That's the easiest reading

occam's razor

when you cast it, it has one target

Just like when you cast Dragon's Breath, it has one target

@Jason_c_o but it is capable of targeting more than one during the spell. I actually don't think reading it that way helps but I'd have to think about it.

But it doesn't ever target more than one. It just changes target

> Twinned Spell. When you cast a spell, you may spend sorcery points equal to the spell's level (or 1 point for a Cantrip) to duplicate its effects. You may choose a new target for the duplicated effect. No creature may be affected more than once by this spell at the same time.

Would anyone object to that revision? It makes the metamagic more powerful, but for being a metamagic that, theoretically, costs more than any other, it seems reasonable.

Target A and Target B are more than one target.

It's not target A and target B

it's a new target A

You can cast Hex and never have it change targets

@Xirema That'd be exactly what I'd run Twinned Spell as at all times if I actually played 5e

Regardless of silliness from Crawford and targeting and jazz. This revision just makes it easier and we don't have to define "target"

@Xirema "duplicate its effects" is not really clear to me actually.

And I do object to opening up twinned to AOEs in general which that would do.

@Jason_c_o I think changing targets is equivalent to more than one target.

I don't at all

it's still only ever focused on one creature

(at a time)

@Rubiksmoose Hence the "no creature may be affected more than once". Makes it worse than Heightened Spell if you could already fit all the creatures inside the AOE range.

Bob and Jerry are different creatures. Different creatures different targets.

@Jason_c_o you can also have magic missle hit only one target. It is specifically disallowed by the errata. It's whether it is capable of targeting more than one which hex is

It's focused Creature X (Target A). THen Creature X is killed and you shift to creature Y (Target A). You have targeted two creatures.

@Xirema But I object to allowing AOEs even with that clause.

Which makes me think about how Protection would work...in that those attacking you with disadvantage aren't targets. You are the target and you being you is generating disadvantage for certain creature types.

@DavidCoffron I still think those are different instances. Magic Missile is capable of targeting more than one at a a time

Hex isn't

@Xirema but some AOEs have smaller effects for balance reasons. If you twin it for double the AOE it becomes very strong (drag someone through two squares of cloud of daggers every turn for example)

now we have apples, oranges, and pears

@Jason_c_o that "at a time" language is absent from the text. It doesn't care when you target those creatures, just that you do

@Jason_c_o Can we combine them into delicious franken-fruit?

@Jason_c_o and now i'm hungry.

@DavidCoffron I would refer you to "no creature may be affected more than once". Maybe revised to "in one turn/round"?

To me the wording is implied. There's a lot in 5e that isn't explicit

and me too, I haven't had breakfast yet XD

@Xirema How would you stop an enemy from trying to pas through a Twinned wall of flame then? would the second just not count?

@Jason_c_o I don't see that implication at all.

Then we'd rule it differently haha

@Rubiksmoose If it's in the same turn, then yes, the second would not count.

@Xirema Even if they chose to go through it?

I think twinned spell still needs a target. Something like a twinned Flaming Sphere would be pretty baller.

@GreySage Ballers.

@GreySage or twinned spiritual weapons

And then they would be able to pass freely back and forth through both walls for the rest of the turn with no ill effects?

@Rubiksmoose They can already do that as-is with a single wall, I think? Since the effect still reads "starts their turn there, or enters the area for the first time in a turn"

I think that's a straw man. Wall of Flame can't be twinned in the first place

Stuff that specifically limit areas/numbers would be a little unbalanced. Like Giant Insects, or Glyph of Warding

@Jason_c_o (We're considering the revision I submitted a few pages up)

my bad

@Jason_c_o It would under the new ruling proposed. Also a straw man wouldn't last long in a wall of fire.

:)

haha

@Jason_c_o how can you add a temporal clause to the text implicitly. Would you also rule that you could cast wish again after losing the 33% because it's not "in the same day"?

@DavidCoffron Wish is one of those instances where it is explicit. "ever again"

@Xirema huh. I think you are right. That is surprising.

@NautArch GASP!!!

IMO, I don't think increasing the AOE of a spell is that powerful. It's powerful enough to justify the expenditure of some Sorcery Points, but it's pretty rare to have an AOE nuke spell that doesn't also have the AOE of a... well, nuke. And if you invoke "well, what about Meteor Swarm"/whatever, I generally consider "balance" a non-factor when 9th level spellcasting is involved.

