Board Game Arena

Nunzio Buonaventura wrote: ↑16 August 2026, 18:55 check this. it happens often https://boardgamearena.com/tableview?table=900267425

If you want to claim something happens often, then you need to identify:

• What the "it" is that you're saying is happening often.
• Data to support that the frequency "it" deviates from expectation by a statistically-significant amount.

One example of something (what?) happening proves absolutely nothing, except that something happened once. I could roll double-6 20 times in a row, it doesn't prove anything about the RNG or coding bias - it's a rare event, not an impossible event. If it kept happening game after game, but only against non-premium players (or some other cohort of opponents) then it would be worth raising.

How often do favorable or “convenient” rolls appear? That is what I was discussing before. My issue is not simply the occurrence of doubles. My issue is how often optimal rolls appear at precisely the right moment. I have the impression there is a bias based on the gap size in experience of the players measured by number of games and by elo difference. Sometimes new players get really insanely lucky. I would like to research more about this and I explore these forums to see how other players experience those outcomes in game like backgammon, cribbage, etc

Look at the last sequence in the aforementioned game. My opponent rolled double six twice in a row, and those rolls directly contributed to his victory. The same thing has happened to me: I have also received exceptionally favorable rolls and won because of them. What I find particularly suspicious is that the apparent density of doubles seems to increase after captures (another aspect to check). Why to do all this? well BGA is not a non profitable group, it should answer the question.

Yes, I would like to gather more concrete evidence about this. What bothers me is that defenders of the quality of BGA games often dismiss these concerns with explanations about luck, cognitive bias, or pareidolia, sometimes in a way that makes anyone raising the issue look ignorant even stupid: “You don’t understand randomness,” “It’s just luck,” “You’re seeing patterns that aren’t there”,...

But poor implementations do exist also. Moreover, many people have played these games extensively outside the platform. Even if they do not have all the mathematical tools required for a rigorous statistical analysis, long experience can produce a reasonably intuition. That intuition can sometimes tell you: “Wait a minute—something feels systematically off here.”

Of course, intuition is not proof. But neither should it be dismissed as irrelevant. It can be the starting point for a proper statistical investigation.

Nunzio Buonaventura wrote: ↑17 August 2026, 14:13 How often do favorable or “convenient” rolls appear? That is what I was discussing before. My issue is not simply the occurrence of doubles. My issue is how often optimal rolls appear at precisely the right moment. I have the impression there is a bias based on the gap size in experience of the players measured by number of games and by elo difference. Sometimes new players get really insanely lucky. I would like to research more about this and I explore these forums to see how other players experience those outcomes in game like backgammon, cribbage, etc

Look at the last sequence in the aforementioned game. My opponent rolled double six twice in a row, and those rolls directly contributed to his victory. The same thing has happened to me: I have also received exceptionally favorable rolls and won because of them. What I find particularly suspicious is that the apparent density of doubles seems to increase after captures (another aspect to check).

Why to do all this? well BGA is not a non profitable group, it should answer the question.

Why do you ask "Why do all this?"? You only have a suspicion and no significant data (you said so yourself). So on that basis - shouldn't your question be "Is this what's been done?". When nothing is done - what response do you really expect? BGA has btw answered the question in the FAQ. In several threads the game-developers have answered it as well. But you can't expect an official response in every sub-forum-thread while an official response is already given in the FAQ.

Yes, I would like to gather more concrete evidence about this. What bothers me is that defenders of the quality of BGA games often dismiss these concerns with explanations about luck, cognitive bias, or pareidolia, sometimes in a way that makes anyone raising the issue look ignorant even stupid: “You don’t understand randomness,” “It’s just luck,” “You’re seeing patterns that aren’t there”,...

