Strange Dice Rolls

[ivioin]

I've been playing fg for a couple of months now and I've noticed something. In all of our games played atleast one person has had a day where we roll only 1-5 on d20 rolls. It recently happened to me a second time, does anyone have the numbers on how they do the die. Do we really have a 5% chance on every number? This is ridiculous that i'm pushing this system on people and they are not wanting to play it because the gm is getting awesome rolls and we are getting low numbers consistently. I don't know if it matters that we was playing pathfinder. Do we have extensions to help with this or can someone make one? Thanks!!

[Trenloe]

This gets raised occasionally, but is usually shown just to be a statistical anomaly that you get with anything to do with statistics the more you do it. FG uses a 3D dice physics engine to determine how the dice roll, the position of the dice at the start of the roll is randomly determined.

Sessions where someone has crap rolls happen - they happen in real life games, just as they do in VTT games.

There's nothing you can do to change the dice physics and the way FG determines the dice results. I'm wondering what you were asking for? A way of hard coding probabilities and then take away the random nature of the rolls or something?

Anyway, If it's such an issue with your players, enable manual dice rolls and get them to roll physical dice in front of a web cam and the GM can enter the numbers they roll. Otherwise you're stuck with the FG dice engine and the will of the god of chance!

[Trenloe]

As an addition - get in touch with Doug (ddavison) and see if he can provide you with some campaign game statistics. Info here: https://www.fantasygrounds.com/forums/showthread.php?37768-Campaign-Game-Statistics

[ivioin]

This gets raised occasionally, but is usually shown just to be a statistical anomaly that you get with anything to do with statistics the more you do it. FG uses a 3D dice physics engine to determine how the dice roll, the position of the dice at the start of the roll is randomly determined.

Sessions where someone has crap rolls happen - they happen in real life games, just as they do in VTT games.

There's nothing you can do to change the dice physics and the way FG determines the dice results. I'm wondering what you were asking for? A way of hard coding probabilities and then take away the random nature of the rolls or something?

Anyway, If it's such an issue with your players, enable manual dice rolls and get them to roll physical dice in front of a web cam and the GM can enter the numbers they roll. Otherwise you're stuck with the FG dice engine and the will of the god of chance!

No, statistically each number should have a 5% chance to be rolled. I've played tabletop games for a long time and never rolled a 1 2 3 1 2 4 1 1 2 3 1 1 1 4 2 3 1 respectively. Aftert the last one i pretty much gave up. Eight 1's in 17 rows should never happen and has never happened to me before and i'm not the only one that had rolls like this.

[damned]

No, statistically each number should have a 5% chance to be rolled. I've played tabletop games for a long time and never rolled a 1 2 3 1 2 4 1 1 2 3 1 1 1 4 2 3 1 respectively. Aftert the last one i pretty much gave up. Eight 1's in 17 rows should never happen and has never happened to me before and i'm not the only one that had rolls like this.

8 1s in 17 rolls 100% *should* happen if you play long enough.

[Bidmaron]

Trenloe told you how to use real dice rolls. That is the only alternative. The mythology that computer pseudo random generators do not produce statiscally reasonable results is prevalent but wrong. And your game log should have the real rolls listed

[seycyrus]

Not this again ...

You cherry picked a set of rolls that stood out to you.

[LordEntrails]

Pull out the games' chat log. Strip out ALL the rolls, not just the "strange" ones. Put them in Excel, graph them. Then show us a statistical anomaly.

Statistical analysis have been done, and they ALWAYS fall within statistical norms.

Until you can show more than anecdotal evidence there's really nothing we can do.

[damned]

Another way to look at your statement "each number should have a 5% chance to be rolled" is if you picked up 20 d20s and threw them once. Statistically you might be thinking that rolling 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 is possible. you have approximately 1 in 40 million chance of that actually occurring.
Rolling 3 1s in a row is likely 1 in 8,000 rolls
Rolling 4 is likely 1 in 160,000 rolls.
So while it might look unlikely. Its not just likely. It will happen sometimes.