I didn't know we were running some sort of nudists' colony here....

@nitsua60 Be careful of your 'friends' :)

I mean, you can wipe out entire civilizations with a well-timed use of the Wish spell.

and i think you mean: BLLLEEEEAAAATTT

@Xirema I guess my issue is that it deviates too far from the obvious and intended use of the ability for my comfort. I think we are throwing the baby out with the bath water here. Sure it makes it easier (maybe I still think the AOE thing would lead to weirdness I can't think of exampels right now though) but we end up with essentially a new ability.

@Jason_c_o okay. Bad example, but there have to be cases where reading clauses into the text is just absurd. I don't see the logic of adding the clause to Twinned Spell as a RAW or RAI interpretation

It's just now how I read it. "Capable of targeting more than one creature..." Hex isn't capable of targeting more than one. It just changes targets

that's not more than one

that's one, and then maybe another

I can see the other way, but that's not the path my brain takes haha

I really think it's more about how many targets can the spell affect. Hex/HM is more than one. Dragon's Breath is a single target (but they can use it on more than on person), Haste is a single target (but they can use it on more than one creature), Protection is a single target (but it can have an impact on more than one creature's interaction with that target.)

@Xirema That is right. What Crawford is saying makes no sense.

I think perhaps a modification that fits my idea better would be to change it to saying "incapable of affecting more than one creature on the turn it is cast". What do people think of that?

@Rubiksmoose that would make hex/HM possible?

That's kinda how I read it. I'd twin Dragon's Breath. I don't think it makes sense to include whom the character actually breaths on

Remove the target argument completely and make it say affect since it is pretty clear what that means. Then the turn limit is to prevent confusion about hex and other spells who can expand later.

@NautArch HM?

@Rubiksmoose That would certainly clear up a lot of ambiguities, though it does shut down Dragon's Breath, as cast on yourself would enable you to Quicken it out and then use the action to deal damage.

@Rubiksmoose hunter's mark

@Xirema You can still do that though but only on yourself. I guess if you want to Twin you have to accept extra restrictions.

@NautArch ah ok. Yeah it would allow both of those.

@Xirema by that logic it also keeps the hex confusion since you could kill the target during the same turn

@Rubiksmoose I don't think I'd want that allowed - but that's just me.

@Rubiksmoose I think it takes a more pro-player logic to the ruling and also makes it easy for the DM to adjudicate as there is nothing I can think of that would be unclear with the new parameters.

@Rubiksmoose HM still is

It has the ability to affect multiple targets in one turn but not at the same time

@DavidCoffron Then it would not be allowed.

I'm still not confused by Hex.. "A creature you can see" is one. It can hop targets, but only if the other dies, so it's still one.

@DavidCoffron Thta's a good point. If you have extra attack you could attack/kill, bonus action change target, attack.

And then you've got multiple targets in a single turn.

where does Hex mention it's limited to a turn?

@Jason_c_o Consider this: if a spell said "select two targets" would you allow a player to pick the same creature for both?

@Rubiksmoose Jason has been arguing that it is still only one target

@Rubiksmoose that's a different beast. If you can pick two, it's no longer elegible. But you aren't picking two

@Jason_c_o I was discussing whether Rubiksmoose's supposed revision solves the problem

@Jason_c_o No I mean ignore Twinned for the moment.

@Rubiksmoose it wouldn't matter. It's still capable, just like Magic Missile

you can pick one target, but it's capable of more

Hex isn't

@Jason_c_o why isn't hex?

I would suggest we not discuss Twin and Hex in the same breath, since they said they did not balance all t his for Multiclassing?

i thought both hex and Hunters Mark worked the same

Can Hex hex both creatures at once? @NautArch

No

@NautArch yes, that is how I understand it.

@Jason_c_o Here's what I was trying to get at. Targets in this game are differentiated by creature. Regardless of when you choose the target or how a different creature is a different target. Saying that chosing creature A on turn 1 and then moving it to creature B on round 2 and counting them as one target would be the same as allowing both creatures to be considered the same target if chosen at the same time as well. It doesn't make sense.

@KorvinStarmast did they balance it for feats? You can get hex from magic initiate.

They are different targets, but it's not targeting more than one creature. Ever

it's never targeting more than one target, iow

@Jason_c_o Concurrent Targets !== Targets. Just because Hex only targets one creature at a time, doesn't mean it only targets one creature

@DavidCoffron I guess I kind of missed your feedback, what did you think of the modification?

