Arena 'random' scenario selection?

[jeroenh]

I had to check what the "official" scenarios are according to BGA, and it turns out that is about 65 scenarios.
Like I mentioned, I've played over 500 games on BGA, of which the majority of which are arena games. Still there are some that I've not played and some where I've played more than 10 times at least.

Also, I distinctly remember having seen multiple times that I get the same scenario when I start games quickly after each other that I get the same map.

I'm trying to see whether it makes sense to file a bug report, so I'm trying to gather evidence.

[Jellby]

You also have to ask yourself what the expected results are. What is the probability of never playing some scenario after 500 games? What is the probability of playing some scenario 10 times or more? I bet they're higher than what your gut feeling says. What is the probability of playing the same scenario back to back (this one's easy: 1/65, so in 500 games it should happen around 8 times). And of course, has the scenario selection been random and the number of scenarios constant in previous seasons? Is the random selection supposed to give the same weight to all 65 scenarios?

[Jeronimon]

Also, I distinctly remember having seen multiple times that I get the same scenario when I start games quickly after each other that I get the same map.

But between your two consecutive games there might have started other arena games for other players.
My last two games I started were different and I got Arnhem bridge. Hadn't seen that for a while

[greenjacket124]

Did anyone else experience too that you get the same or similar maps when you start turn-based games quickly after each other?

That is how randomness works. You simply don't understand what random patterns will look like in terms of individual experience. Its very common not to understand this, most people don't, which is why there is a constant stream of conspiracy theories about the site 'rigging' the dice.

While its possible for the programmers to have screwed something up its unlikely they would not simply be using a random variable to select a scenario from a set (and in all likelihood this is a site feature, but hard to confirm without a developer stepping in).

Thank you! At least someone here understands how randomness works.

[jeroenh]

Did anyone else experience too that you get the same or similar maps when you start turn-based games quickly after each other?

That is how randomness works. You simply don't understand what random patterns will look like in terms of individual experience. Its very common not to understand this, most people don't, which is why there is a constant stream of conspiracy theories about the site 'rigging' the dice.

While its possible for the programmers to have screwed something up its unlikely they would not simply be using a random variable to select a scenario from a set (and in all likelihood this is a site feature, but hard to confirm without a developer stepping in).

Thank you! At least someone here understands how randomness works.

For the record, I understand very well how randomness works. I also know that it is really hard to have a good quality entropy source that is able to provide randomness at a very large scale. Very often cloud services are running on virtualised resources, which don't have a direct feed to physical entropy sources. And even with good quality entropy sources, it is really hard to handle those correctly programmatically and make sure that you get good quality randomness at a scale that BGA needs.

This is why I was asking for experiences from other players, and not filing a bug report directly.

Please don't start explaining things to people who did not ask for it. If you feel the need to do that, please ask first.

[jeroenh]

And, there is actually reason to doubt the randomness of BGA. Looking at the code of BGA Memoir (https://github.com/bga-devs/bga-memoir), you can see that the developers are using the PHP function array_rand(), which has a clear warning in the documentation:

Caution This function does not generate cryptographically secure values, and must not be used for cryptographic purposes, or purposes that require returned values to be unguessable.

If cryptographically secure randomness is required, the Random\Randomizer may be used with the Random\Engine\Secure engine. For simple use cases, the random_int() and random_bytes() functions provide a convenient and secure API that is backed by the operating system’s CSPRNG.

https://www.php.net/manual/en/function.array-rand.php

To see what that looks like, consider the following: https://www.random.org/analysis/
The example there is the result of PHP rand() on Windows, on other platforms it is somewhat better, but still not cryptographically secure, or unbiased randomness.