how many times do you roll "1"'s?

[Dingleberry]

There is an recognised hex that is banned at our gaming table, no one is ever allowed to say.

"You need anything but a 1."

A similar situation led to one of the funniest moments at our gaming table in recent weeks. The fighter/rogue needed to roll a very easy Reflex save to avoid being blinded by the cleric's blinding shield, and the cleric's player said "What are you worried about? You only need what, like a '2'?"

When the "2" came up, the fighter/rogue player launched into a mock tirade: "Why would you say that out loud!!! You've been playing D&D for 20 years and you say THAT out loud!?! You KNOW the GDI* are listening!!!"

*At our gaming table, the "GDI" refers to the "Gods of Dice & Irony", who we believe to be in ultimate control of the dice. The dogma of the GDI is long and convoluted - I'd be happy to expand upon it if anyone's interested.

 

[Gryphon]

Statistically you have the same chance each time of rolling a 1. The chance is 1 in 20. There is no difference between the first 1 roll and the 100th in a row, the chance of rolling a one is the same every time.

In Game world every 20 has to be ballanced so get ready for the 1. The God of Dice decrees that you will roll a 20 when you least need it and a 1 when you most need a 2 or more.

Hence classic "Damn what a waste" when rolling a 20 when you are rolling for a reduction in price of your lodgings.

 

[Drawmack]

Actually applying the laws of chaos theory previous rolls do affect the present roll, but not enough to make it worth while to do the math.

Here's an interesting test your group can do. Each player takes a piece of notebook paper and does this

Name:_______________________

d4:

d6:

d8:

d10:

d12:

d20:

Then everytime you roll a die jot down the raw result. At the end of the night everyone total up each die's rolls and divide by the number of rolls, then add the results for each die together and divide by the number of players. I've done this a couple of times and the results are always interesting. Lets you know how, on average, each individual and the group rolls. The GM can do it too, but he should keep his scores seperate from the party.

One group that I was in had a mathematician in it and he would have us do this about once a month or whenever someone got new dice. He knew the odds of every roll working and he was almost always right. It's really neat what you can do with a bit os statistics.

 

[alsih2o]

Take the simple average and you should have a value pretty close to 10.5 (for a d20).

i am also not a mthemetician but this is not what you are looking for. rolls of only 12 and 9 will also avg 10.5.

i think what you are looking for is more your distribution of rolls and not just an avg.

 

[Alchemist]

i am also not a mthemetician but this is not what you are looking for. rolls of only 12 and 9 will also avg 10.5.

i think what you are looking for is more your distribution of rolls and not just an avg.

Hrm. Yeah. You're right.

Time to roll some dice and whip out excel for that bar-graph so you can examine the distribution.

But the big question then is what is a reasonable variance for 100 trials? or 200 trials? On 100 trials one would expect to get 5 of each number come up, but what is reasonable deviation? 4-6 times? 3-7 seems too much, but 4-6 might be too much on 200 rolls. Stupid statistics.

 

[alsih2o]

But the big question then is what is a reasonable variance for 100 trials? or 200 trials?

i would like to stress again that i am NOT a mthemetician...

*alsih2o goes and hides in his studio, where none expects math from him*

 

[Alchemist]

i would like to stress again that i am NOT a mthemetician...

*alsih2o goes and hides in his studio, where none expects math from him*

Hehe. That wasn't really directed at you, but at the more mathematically capable amongst us. I'm only as much of a mathematician as I'm required to be for school, which is more than I would like to be.

 

[Olgar Shiverstone]

But the big question then is what is a reasonable variance for 100 trials? or 200 trials? On 100 trials one would expect to get 5 of each number come up, but what is reasonable deviation? 4-6 times? 3-7 seems too much, but 4-6 might be too much on 200 rolls. Stupid statistics.

For a d20 rolled 100 times, a given value is expected to appear 5 times with a variance of 4.75 (std dev of 2.18).

For 200 times, the expectation of a given value is 10 with a variance of 9.5 (std dev of 3.08)


(Math: this is a Binomial distribution with p=0.05; q=0.95. E(x)=np, Var(x)=npq, variance is standard deviation squared)

 

[Ferret]

5% if you roll 20 at once, roll one 5% chance for a 1, roll again and there is still a 5% chance, not 10%.

But a player of mine rolled a 1 initiative the other day.

 

[Brown Jenkin]

To make some of you feel beter, a couple of weeks ago my die was realy against me. The night started with 4 consecutive 1s followed by a 12 and then another 1. At least the night ended better with only 50% of all my rolls being 1s.