Just rolled up my new character and (someone with math skills needed)....

[MerakSpielman]

I used the dice rolling method the DM told me to use. I got absolutely absurdly high scores.

My results: 17, 16, 17, 18, 14, 15.

That would be the equivilent of a 66 point buy!!!

The method she instructed us to use was the 3-2-1 method. You roll 3d6, reroll the lowest two dice, and reroll the lowest one of those. All rerolls are optional.

Can anybody with math skills tell me the average scores that would be generated by that method if you rerolled all results of 1, 2, or 3?

Am I just freakishly lucky or is my DM very generous?

The main problem I have is I already made up this guy's backstory. He was raised by a cleric who intended his sons to follow in his footsteps. When I went to the temple for training, they tested me and discovered I was incapable of casting spells (WIS too low). In shame, I ran away and lived in the wilderness, eventually finding a mentor who trained me as a barbarian.

I was going for a smart-strong-unwise character. If I follow exactly what I rolled (and I don't like fudging numbers) my backstory will be unworkable. I was more or less counting on having one score below 10 so I could put it in wisdom.

 

[Zogg]

I don't quite have the mind I did in high school (when I was a math FREAK) but based on the system your DM uses you should consistently get really high rolls like that. Just roll up another few characters using that system and you will see that those results are not uncommon.

 

[Tabarnak Smokeblower]

Since rerolls are optional, figuring out the average of 3d6-reroll2-reroll1 would be a little complicated. You'll have to establish rules (reroll all 2's and 1's, keep the rest, for example).

But I think your DM is being EXTREMELY generous.

As for your backstory, I don't think your DM will mind if you reduce one of your attributes. Hell, generous as he (she?) is, maybe he'll let you boost one stat if you reduce another

I'll try to statistically analyze the problem this evening (but don't count on it too much. I have tons of house work to do AND dvd 3 and 4 of Futurama season 2 to watch )

TS

 

[Dakkareth]

Transplanar double-post. Damn down-times

 

[Dakkareth]

A quick and dirty program I wrote gave the following result:

3-2-1 reroll gives an average of 13.5 (equivalent of 39 point buy)

as opposed to

4d6 drop lowest giving an average of 12.2 (equivalent of 28.5 point buy)


Note: I set the program to reroll, if a result was less than 4. This gives the optimal end result.


Conclusion: It's a method that brings powerful scores, but you also were very lucky.

 

[Tabarnak Smokeblower]

Might i inquire as to how you constructed your program?

TS

 

[Cedric]

I can't speak for the odds and dice results...but explaining your character background is easy.

The church was wrong. Maybe they got you confused with someone else? Maybe the headmaster was just incorrect? But they thought you weren't able to handle the strength of mind necessary to power divine magics...and they were wrong.

But my favorite explanation for it? Your character has a special destiny of somekind that did NOT involve him starting life as a cleric.

So, an agent of the deity, knowing of this destiny, simply refused you the initial orisons and ability to channel divine powers.

Cedric

 

[Conaill]

A quick and dirty program I wrote gave the following result:

3-2-1 reroll gives an average of 13.5 (equivalent of 39 point buy)

Note: I set the program to reroll, if a result was less than 4. This gives the optimal end result.

I got an average of 14 off the top of my head, using always rerolling. If I got it right, expected value for the first die will be 5.5, 2nd die 5, and 3rd die 3.5.

BTW, rerolling on a 3 or below is only optimal for the *last* roll. for the two reroll, I believe optimal would be to reroll on a 4 or below.

Either way, it does sound like you got lucky on the rolls...

 

[Tabarnak Smokeblower]

(my replies don't seem to be coming thru, i'll try anyway)

Just ran some numbers, reroll if < 4 method, and I, too, come up with a 13.5 average on rolls.

TS

 

[Dakkareth]

Might i inquire as to how you constructed your program?
TS

1. Take three random integers between 1 and 6
2. Use the maximum and remove it from the list.
3. Reroll the others if they're <=4*
4. Repeat from step 2
5. Do 1-4 10k times and calculate the average

*
Depending on which programming language I use. Python seems to have slightly different pseudorandom results than Pascal. With Python I'd use <4, with Pascal <=4 for optimal results. That could be some error of mine, though.

Also only rerolling the third die on a three or less seems to have a small, but noticable positive effect on the result.