The problem is that it happens soooo much that people have a "suspicion" like yours. Because it is very human to have cognitive biases and to see patterns in randomness. And we're all human. The feeling "something is off about the RNG" is too widespread. But the funny thing is - they see all different kinds of things. They see lower elo players being favored. They see higher elo players being favored. They see new players being favored. They see old players being favored. And they are all right to some extent: all groups of players can be observed being favoured for a while. Because it's normal that randomness creates such short-term-patterns (and humans tend to think something is a long-term pattern when in fact for examining randomness these are still very short runs). And then in many cases the bias kicks in and they think they see it even more then they actually did.

Also there were cases were a concern wasn't immediately dismised and turned out there was something to it. Here is a recent example: https://boardgamearena.com/forum/viewtopic.php?t=49561
So why wasn't the first response "you're biased / it's just lucky" or similar? Because OP didn't come with some vague suspicion without even defining what "it" is. They said pretty precisely what the issue is. What you can look at to check it out. So i checked it out and didn't dismiss it and they were actually right (and it probably wasn't an issue with the RNG but my guess is the developer forgot to include some cards in the shuffling).

But poor implementations do exist also. Moreover, many people have played these games extensively outside the platform. Even if they do not have all the mathematical tools required for a rigorous statistical analysis, long experience can produce a reasonably intuition. That intuition can sometimes tell you: “Wait a minute—something feels systematically off here.”

Of course, intuition is not proof. But neither should it be dismissed as irrelevant. It can be the starting point for a proper statistical investigation.

Sad thing is: the ones raising these "issues found by intuition" - they never follow up with collecting data. They only want to raise questions from their feelings but then they want others to do the work for them without even saying exactly what they think is the problem.

Well put, Silene!

Only things to add:

Nunzio Buonaventura wrote: ↑17 August 2026, 14:13 How often do favorable or “convenient” rolls appear? That is what I was discussing before. My issue is not simply the occurrence of doubles. My issue is how often optimal rolls appear at precisely the right moment.

Skilful backgammon players play in a way to maximise their future opportunities - try to increase the number of different roll possibilities that will give them good moves on their next turn, not depending on one or two lucky rolls. To a less skilful player this can seem like biased dice because the experienced player seems to be getting "lucky rolls", not seeing that their opponent is putting themselves in a position where lots of rolls give them good moves. I don't operate at that level of skill, but I've been working on learning to do it more and recognise how my opponents do it.

Nunzio Buonaventura wrote: ↑17 August 2026, 14:13 Look at the last sequence in the aforementioned game. My opponent rolled double six twice in a row, and those rolls directly contributed to his victory. The same thing has happened to me: I have also received exceptionally favorable rolls and won because of them. What I find particularly suspicious is that the apparent density of doubles seems to increase after captures (another aspect to check).

Double six twice in a row is not that odd an event. 1 in 1296 rolls - and if I've already got a double 6, then the probability of the next roll also being double 6 is 1 in 36, not 1 in 1296 or something less likely. Our brains are wired to recognise and remember these patterns, so they feel more common and significant than they are in reality. I can remember multiple times (on another site, not BGA) where I rolled double 6 3 times in a row as soon as I had a piece on the bar - it felt like a pattern, but that doesn't mean that objectively there was a pattern.

Nunzio Buonaventura wrote: ↑17 August 2026, 14:13 Even if they do not have all the mathematical tools required for a rigorous statistical analysis, long experience can produce a reasonably intuition. That intuition can sometimes tell you: “Wait a minute—something feels systematically off here.”

Unfortunately for your argument, science tells us that peoples' intuition about such things is not always reasonable, for reasons Silene has already mentioned. For those that haven't studied statistics and probability (or even those that have!), there are plenty of scenarios where the common intuitive outcome is actually quite unlikely, and the the likely outcome contradicts intuition. And when you've seen literally hundreds of complaints that present intuition as fact and don't look for any objective evidence (or even ignore the objective evidence that contradicts their claims - and rarely search for other threads with the same claim before starting yet another one), it is hard to take each new one seriously - but equally not responding at all is often taken as "evidence" that the claimed problem is true.