If we all rolled the normal way for stats, how come he has three 18's?

[JDJblatherings]

I'm sorry, but I don't like your comparisons. Hole in one, bowling, injury from shaviing, winning an olympic medal are all examples in which you can actively try to improve those odds (practice). Rolling dice is a pure random event in which you can't train yourself (unless you train yourself to cheat).

Well, clearly you've never seen me bowl or golf ;-) . I assure you those were random events.

I also list the probabilities so some folks can get them inside their heads as not everyone is good at visualizing odds (if they were there would be no state lottery ).

 

[Thurbane]

Odds of injury from shaving: 6,585 to 1

You're joking, right?

Either you are defining "shaving injury" in the Pythonesque manner of self-decapitation, or you're shaving with a piece of wet spaghetti...

 

[JDJblatherings]

You're joking, right?

Either you are defining "shaving injury" in the Pythonesque manner of self-decapitation, or you're shaving with a piece of wet spaghetti...

I pulled the stats from a site about probabilities. By injury they could mean any scratcch that requires some treatment beyond dabbing at it with some bathroom tissue.

 

[orsal]

for an almost 1 in 10 chance of an 18 in any one ability score. So 1 in 1000 will have 3 scores of 18 when rolling 4d6 drop the lowest.

Your combinatorics is a little weak. You cubed 1/10 to get 1/1000, right? That would be the way to compute the probability that three consecutive characters will each have an 18 -- not the same as one character having three 18s at all.

Here is the correct computation.

On one 4d6 roll:
P(18)=21/1296=.0162
P(<18)=1-.0162=.9838

On six such rolls:
P(no 18s)=.9838^6=.9066
P(at least one 18)=1-.9066=.0934
So far, your estimate is pretty accurate. But now:
P(3 18s)=(20)(.0162)^3(.9838)^3=.0000810, about 1 in 12,000
You're off by a factor of more than 10.

The reasoning: There are (6 choose 3)=20 possible ways to choose which 3 of the six stats will be 18s. For each such way, there are P(18)^3 * P(<18)^3 ways to get that result.

 

[woodelf]

Why this self-deception? It seems entirely common in the D&D community. If people want super-high scores, why don't they just use a different means of stat generation? Now, if I hear people rolled for stats, I assume something shady is going on.

Well, what you're describing does seem a bit iffy. But don't ignore the possibility of luck in this. I'm one of those people who consistently rolls poorly--and it's not *just* selective memory. Went through an entire D&D game without rolling double digits on the d20. My most recent D&D session was similar: most memorable was the attempt to pick a guy off with a bow, where i needed a 9 to hit him, and a single hit would've taken him out, and, at two attacks a round, it took me 4 rounds to finally hit him.

And then there's my friend, with a similar affliction, who recently rolled up a character for a
D&D game. 4d6-drop-lowest, and he ended up with a net attribute modifier of +1. When people didn't believe him, he proceeded to use an online die roller (so, no chance of biased dice), and ended up with a net attribute modifier of -1. Think about the odds of 4d6-drop-lowest producing an across-the-board below-average result for 6 rolls. And yet it happens. That's why it's called probability.

Heck, i remember back in the day watching a player in my game roll 5 18s in a row--followed by a 9, i think.

 

[shilsen]

And then there's my friend, with a similar affliction, who recently rolled up a character for a D&D game. 4d6-drop-lowest, and he ended up with a net attribute modifier of +1. When people didn't believe him, he proceeded to use an online die roller (so, no chance of biased dice), and ended up with a net attribute modifier of -1. Think about the odds of 4d6-drop-lowest producing an across-the-board below-average result for 6 rolls. And yet it happens. That's why it's called probability.

Yup. Just rolled stats using the 4d6-drop-lowest method for a PC in a PbP game: 10, 12, 9, 6, 9, 12 (net attribute modifier: -2). It happens. Luckily the DM was allowing us to roll 2 sets and choose 1, so my second one was much better.