Dice Roll Results

[PTBBC.ORG]

So my Sunday night group has been seeing a pattern for the last couple of months with dice roll results. Sometimes an attack roll for a specific character will get stuck in a rut and roll the same number every round.

We have a cleric that had to do some checks, she rolled a 2, then 2, then another 2, one more 2, then got a 4, 6, 8 and 10 in that exact order. It became a running joke.

My players say "Well, thats typical for FG." This software costs a lot of money to be able to play. I could have gone with ROLL20 (sorry), but I didn't because I liked the appeal of things to come (hopefully) with FG. But, dice results sometimes are predictable because of what we see the previous round.

So, is there something I can add (like an extension) that will random the dice rolls better? Because they are not random enough for me. Or is this some sort of glitch and we should just use real dice....

Also curious if anyone else has noticed this, or just take the multiple 8's in a row as "normal"

[LordEntrails]

I've never seen this. Been playing weekly for about 15 months. But I'm sure this would be of interest to the devs. They'll probably need you to grab the chat log and maybe some other things as well.

[TheoGeek]

Actually, multiple 8's in a row should be expected. If you had a program that generated a random number between 1 and 8 and you generated 8 numbers and none of them were the same, that would almost (almost) prove bias, not randomness.

Your example does look fishy, especially if you can predict die rolls, but personally, I've never seen this. I have seen the same number come up a few times in a row, but that is to be expected.

Perhaps what you want is not "more random", but actually "less random" as shown by the iPod's "shuffle" implementation - people thought it wasn't random enough because they would get multiple songs by the same artist sequentially, so Apple actually added an algorithm that made it keep track of that kind of thing and select different songs, not randomly, but with a bias against past performance.

[PTBBC.ORG]

Hmmm... Ok, that makes sense in a way. In my head it just seemed like the randomness wasn't happening, but you are saying that it is and its just coming up with the same number.

I had our cleric do an INT check, and someone said "Why, it will just be a 2" and it was. So that's what prompted this thread. It just "feels" like sometimes the dice rolls are "stuck" if that makes sense.

[Zacchaeus]

You can't really say that a random number sequencer isn't random until you can show via thousands of rolls that there is a bias towards certain numbers. Additionally just becasue you rolled a two last time does not guarantee that a 2 won't come up again this time or the next. Every time you roll any single dice each number has exactly the same chance of coming up. It's different if you are rolling two dice and adding the result (e.g 2d6 has a huge bias towards 7 since there are lots of combinations which add up to 7) but on a single dice there's no bias.

Whilst your number sequence is remarkable it can't be taken as any kind of evidence that the dice rolls in FG are anything other than random.

[kalmarjan]

It's nothing new. This has been a complaint since the beginning of FG (back in 2004.) Here's the thing. The other night I was playing in a home game of holdem poker, and even though we used a card shuffler, we were dealt the same hand three times, and even had a river of 666.

I can tell you about times where we rolled nat 20s four and five times in a row using dice.

It comes down to chance, and sometimes weird things happen when the dice roll.

There's also nothing stopping you from rolling your dice separately and telling the doe roll. In fact, isn't that feature coming in 3.2?

[Maspalio]

My players oten complains that they get too much 1's, more than with a real dice. From my point of view, it is just an illusion, since they don't complain when they get successes after successes in a row. The random number sequencer is plain random after all the test i did. It just comes sometimes with a funny randomness, just like with a real dice. I got plenty memories of game session at the table where someone rolled a 20, then another 20 with everyone jumping around in laugh. How about the chance to get the right number at the lotery ? 1 chance on a billion ? But people wins. The same with FG imo : it happens that you roll a 2, then a 2, then a 2, and it's still random.

[Zacchaeus]

There's also nothing stopping you from rolling your dice separately and telling the doe roll. In fact, isn't that feature coming in 3.2?

Yes

[Trenloe]

Manual rolling is a massive overhead, especially on the GM. It is primarily aimed at face to face games where the GM is keeping track of data in FG and perhaps sharing a map on another monitor, and they want a way of getting damage results and some other rolls into FG's automation. I wouldn't recommend it ad a replacement for FG's dice when players are connected through their own client.

[PTBBC.ORG]

Appreciate it guys. It was just kind of funny and we joke about it all the time, so wanted to post the question and get thoughts on it.

