wanted to see what /tg/ thought about such an idea. I didn't mean to cause such a fuss.
>>45494061
The chance of one d20 not landing on 20 is 19/20.
1 - 19/20 = 1/20
The chance of neither of two d20s landing on 20 is 19/20 * 19/20 = X
The chance of either (or both) d20s rolling a 20 is 1 - X.
The chance of three d20s not rolling a 20 is 19/20 * 19/20 * 19/20.
Etc.
I almost failed Probability by the way. I just Googled this, which you could have done.
>>45505690
OP here again. You could have actually read the thread before making that reply just as easily...
I was unclear in the OP but I've clarified it several times since.
>>45505690
>One person is a liberal arts major
>Another is boasting about failing a class
Where I come from, getting a worthless degree and failing basic math are things that are looked down upon, not praised, and certainly not something people talk about like they should be proud.
>>45509482
I'm hardly boasting, I'm just putting into perspective that it doesn't take a math genius to understand probability. Math, and particularly probability, is fascinating to me, but I'm not inherently good at it. I'm a bio major, man. I just look at cells and species relationships all day.
Unless your system has successes per die as a core mechanic like World of Darkness, or something, rolling multiple dice is just silly. If you want granularity use percentiles, otherwise just change the target number for whatever your d20 roll is. Seriously, why do people never think to just change the target number?
>>45510737
>Bio Major
At least it isn't an English degree. Keep up the hard work, faggot.
>>45511183
Because the target number is the attribute of the character rolling. I'm not counting successes because the actual scale of attributes will be rather low and the reason I'm rolling multiple dice is because I've associated dice with combat actions. If I wasn't doing all that, I would just count successes. A different core mechanic for a different system.
>>45494061
>But if I can predict the chances of at least one d20 hitting the target number, does it really matter if I have the players roll say 2-5d20s and keep 1?
The odds are different from hitting it on one d20, duh.
If you tailor all DCs so that the odds of hitting success are maintained, i.e. 20 DC on 5d20keep1 is equal to 17 DC on 1d20; 19 is equal to 13; 18 is equal to 10 and so on, then yes you have made dicepool completely pointless and even actually harmful because it reduces granularity to no real gain.
(quests use multi dice+keep one system in order to make several anons feel like they are participating in a roll and to pad the numbers to make them feel success big and important even if it's still actually 50+% chance)
But keep one is not how actual dicepool systems work; they use different interaction mechanic between the dice rolled and generally reduce the die size from 20 in order to make it simpler.
The point of AnyDice is to calculate the odds of hitting a value on a roll, yes. Because XdY don't stack up linearly and that gets into statistics calculations deeper than is in common parlance.
>>45513563
The point is that it isn't supposed to be a dicepool in the manner that you're suggesting. There's supposed to be an initial large increase when you throw more dice at specific actions, but the returns are supposed to be diminishing with higher attributes and there's always a hard limit to how many dice you can roll per turn. Couple that with the fact that the damage roll is part of the attack roll and it actually fits what I had in mind quite well.
I'm also assuming people won't spend all of their dice on any one action each turn, aside from those who are outside of the action and can afford to. That may not always be the case, but I figure I can adjust how ranged weapons work (maybe they cost more die?) after I start testing.
Keep one is not how actual dicepool systems work, but this is not a dicepool mechanic that depends on "total dice pool result equals or exceeds minimum DC". I'm not worried about doing that wrong because that's not what I was trying to accomplish in the first place.
Again, my point is that statistics results from AnyDice can be used to work backwards to determine the comparative success rates of X vs YdZ. It might not be a beautiful bell curve, but it doesn't always have to be as long as it fits the margins of success/failure you have for your system (assuming those margins are fair). You don't need to get into statistics and probabilities with your players because you've already worked them out in your core mechanic and once explained to your players, they understand simple abstractions such as "high numbers equal good"!
tl;dr It isn't trying to accomplish the things you keep comparing it to, keeping things "simple" is relative, and I'd like advice specifically about how I might proceed from here and/or any ideas people might have.
>>45496309
>>45496690
>>45504605
>>45515516
This is actually pretty neat. I'm gonna steal it for my homebrew.
As for ideas...make melee and ranged combat skills separate from character stats. That way you can use different numbers of dice in combat and out. If you plan on rolling 2 dice minimum for most standard actions, that should inform the value of whatever skill you plan on using. You can use raw attribute/skill modifiers versus a single die outside of combat for simplicity's sake.
>>45516633
OP yadda yadda, I hadn't thought to make the skills separate from character attributes. What else does that? Dark Heresy?
>>45513563
Just how granular can a d20 dice pool be anyway?
>>45494061
frankly i feel that if you're using one of those systems where you throw a fistful of dice then you should stick with d6's. they're quick to roll and read and everyone has a pile of them, and the target number should be firmly fixed and certainly not shifting around from 1 to 20. early editions of shadowrun and world of darkness taught us that you only fuck with the roll along one axis - either number of dice rolled or the number they're aiming at - or else your odds of success become completely inscrutable.
>>45509482
>>45511688
You're very angry about people getting non-STEM educations for some reason. What's with that?
>>45509482
Where I come from the value of education is rapidly declining and people are entering school for fields that're rapidly growing, only to graduate into the same field that's now rapidly shrinking.
>>45520363
Thank you for the advice.
>frankly i feel that if you're using one of those systems where you throw a fistful of dice then you should stick with d6's.
Generally, I think comparison between a die result and a number on the sheet is just as a easy as reading a d6 and my system is mostly comparison from what I've thought up so far.
When it comes to counting (whether it be for damage or modifiers or having multiple dice or whatever), adding is the same no matter what die you use, though numbers may be bigger.
>The target number should be firmly fixed and certainly not shifting around from 1 to 20
Target number will be based on character attribute/skill (have to work this out more) and will not be shifting around from 1 to 20. In fact, it will most likely be hard-capped at a specific number to create some sort of bounded accuracy. What that number will be, I'm not sure yet, but hopefully I will it shortly.
>early editions of shadowrun and world of darkness taught us that you only fuck with the roll along one axis - either number of dice rolled or the number they're aiming at - or else your odds of success become completely inscrutable.
I will generally try to contain roll modifiers to either the Target Number or to the # of dice rolled, but I will say that other systems do use both axes (with mixed results) and they usually make sure the math works out. The odds of success might be a inscrutable for the player, but generally higher numbers in a statistic mean a better chance of success and as long as the chances are appropriate for the situation, AnyDice or similar programs can help create a guideline to follow when it comes to modifiers and the sizes of dice pools when balancing such things.
How else would things like Heavy Gear move from 1D6+1 to +2 to 2D6 if they didn't make sure the math worked out otherwise?
>>45494061
The point of a dice is to generate a number that cannot be predicted. Sure, you can build some elaborate sensoring system that might be able to predict the dice roll 99% of the time, in which case, if you can afford that, you might as well just go buy a tool for measuring nuclear decay and you cant predict shit because !!SCIENCE!!.
>>45523739
>The point of a dice is to generate a number that cannot be predicted.
That's blatantly false. The point of dice is to generate a number that is RANDOM. Dice probability is predictable. You don't need some elaborate "sensoring" (what the fuck?) system to predict those probabilities. You have anydice.com
I've got it.
First have them roll a D6.
Then, roll 6D20, arranging them in order from lowest to highest. Then, they take whichever the nth D20 where n is the number that they rolled on the D6.
>>45525398
Seriously?
>>45520490
Why wouldn't I be?