I think it does. I'll accept you don't agree

@Rubiksmoose I could see it being used but it does change how it works fundamentally.

It only targets one creature at a time, but over the lifetime of the spell it has 1+ targets

Hex is different from Magic Missile, and I would twin Dragon's Breath as well

@Jason_c_o And neither can Hunter's Mark, right?

@DavidCoffron Does it? It makes it more permissable sure, but the underlying concepts are the same about it I think. And it is much easier to use potentially.

@Jason_c_o It could target more than one in a single turn.

I don't see it mentioning turn anywhere

I think tha'ts a detail you're caught in that doesn't matter

it never targets more than one

it's not capable of targeting more than one

@Jason_c_o They are talking about my revision I think

@Rubiksmoose yeah. I would certainly not protest to it as a house rule. Can't think of any examples that it changes other than removing some ambiguity on DB. I'd say it's a good revision

maybe? I've lost track XD I'm talking vanilla, if it matters

@Jason_c_o The only way this statement can be true is if you accept your thesis that "at a time" is implied. If you disagree with that then that statement is false.

But there is almost no way to prove that it isn't implied so it is not really fruitful to keep arguing about that point with you.

I've already said I think it's implied. It's the way that makes sense to me, and it's the reading that causes less confusion.

@Jason_c_o If we're talking vanilla, i really don't think it works. The difference is in your reading of it as "one creature at a time" and i'm reading it as "more than one creature targeted for the duration of the spell"

but then you're assuming the "duration of the spell" is implied

you're guilty of the same thing

Neither are explicit or implicit (or right or wrong)

@Jason_c_o take Channel Divinity. "Beginning at 6th level, you can use your Channel Divinity twice between rests." Is there an implicit "[at the same time]" after "twice"

@Jason_c_o You did, but I think if we are going to continue this discussion we are going ot have to buy into one or another's reading. Arguing abou the same base point will get us nowhere. You read it one way, the rest of us read it the other.

@DavidCoffron I don't think so, feats are an optional rule/variant.

I think the other way causes the confusion haha. I'll accept I'm the odd one out, but we're both adding something. Either "at a time" or "for the duration"

but the intent based on tweets is not wanting more than creature targeted.

"At a time" is a pretty large assertion to add to something as implicit. If you came across a spell that read "make an attack against one creature. If it hits, deal 1d8 psychic damage, and then you may make another attack against another creature in range, dealing the same damage. You may repeat these attacks until you miss or run out of targets".

@Jason_c_o I'm not adding "for the duration". It says if "the spell targets more than one creature". Hex targets more thsn one creature

no it doesn't

It's pretty evident that this spell has multiple targets, but if you go by your reading of adding an implicit "at a time", then the spell only has one target

it targets one creature and that creature can change if one dies

you don't even need "at a time"

@Jason_c_o that makes 2 creatures.

2 distinct targets

oy vey

I don't want to go in circles too much here

@DavidCoffron Not as far as I can tell from the text in front of me. Channel Divinity takes an action to do, for a cleric, right?

Ok can we not keep doing this please. I think there are other interesting aspects to talk about. @Jason_c_o is reading it their own way and there's no way to convince them otherwise. We are reading it another. If we continue to talk about this we have to agree to one or the other.

If I fired a fun at a range and then fired the gun again down a different lane, did I hit "one target"?

You fired twice

If we keep cutting in and disagreeing on the same unresolvable point we cannot have a discussion.

@DavidCoffron Yes, because your accuracy isn't that good.

that's not comparable to Hex

@Jason_c_o ok. The bullet ricocheted and hit another target

Different beasts

@DavidCoffron 2 targets by our reading, 1 target "at a time"

1 target at time of casting

@Delioth chain lightning is one target at a time.

@Delioth exactly.

@Jason_c_o I don't see how it's different. I fired one bullet (the casting). It hit the one target

Then it ricocheted (the target died) and hit another target (creature B)

@DavidCoffron Was the bullet made of ice? I think that's relevant here.

Hex doesn't work like a bullet

ceding one one isn't ceding on the other

So if I fling a fireball at one guy who (from my perspective) is standing in an open field, is that only 1 target?

