Players are complaining about low rolls (they think it's a FG problem)

[Erin Righ]

Well I have to say, my dice favor my players, if I ever run a game, play, I actually have to fudge rolls to make things harder for my players, if I don't they don't feel that I am challenging with the amount my dice fumble in their favor

[PTBBC.ORG]

The only thing we ever see that my players talk about with dice rolls is the same number coming up several times in a row. Last night I rolled 3 D20's for checks and got 3 8's in a row. We see that from time to time, but we don't complain. Just laugh and roll with it.

[Donrav]

Seems that when you roll the dice manually into the box it's all fine. Hence the reason most of the investigations seem to have the proper outcome.

However; I did notice that each session at least 1 player has the "lucky rolls" and at least someone has the "bad rolls" when using the roll clicking from your character sheet for combat actions.
It changes each session who it is and who is affected, but rolling 4 times a 16 in a row, and almost only 17, 18 and 19 on fight related rolls by the DM or any player stands out and can happen once, but is very unlikely to happen multiple times across only a few sessions.
So with the physics in mind, it could be that the starting point of the dice is determined per session per character sheet only for character sheet based rolls and thus effects the overall outcome per session.
This will be very hard to prove as you would need to record behavior across multiple sessions with multiple characters only on the fight related rolls.

However, we do seem to be able to determine at the start of a session who has the lucky rolls and who doesn't.
And that is very suspicious of some code interfering with the actual supposed randomness to me.

So I'm very biased on the actual coding precision in FG when relating to the character sheet actions.
i would love to see the actual programming code of the dice behind the character sheets at some point.

[amerigoV]

I suggest a different approach. Accuse your players of using loaded dice when they play in person. Make them provide their real dice are not biased.

(tongue in cheek, or course)

[tahl_liadon]

...I suspect there are fewer complaints when rolls are higher

X-D true. true.

to gm with complaining players -- tell them there are 3 proven scientific methods to improve rolls:

1) change dice color at least 3 times during a session; best result: immediate after a roll of 1
2) "flush the dice" -- set d20 to roll *at least* 20x in a single roll in chat window; best result: when d20s is set to roll all at once and fill the entire chat window
3) (this one works every time) say out loud "20" whenever player wants a crit

works for me.

[LordEntrails]

Their is a 'chatlog.html' file saved inside the campaign folder for each campaign. You can use the chatlog scrubber (found on the forums via a search) and modify it to pull out all the rolls and do statistical analysis on the rolls.

Then you actually need to go study statistical analysis and distribution. What you think you are seeing you will learnis within statistical norms (see the results of your statistical analysis of your actual rolls) and has to do with the human psyche and the fact that humans tend to take note of those things they think prove their point and ignore those facts that disprove their thesis. i.e. One player comments that they always get low rolls, so everyone notices the low rolls they make and ignore the high rolls they make.

[Speculi]

I extracted every D20 roll once from the chatlog in a D&D 5e campaign, counting also the dropped dice of advantage/disadvantage. By now I should have even more data, but I don't see a point in going through this again. The theory of per session bias can't be addressed with this method, so maybe that's something one could look into... but in general the probabilities seem to be right.

2231 rolled D20 in total. Each value rolled with relative amounts in a range of 4.08% to 6.01% seems reasonable close to the expected 5%.

Here are the complete results.
https://docs.google.com/spreadsheets...EMd2kXWDofCnWE

[Bonkon]

Good Day All
Last nights session was a classic example of a bad streak of rolls. Our Monk (5E) was attacking and rolled three consecutive 1's in the same turn. To make matters worse our DM is using the Crit/Fumble tables The player is convinced the DM has a way to force a fumble!! LOL

[Mgrancey]

I have either co-players or players (forget which campaign) first they they do after logging in is to roll something like 20d20s to see what they spread is and if they need to change the color. Personally, I have found that changing the color and the text of the dice so I can't see what i actually roll is reported seems to benefit me (so white dice on white background OR black on black), mind you this ISN'T something I have scientifically tracked, just something that I have run thru my head.

[JohnD]

When in doubt always change the dice color... scientifically proven to make a difference.