[Nylanfs]

There's also the added complexity that FG uses a physic's engine to determine the roll result as opposed to a standard random number generator.

[kalmarjan]

I was just about to write this. My understanding is that it is combination of an algorithm and a physics engine...

[JohnD]

Yes
Is this the die rolling "fudge" mechanism that was talked about a while ago where the roll is just for show and the result/total is predetermined?

[damned]

I have *never* seen this behaviour in 6 years of regularly using the software. Are you sure the player wasnt doing anything incorrectly? What ruleset are you using? What was your player clicking on? Was it the same dice roller every time?

[Moon Wizard]

John,

No, this is just way to enter die rolls manually, for people who want to roll dice at a real-life table, but still use the automation.

Regards,
JPG

[JohnD]

Ah. Too bad. The ability to fudge results would be nice to have, especially since FG shows hidden modifiers.

[Nickademus]

I had a player roll four natural 1's in a row last game. He was rolling in the dice tower, so it was my dice rolling. Luckily he wasn't in combat (and the chest wasn't actually trapped). But I quite shocked.

[Moon Wizard]

John,

I think the answer is to look at the modifier visibility rather than creating an interrupt solution. In addition to the difficulty of pausing a roll on a remote machine to "fudge", it would slow the game drastically to interrupt every roll to allow fudging. Plus the players know you're fudging if you don't have it on all the time.

Cheers,
JPG

[Nickademus]

You are implying that we wouldn't be fudging all the time...

[Moon Wizard]

You might as well play diceless, since it doesn't matter what they roll if you're fudging all the time. ;)

JPG

[Nickademus]

But then what would the players blame their bad luck on?!

[JohnD]

If an effect is marked GM only I feel it shouldn't show up in the summary line on any roll it has impacted.

There's not much point being "secret" if the program tells people something is rotten in Denmark.

[Maspalio]

I had an IRL vampire session last night. I rolled 7 ten-sided dice for an attack, and i got 10,10,10,10,10,10, 8. What are the chances to get such a golden roll even if you wants to ? **** happens. Good stuff too. And though it can seems biased, weird or unfair when this arrives, you still have a probability to chains five times a 2's throwing a dice. It's unlikely, but it still can happen.

[kalmarjan]

That reminds me of a SW game where a guy rolled 98 damage from a shotgun blast. Gotta love the exploding damage die!

[Zerbious]

I had a player roll four natural 1's in a row last game.

That is insanely unlucky! We had that happen to our ranger in a RL game using actual dice last night, twice... one when he had ADV and rolled double 1s!!!

The DM was even nice enough to leave a message in our group chat this morning, LOL:

Double 1's.
Never Forget.

[Nickademus]

Had another player last game make two attacks (during a major combat that they were struggling with); he was the heavy hitter. Both attacks were 1's.

Played in another game a couple weeks ago where a player had advantage and rolled double 1's. In the same game (maybe the week before), I tried something bold, a.k.a. disadvantage, and rolled double 20's.

These things happen more than you would think.

[LordEntrails]

:)
We roll a lot of dice. It's just not very memorable when you roll a 7, 4, 15, 19, 1, 13
But when you roll 20, 20 and slay the BBEG on round 1, well, that's memorable :) (Like my players did to me today!)

[kalmarjan]

Bah ha ha

"You round the corner and see a dark figure with a black mask, mists swirl around it, while an ominous low tone Huns from the magic sigils adorning it's cloak...

"Yeah, I smack him with my double handled great axe... 20! Roll to confirm... 20! Woo hoo! I deal him... Dang! 32 damage plus 14...46 damage...

"Uhhh....well, he goes down with a wet slurp.... "

[MarianDz]

This "miracle" sometime happened, but can you count on it? I think its luck, and if never happened then we cant get down strong monster quick.

[kalmarjan]

That's the thing. It is luck. Sometimes even when you're holding pocket aces, if everyone at the table is still playing in the point, it's a coin toss. You'll probably lose.

Same goes here. Sometimes, no matter what happens, the players will get lucky. This is where it gets tricky. For me, I let it stand because I didn't want to deny them the awesomeness of rolling double 20s...

But that just meant a story adjustment. Everyone wins. Players and their characters have a badass moment... You get to enjoy that badass moment with them, but you also get to challenge them another way. Win/Win.