@ColinGross ice knife is another good counterexample actually. The initial casting targets one creature but then does an AOE

@Jason_c_o But six shooters have a HEXagonal arrangement of bullets. Soooo...

@DavidCoffron AIR?

good luck getting a bullet to ricochet off one person into another. I'd hardly call that practically capable

Oh

You go out and consistently hit two targets with one bullet

and I'll consider your analogy

@Jason_c_o What about just go through and into the person behind?

What if the "bullet" was made of lightning and did it really fast?

@Jason_c_o Needing to bring actual physics into a debate about magic is kind of a bad sign

that depends on a lot, and usually bullets don't go straight through

@Jason_c_o I have an image suddenly of someone putting an unshot bullet between their knuckles and punching people with it.

@Jason_c_o Depends on how hard they try

@Jason_c_o it's an analogy. I can hit two targets with a frisbee. Now will you consider it?

That's gotta be an Anime character, yeah?

@Delioth That's what I'm saying. You can't compare Hex to abullet

It could just as easily be me throwing a ball, bouncing if off the barbarian's head, and hitting the wizard

@DavidCoffron The analogy needs to be taken out back and shot. Then we can play frisbee

This poor horse.

I think it's time to move on.

Yeah the analogy sucks, is what I'm saying

@Delioth Off the barbarian, into the wizard, over the rogue... nothing but net.

@ColinGross But we can't shoot two analogies with the same bullet.

too bad

@Xirema We can, but they can't be twins.

but if we can't, then we can twin it

@Jason_c_o It doesn't suck, but it also doesn't help the discussion at all since it doesn address the assumption that we can't disprove.

then it sucks for it's intended purpose

My takeaway from this conversation is that we should all go out and start shooting twins. Everyone in agreement?

Okay, excellent.

@NautArch I agree. Lets do a reducto ad comedium like usual

@Rubiksmoose Undisprovable assumption.... "undisuption"

I will be soon suggesting a new metamagic: Tripled Spell. It will be much easier to implement.

@KorvinStarmast Can I twin and triple a spell?

@DavidCoffron That only affects small to large analogies. Small, tiny, huge, and gargantuan analogies are unaffected.

> Tripled Spell. Spend 3 Sorcery Points. Your DM immediately convulses over in an uncontrollable fit of sobbing and wailing. You instantly win D&D 5e.

> When you cast a spell that targets only one creature and doesn't have a range of self [...] To be eligible, a spell must be incapable of incapable of affecting more than one creature on the same turn on which it is cast, at the spell's current level.

So that is my proposed rewording.

And it also happens to bypass this entire debate.

@Rubiksmoose Wait, the spell has to be cast at the creatures level?

"be incapable of incapable of affecting ..."

I'd change "on the turn it is" to "as it is"

@ColinGross if I reduce a large analogy to a medium one but it's still bigger than other large analogies is it medium or large?

@Delioth So it HAS to affect more than 1 target. genius!

@DavidCoffron Depends on how many identical analogies come along with it.

I'd polymorph the analogy into a chicken and then hit it with a twinned hex

@Jason_c_o Truly

@ColinGross is a swarm of analogies 1 creature or multiple?

oh no NotLikeThis

@DavidCoffron Yes.

> Tripled Spell. Every number specified by the spell is instead treated as though it were multiplied by 3.

@DavidCoffron That's actually a bird. Swanalogy.

@DavidCoffron That's how we get six packs, so yeah. Beer is good for D&D.

@Xirema Can't wait to roll my d18

@DavidCoffron You are truly a monster lol

@Rubiksmoose no. Multiple monsters #SwarmOfDavidCoffrons

@Delioth That's just 2d3 and a coin.

@Jason_c_o But what if you ate the chicken wings before undoing the polymorph?

@Xirema What if it's "a creature", not "one creature"?

@ColinGross that's not even close to how dice work lol

then it dropped to 0 and I could change targets @NautArch

@DavidCoffron [screeches in panic]

@DavidCoffron You need 18 unique combinations, then that's how you get em.

@ColinGross it needs to be uniformly distributed to be a d18

You don't like value lookup tables? Not pro-dictionary, eh?

2d3+1d2 is not

@GreySage Depends on the language/culture you're playing in. 5e now has locale-specific mechanics.

@DavidCoffron 2d3 and a coin give you 18 combinations. Each is just as likely as any other.

