...

Forum rules
Please DO NOT POST BUGS on this forum. Please report (and vote) bugs on : https://boardgamearena.com/#!bugs
User avatar
Tisaac
Posts: 2736
Joined: 26 August 2014, 21:28

Re: "dubious" luck

Post by Tisaac »

Where are the stats Jim ?
User avatar
CaractacusPots
Posts: 90
Joined: 31 August 2020, 18:08

Re: "dubious" luck

Post by CaractacusPots »

Stats are stuck in game histories which are a ball ache to down load and process.

Give us a simple "export ALL histories to CSV" function please
User avatar
euklid314
Posts: 680
Joined: 06 April 2020, 22:56

Re: "dubious" luck

Post by euklid314 »

CaractacusPots wrote: 25 September 2020, 17:31 More BGA "randomness"

Latest game 4 consecutive dice throws,

6-3, 6-3, 6-3, 6-3,

That's a 1 in 5,832 chance of happening

Of course it dosn't matter because here anything can happen as it would not be random if it didn't right ?!!! lol

So 10 consecutive dice rolls all the same would be just fine ! In BGA world . . . .

It's random Jim, but not as we know it
You should be complaining that you got the 4-fold repetition not earlier: Since you have played 140 games till now you have played approx. 7000 moves on BGA (together with your opponents). So why do you wonder that you have seen a 1 in 5800 event?
User avatar
Jest Phulin
Posts: 1856
Joined: 08 July 2013, 21:50

Re: "dubious" luck

Post by Jest Phulin »

CaractacusPots wrote: 25 September 2020, 17:31 Latest game 4 consecutive dice throws,
That's a 1 in 5,832 chance of happening
As I am writing this, there are 730 active games of backgammon being played.
With an average of 26 rolls per player, and 2 players per game, that means there are over 37,000 rolls happening. So you should be complaining that a 1 in 6,000 event is not happening enough.

CaractacusPots wrote: 25 September 2020, 17:31 It's random Jim, but not as we know it
Jest Phulin wrote: 21 September 2020, 01:47 Showing perfectly well how little you know randomness.

Wait, wait, wait. I just checked my last game (https://boardgamearena.com/gamereview?table=112837100). It started off:
3-6, 1-4, 2-5, 5-1, 5-5, 6-4, 6-4, 3-5, 4-5, 2-1
The odds of getting those exact rolls are 1 in 2,313,662,762,852,352. Maybe there is something going on with the RNG.... (Hopefully people understand sarcasm.)
User avatar
Chauff
Posts: 99
Joined: 27 March 2020, 11:30

Re: "dubious" luck

Post by Chauff »

OK, until now I thought the RNG was perfectly fine, but Jest aroused my curiosity and I did the same exercise on my last games. And what I discovered baffled my mind.

Last game : 6-4, 4-6, 2-6, 6-3, 6-4. OK, that's weird. I think all of you recognized the exact score of Djokovic defeating Dominic Thiem in 2020 Australian Open final. I didn't calculate the odds, millions of games happen on this site and, as Douglas Adams said, an infinite number of monkeys will want to talk to you about their script for Hamlet. But still...

Previous one : 6-3, 6-2, 6-3. COME ON! Now that's insane. As you probably realised, that's the exact score of Novak vs. Rafael Nadal in 2019 AO final. That cannot be.

I looked at the previous one, terrorized, already knowing what I would find. And I did : 2016 Australian Open finals' exact score, vs. Sir Andy Murray. Game, set and match : 6-1, 7-5, 7-6.

I stand my case, guys, the RNG IS BROKEN: the RNG is a Novak Djokovic fan.
User avatar
CaractacusPots
Posts: 90
Joined: 31 August 2020, 18:08

Re: "dubious" luck

Post by CaractacusPots »

Derision is not necessary

The RNG is pants imo. The lack of any simple means to download game histories in bulk is telling. Would take a dev probably 5 mins to code.

Let's have it, then I'll happily analyse the data as we'll see if the RNG is random or not
User avatar
Jest Phulin
Posts: 1856
Joined: 08 July 2013, 21:50

Re: "dubious" luck

Post by Jest Phulin »

CaractacusPots wrote: 26 September 2020, 23:30 The lack of any simple means to download game histories in bulk is telling.
The means are there. We cannot do anything about your laziness. There are companies out there that will write web scrapers for you. Hire one of them to 1) load the backgammon game history to get game numbers, 2) load the replay of a game, 3) scrape the log for die rolls.

You are the one making the claim. Based on English Common Law, you are responsible for proving the claim. It is not up to the defendant to prove their innocence (as Napoleonic Law often requires).
User avatar
CaractacusPots
Posts: 90
Joined: 31 August 2020, 18:08

Re: "dubious" luck

Post by CaractacusPots »

Jest Phulin wrote: 27 September 2020, 00:13 The means are there. We cannot do anything about your laziness.
Oh for goodness sake be real.

Loading up one game at a time, cut-pasting the game text, dumping it into a "program" to parse it and loading the actual numbers into a database. . .

Would take a ridiculous number of hours to achieve for the sample size needed to undertake meaningful analysis.

You need 10s of 1000s of game histories to check the RNG spread

Stop hiding behind excuses. Just provide a simple download option for ALL game histories say between 2 dates, for any player.
User avatar
Jest Phulin
Posts: 1856
Joined: 08 July 2013, 21:50

Re: "dubious" luck

Post by Jest Phulin »

CaractacusPots wrote: 27 September 2020, 00:24
Jest Phulin wrote: 27 September 2020, 00:13 The means are there. We cannot do anything about your laziness.
Oh for goodness sake be real.
I am real. I understand that it is difficult, but it is possible. And, as I stated above, if you are unwilling to do the work yourself, there are companies that can be hired for this purpose. Stop hiding behind excuses. :)
CaractacusPots wrote: 27 September 2020, 00:24 Just provide a simple download option for ALL game histories say between 2 dates, for any player.
This appears to be directed at BGA staff. Please realize that none of the responses so far have been from BGA, but from the community. The closest affiliation anyone in this thread has with BGA is that they are a volunteer developer.
User avatar
Tisaac
Posts: 2736
Joined: 26 August 2014, 21:28

Re: "dubious" luck

Post by Tisaac »

To make thing more precise, even as a community developper, we dont have access to the framework code so we cant add this feature.
That being said, it is still totally doable to do a script that would parse finished games. You could hire any dev to do the job
Post Reply

Return to “Backgammon”