maybe improvement for randomizer necessary (better seed?)

[sesselfussel]

Hi,

I just started the new games in the actual Backgammon tournament in Arena Mode and it is very weird. All new games that I was offered all first dice rolls showed a 4 and a 1. 4 games straight in a row all the same opening with my dices. You can tell me what you will, this is not really random! I had the same experience more often in the past but I realized it but thought that it was just coincidence. But now with all games straight after the other I think that the randomizer for the dice rolls should be improved.

My lectures in programming in computer science were not just yesterday but I can recall that this is not really good. Especially when I had realized something like this a couple of times now ...

[RobertBr]

My lectures in programming in computer science were not just yesterday ...

Don't worry, we can all tell you have not received any recent education in a relevant subject.

The chances of a pair of dice showing the same result in the first roll of four games (assuming you do not care which die shows which) is 1 in 5832. There are 1868 games currently occurring of Backgammon, if that reflects daily activity the behavior you saw will occur to someone every 3 days or so. And if that person does not understand probabilities they will immediately run to the forum to add fuel to the fire of the various conspiracy theories about the perfectly normal random behavior games exhibit on the site (because they all use the same standard function to generate dice rolls).

Please do not do this, it is actively harmful.

[steve5332]

The chances of a pair of dice showing the same result in the first roll of four games (assuming you do not care which die shows which) is 1 in 5832.

Not quite. Given a non-double first roll, the chances of the dice showing the same result the next three rolls is 1 in 5832. The chances of a pair of dice showing the same result in the first roll of four games is 1 in 6828 (rounded), assuming my calculations are correct. Still very much in the same ballpark of course, so your point regarding the frequency still stands.

[dschingis27]

It is a common misconception that you could easily observe as a human player if there is something wrong with the RNG. That's why usually there are extensive validations like these: https://boardgamearena.com/forum/viewto ... 1&start=10

It's also a misconception that you could easily mess up the RNG in the programming code of a game. If you mess sth. up there, it will get super-obvious quite quickly and noticed by many players at once or shown in the stats collected by BGA anyway. It is really hard to mess up the code in a way that the RNG is rigged ever so slightly to just get noticed by a few ones or just get shown in very specific metrics like getting the same starting roll 4 times in a row.

[veggivet]

If anyone believes the dice rolls on this site are random, I have some prime swampland to sell to you

[dschingis27]

If anyone believes the dice rolls on this site are random, I have some prime swampland to sell to you

Lol So what is your theory about the fact that the validation I just linked to worked out pretty well? And also do some research on how good humans are in recognizing randomness. But probably "because science" is not the answer you want to hear.

[veggivet]

I happen to have a science background, but you don't need one to recognize the obvious fact that playing backgammon in real life is VERY different than playing 'backgammon' on this site.

[euklid314]

I also have a science background and I am a (real life and online) tournament player.

For me it is obvious that the RNG is working fine on this site and for me it is clear and obvious that BGA backgammon and real life backgammon (with precision dice) are absolutely the same (in any aspect that is humanly possible to distinguish).

[Romain672]

@veggivet: just one thing, you did 9 games of backgammon during the last 24 hours.
Now compare it with the time for you to do 9 games of backgammon IRL.
If the answer is still low like 7 days, that mean that you will see 7 times mores weird games on bga than IRL.
If your answer is like 2 months, that mean you will see 60 times more weird games on bga than IRL.

2x more is already really high. We are way above this.

[TheViolator]

I played a lot of games over the past few days and I can tell you for sure, the dice roll on this website is freaking disgustingly broken. It's so, so, so, irritatingly moronic that it is definitely due to human interference. There is no way a player can get 6-6 doubles three times in a row if the algorithm is not f....ed in the a.. You definitely dont need a science degree to notice how amazingly idiotic dice rolls are in this game.