@DavidCoffron You just need to differentiate the d3's, and use a table

A,B,C * 1,2,3 * H,T

Baby you and me now

@ColinGross oh k gotcha now

@Colin Gross, to get a d 18 you roll a red and a green d6. On the red one, 1, 2 is for 1-6, 3, 4 is for 7-12, and 5, 6 is for 13-18.

I thought you were multiplying all.the dice

the green one is the 1d6 variable value

@KorvinStarmast Oooo I like that!

now roll that with advantage

That's substantially easier

Technically, you can get dX from any number of dY dice, given enough of them.

@Xirema Does that work for primes?

@ColinGross It should, you just need enough dice.

> How to roll d17: Roll a d20. Reroll on a roll 18 or higher.

@Xirema Roll number of times = least common multiple and divide

Done.

@Delioth You can get a power of 2, but you need to drop a value.

(this is how PRNG algorithms give smooth distributions between arbitrary user-supplied domain values, FYI)

@Xirema Depends on your implementation. Modulo division doesn't work for non-uniform prng values.

@Xirema which is silly. The universe should have better math than "just ignore the results that you don't like"

@ColinGross This isn't modulo though. Modulo would be something else.

@DavidCoffron But it works.

@ColinGross it works if you want to waste precious microseconds

> How NOT to roll a d17. Roll a d20. Subtract 17 from any results of 18 or higher

(that would be how modulo division works)

You lose like 0.00004 US cents for each skipped value in electricity costs

@DavidCoffron Someone please think of the poor Capitalism.

@Xirema Its cheaper than developing something faster

@GreySage and it's much cheaper than creating an entirely new universe with better math

Only AO could do that

I like that. Better math. The Best math describes the universe perfectly, so that implies better universe

@DavidCoffron Typical lazy Centrists, unwilling to even consider more radical proposals like recreating the entire universe under new laws of physics.

@Xirema I thought the bounded result on top of mersenne was by modulo division not throwing out values that were outside of bounds.

@Xirema gotta save the hyper-semi-nano-dollars

@ColinGross PRNGs don't have values that are "out of bounds". If they're valid, they return uniformly distributed results in the range 2^N.

But the problem is that people don't think about what they're doing when they use those results.

So they write things like int value = rand() % 10;.

@ColinGross Modulo isn't good for that, because unless your values are a perfect multiple of your modulus you'll skew results

@Xirema yeah, so when I make a call for a random number between 17 and 48, the underlying implementation gets a random unsigned int back and makes the value fit with modulo division.

Not realizing that in doing that, they've made values 0,1,2,3,4,5, twice as frequent as 6,7,8,9

@Delioth you could use modulos if you find the least common multiple and expand your terms out to the right number of digits

@Xirema you have to make the search squares divisible by the range (32)

@Xirema I think you get uint32 or uint64 back depending on the implementation.

But many people just call whatever function gives them a random number in the specified range.

So in base 2 you need 5 bits. In base 8 you need 2. In base 10 you would need a whole lot

I prefer the tried and true function rand(){ return 4; }

Verified random by physical dice roll :P

@Delioth Works for me as long as you roll the die into a bowl of pudding.

@Jason_c_o Didn't want to put up a thread in the comments, but did you read the meta? I understand it being helpful, but there can be good reasons NOT to explain the downvote as well.

Very true. I've read it before but not recently. There may be good reasons, but the poster was a "new contributor" asking for reasons. It's good to point them there, though, I'll admit

@Jason_c_o There have been times i've explained my reasoning and it's gone to a comment war.

@Delioth ? Why are you trying to force a 4? Lol that's a weird number to prioritize

@NautArch Unfortunate.

@Jason_c_o And I'd rather them focus on getting answers to their questions rather than on downvotes.

I'd like for them both to get an answer and learn what makes a better question :)

@NautArch To be fair, I'm not actually sure why the downvotes are occurring. I think it is fairly interesting and well-put.\

Yeah I was curious too.

@DavidCoffron What are you talking about? The function's 100% random, it demonstrates the results of my random physical method. I rolled a die, the function returns those random results (xkcd: xkcd.com/221)

@Rubiksmoose I think Jack's reasoning may be part of it...but i'm going to put another comment up.

@NautArch I personally would not consider downvoting for that, and it seems a bit unlikley. I wonder actually if it is the title and people aren't really reading the question.

"How broken is..." is not usually the beginning to a good question in my experience.