[Hurske]

I have seen issues where it seems like the dice rolls in a predictable pattern myself, but I think it's just the way randomness can be, I've seen the same thing in real rolls, and I even work as a dealer in a casino and see some head scratching moments on random numbered games...

[Nickademus]

One thing that my players and I have noticed fairly often (maybe 1-3 times a session) is the die value not matching the dice. For instance, the d20 will roll and end on a 19; the side facing up says 19 and the players say something about a 19. But when the die fades into the chat window and the roll is registered, it will say the roll was an 8. Then the players are like, 'What, that was a 19! What happened!'

[Fudly]

John,

I think the answer is to look at the modifier visibility rather than creating an interrupt solution. In addition to the difficulty of pausing a roll on a remote machine to "fudge", it would slow the game drastically to interrupt every roll to allow fudging. Plus the players know you're fudging if you don't have it on all the time.

Cheers,
JPG

This would be great. Currently, my players roll their attacks and damage, see the result, and then I describe the result. The order wrecks the suspense. It's like telling a joke by leading with the punch line.

I'd much rather it be: my players roll attack and damage, I see the result (including the attack roll, damage, effects, and mob's new status) in a window that lets me adjust cover or fudge it, I describe the result, I click yay or nay to apply the attack (in case it's erroneous), and then my players see their rolls. This would take almost exactly the same amount of time for me since I describe the result either way. Hell, you could make it faster by not forcing me to wait for the physics engine.

[Nickademus]

I don't think that FG knows what the roll is before the physics engine is done (that's kind of the point of the physics).

Is just turning off the option to show results to players not enough? They are going to see their dice roll anyway and know what their bonus is, so they are going to know what their end roll would be. Turning off the results would allow you to say whether their roll was good enough or not without FG saying so.

[TheoGeek]

Talking about lucky rolls, I am taking my son and his friends through a make it up as we go along adventure. I'm letting them tell their own story and providing the environment for it to happen. But there was one encounter that I wanted to go a certain way, and because of lucky die rolls, it went in a completely different direction.

They were in a sort of Oliver Twist-ish story - a bunch of teenagers being used by a criminal to steal from and swindle the people in the town. I was going to try to extend that story and use it as a way to get them up to level 5 or so before they would eventually be compelled to rise up against the oppression and uncover the criminal element in the town and save it.

...weeeeeeeelllll....

They decided far too early to rise up. And the encounter that, at the time, should have easily defeated them and landed them all in captivity, turned out FAR differently. Due to literally 4 critical rolls, and even a 20 on a DC check, the party surprised and killed the ringleader (including blinding him with an arrow in the eye (that DC20), burned down the building they were staging from and fled to the woods where they eventually tracked down that leader's henchmen, killed them (that was a little less one sided), and took over their camp.

Now, those teenagers are starting to build their own criminal underground, and have eyes on rooting out the other criminal elements and becoming THE criminal underground.

The Warlock is even running for mayor on the platform of destroying the criminal underground.

It's actually pretty awesome how a few lucky rolls can completely change the course of the campaign! :)

Oh, and by the way, all of those 20s were the product of them rolling actual, real, physical d20's from different sets of dice, NOT using FGs die roller.

[LordEntrails]

... They are going to see their dice roll anyway and know what their bonus is, so they are going to know what their end roll would be.....You mean just like if you were at a table and they rolled the dice? :)

[Fudly]

I don't think that FG knows what the roll is before the physics engine is done (that's kind of the point of the physics).

Is just turning off the option to show results to players not enough? They are going to see their dice roll anyway and know what their bonus is, so they are going to know what their end roll would be. Turning off the results would allow you to say whether their roll was good enough or not without FG saying so.

Turning off the results is far worse than what I've suggested. The player rolls. The player asks if they hit. I respond. The player rolls damage. It still says if the monster dies. Then I describe the attack. Having 2 response cycles is way slower, and it didn't solve the description order problem.

The physics engine works similarly to a position function. p(t) = -16t^2 + vt + h. Rather than display the die's position at 60 fps intervals, they could fast forward me to the end result. As a DM, I'd love to spend that time thinking of the attack description rather than waste it on fluff lag.

Plus, as you've noted the physics engine doesn't always agree with the die result, so the physics engine could be complete fluff.