@Rubiksmoose I didn't up or down - I don't think it was worth either. I think there's a lot of knee-jerk reactions to asking questions about upending existing rules.

That could be it. They seem to be common lately, along with balancing homebrew

Well I modified the title.

@Rubiksmoose might be like meta downvote. Disagree with the idea

@DavidCoffron Could be.

@ColinGross That is how we did 30 sided dice some years ago. d6 (used like that red one) and a d10 or d20.

@Jason_c_o I'm all for the homebrew reviews, but folks are not doing their due diligence before posting and that frustrates me.

I do really hate the use of the term "broken" in almost every case though.

@KorvinStarmast d6 and d20 doesn't map to d30...? Does it?

Oh it does. It's just messier

@Rubiksmoose Hoping they respond to my last comment.

@NautArch I wasn't arguing against them. Just saying they're more common, and don't always do their due, as you say

@DavidCoffron Our d20s didn't have teens on them. They had two zeroes, two 1's, etc.

@KorvinStarmast ... Oh

We would use colors to fill in which one's were high. But for percentile dice, or 1d10, we ignored the color.

That's how I was thinking you'd map it though. Take the mod 10 of the d20

a D10 has been a useless die for me, but I use them now. ;)

@KorvinStarmast I miss my 2d6 from my greatsword now that I use the 1d10 glaive.

@NautArch Yeah I'm not sure that it would really helpful for them to do so though. They are basically saying "I play with this rule that seems to widely be seen as overpowered. What issues are there, if any, that I am not seeing?" essentially.

And a self-answer does not further that.

@NautArch d10+3 is like way better than 2d6 tho

@NautArch I miss AD&D 1e Two Handed Swords 3d6 to larger creatures. :( But I have adapted.

@NautArch I'm hoping they respond to the DM/PC question though. I am unclear what is going on there.

@NautArch Then we close them and point them to meta.

That would prob be why it was facing downvotes

Are there people here who are better at combinatorics than I am right now? If more than half of the 22 players in my organized play group picked 3 of the 12 available classes, is that surprising? I think it's $\binom{12}{3} \sum_{i=11}^22 \binom{22}{i} (1-3/12)^(22-i) (2/12)^i$, so about 2% chance to happen randomly, so surprising; Am I correct?

as that one comment pointed out

@Rubiksmoose Except that if we all assume it'd be OP, but they've been playing with it and hasn't...that's good for us to know.

@DavidCoffron Well, technically it was a 2d6+1 (later turned to +2)

@NautArch 10d6s and map to a d10. So many dice

@NautArch sure but it doesn't help their question which is "what problems are other people seeing" as I read it.

But yeah, still better :) But I'd had that maul from level 5 (when I joined) to level 15 or something.

@Jason_c_o are we still talking about the proficiency question or the homebrew questions from the last c ouple of days?

@Rubiksmoose And we may have found the downvote reason :) If they've been playing with no problems, but they're asking where the problems are, that's some cognitive dissonance.

@Anaphory wait half of them picked the same 3?

@KorvinStarmast We try, but folks like to help and I think there's more of a rush to answer than a rush to close.

@Rubiksmoose I thought we were talking about the proficiency question

@NautArch agreed

@DavidCoffron Yep, Paladin and Wizard together have been picked by 10 out of 22 players.

@Anaphory blinks three times. Those aren't even the ones I expected lol

@Jason_c_o but that isn't really homebrew and it seems to have done their research as well.

@Anaphory but yes your math checks out fmpov

@DavidCoffron So Paladin plus Wizard plus anything-but-Warlock (which was picked by no-one) are half of the player base.

@Rubiksmoose I didn't downvote it and didn't call it homebrew. I said there have been an influx of balancing questions, including homebrew. It's just getting lost in chat.

@Anaphory Rogue is the big winner in my local group.

(Although I just used an online calculator for the stat distribution)

I've played the only paladin and only bard. We've had a lot of rogues, druids, and clerics.

@Jason_c_o oh I know you didn't but others did which is why I was confused lol. (and still am I guess). NBD.

@Rubiksmoose no worries :)

@Anaphory are they new players? There's a lot that goes into picking a class, and paladin and wizard appear a lot in pop culture.

They could just be influenced to pick what they know of

@Jason_c_o To some extent they are new. I guess that's definitely one point in favour of the wizard.

@Anaphory umm... that is a much lower percent chance

(If random)

@Rubiksmoose I was actually at one point working on a comment linking to the homebrew meta...but then i realized this was a houserule and not relephant.

@DavidCoffron Well, yes. But that's not what I was asking.

@Anaphory ofc. Just shocked

@NautArch Good call I think. Yeah I was just confused what all the cool kids were talking about without me ;)

I wanted to know whether I should be surprised by the most frequent 3/12 classes making up 11/22 players or more. I'm surprised that the chance for that might just be 2%, I feel like my calculation is missing some factor of looking for the most frequent three.

@Anaphory umm... I got .2 (misread the first time)

Using this calculator. Prob of success is 0.25, 22 trials, more than 11 successes

@Anaphory I think looking at it like it's truly random is where you're getting tripped. The choices are all weighted, and differrently so for each player.

Not saying it isn't an anomaly though haha

With 11 or more its about 1%

@Jason_c_o Sorry, Bayesian thinker. I have a prior for how those frequencies are distributed, and I want to know the likelihood of the outcome I see under my model of equal weight, so I know how much I need to adjust my priors.

@DavidCoffron Why do I not need to take into account that there are \binom{12}{3} ways to pick my best 3 classes?

@DavidCoffron Ah, also that is 0.0002=0.2%, right? I expected the probability that this happens to be somewhere between 5% and 20% – unlikely, but not very surprising.

@Anaphory Because which 3 are the best is irrelevant, I'd guess. E.g., it doesn't matter to the distribution if it's Rogue > Wizard > Fighter vs Warlock > Cleric > Monk

@Anaphory WoTC did a survey with a PC class breakdown. We discussed it a few months ago at GiTP.

@Anaphory it's just the chance that any 1 player picked 1 of the classes

3/12 or .25

dnd.wizards.com/articles/features/dd-survey-results-summary

@Anaphory Classes- 1. Fighter 2. Cleric 3. Rogue 4. Wizard 5. Paladin The paladin just squeaked by the warlock to capture fifth place. However, unlike with the races, the classes were much more clustered together. To draw a comparison, the least popular class (the druid) was just a point or two behind the dragonborn in usage.

Races 1. Human 2. Elf 3. Half-elf 4. Dwarf 5. Dragonborn The aarakocra turned out to be the least popular race, while of the non-core races, the goliath led the pack. As well, there’s a big gap between the dwarf and the races that followed, with the tiefling and a few other races following right on the dragonborn’s heels.

@KorvinStarmast I don't think I have race stats to compare this with. We took out asking registrants for their races because we didn't care, really.

@Anaphory what are your criteria for 'best' in the case of classes.

@KorvinStarmast Oh, did I phrase it like that? I meant “most frequent”.

@Anaphory if I can find the GiTP thread, I'll post a link. There was some interesting commentary on methodology, and "where did they get their data" that you may find interesting.

@KorvinStarmast I think there's some confounding factors here though. I know plenty of tables that outright ban aarakocra because flying.

AL appears to not like them either.

For the same reason, I believe.

Flying does tend to screw with... a lot of things

@DavidCoffron For any combination of 3 classes, of which there are \binom{12}{3}, the chance is that about-a-percent, coming from that binomial distribution with p=0.25. That's the $\sum_{i=11}^22 \binom{22}{i} (1-3/12)^(22-i) (3/12)^i$ I had in my formula above.

@Delioth Yes. Our organized play is not AL-conform, but we also banned Aarakocra for the same reason.

@NautArch Whenever I see the term "overpowered" in a 5e question, I have to fight back the urge to reflexively down vote.

Actually, I mistyped that. it should have been (3/12)^i, and then it's p=2.2, which is clearly not a probability.

@KorvinStarmast It originally said "How broken is..." I changed it. DO you think it is an improvement?

@Rubiksmoose Which one? Skill prof?

I feel the same way about broken honestly and it is usually a sign of poor quality if it is being used in my experience.

@KorvinStarmast Yeah

@Rubiksmoose Amen, and I agree with you on the correlation.

@Rubiksmoose That edit was well advised.

@KorvinStarmast cool cool, I was just trying to help OP past knee jerk reactions to the title so I just wanted to make sure I hadn't unintentionally made it worse (or not helped at all).

@Anaphory yeah. I was thinking about it while mopping the restroom (you don't want to know why) and realized that what I had done is the chance of a specific 3 being chosen 11/22 times, not any 3