Two Plus Two Publishing LLC
Two Plus Two Publishing LLC
 

Go Back   Two Plus Two Poker Forums > >

Notices

Backgammon Forum hosted by Bill Robertie. Discussion of anything related to backgammon: strategy, problems, books, clubs, and tournaments.

Reply
 
Thread Tools Display Modes
Old 07-15-2020, 05:33 PM   #26
echidna
stranger
 
Join Date: Mar 2020
Posts: 6
Re: Backgammon Galaxy dice are rigged

Quote:
Originally Posted by Robertie View Post
Excellent point. I'm particularly curiously as to how sites select the 'unlucky' people. Are they chosen at random, or are darker forces at work?
I was selected because I'm 5'1", 240lbs, have the face of a donkey, and a microphallus. I suspect that's why others were chosen too.
echidna is offline   Reply With Quote
Old 07-17-2020, 10:08 AM   #27
Gobbs23
newbie
 
Join Date: Apr 2019
Posts: 36
Re: Backgammon Galaxy dice are rigged

Just to update my previous post:

Remember, we would expect a particular double to occur once every 36 rolls (2.78 percent) and any double to occur once every 6 rolls (16.67 percent). I analyzed all 56,930 rolls (28,483 by me and 28,447 by my opponents) over my last 256 matches. Therefore, we would expect a particular double to occur just over 1,581 times and any double to occur just over 9,488 times. The results are:

11 - 1,613
22 - 1,614
33 - 1,596
44 - 1,559
55 - 1,626
66 - 1,490
Total - 9,498

Overall, over the course of 256 matches (about 1,800 games, give or take) and almost 57,000 rolls, my opponents and I have rolled doubles only 10 more times than the expected value - well within expected variances and deviations and definitely close enough to show any perception that doubles occur too often is incorrect. In fact, together, we rolled significantly fewer double sixes than expected. A quick check showed this is not one-sided, we both rolled fewer double sixes than expected.

So, I think it's safe to say that the dice are not rigged to roll more doubles. It's just people's skewed perceptions.
Gobbs23 is offline   Reply With Quote
Old 09-27-2020, 12:37 PM   #28
TweedleDum
stranger
 
Join Date: Sep 2020
Posts: 2
Re: Backgammon Galaxy dice are rigged

And that's the problem...if BGG dice aren't random, how and why are they skewed? Cui bono?

My regular playing companion, a mathematician who teaches probability, constantly complains that the algorithm is biased against him, and uses as an example the almost predictable and frequent 6-6 or 5-5 rolls when trapped on the bar, and the miraculous saves as perfect dice are rolled in tough situations.

The doubles rolled regularly when I'm trapped on the bar are disturbing (they don't feel random at all) but I can't for the life of me figure out why BGG should alter the randomness, so I maintain it is, in fact, random, and I'm simply more aware of big doubles rolled when i can't use them. And I'm too lazy to test the theory scientifically. I don't keep a random dice generator next to my computer, and I can't see myself recording thousands of rolls.
TweedleDum is offline   Reply With Quote
Old 10-01-2020, 09:54 AM   #29
Puzzled
stranger
 
Join Date: Sep 2020
Posts: 6
Re: Backgammon Galaxy dice are rigged

I'd like to give the frustrated people who instinctively know the dice are rigged (in just about every online BG game) something to think about. I make no assertions about any BG site being rigged. I shall just provide a theory on how it might be done.

When people analyse numbers they usually just look in the obvious places.

Were there the right number of doubles?
Did each of the 36 possible throws come up the expected number of times
and so on

To imagine that a multi-billion dollar gambling company would not have such obvious bases covered is just idiocy.

Here's where I believe you need to conduct your analysis.

You should look at every instance of a SPECIFIC dice throw, say every instance of 5-2 and then look at the NEXT roll that occurred after that. If the dice are true, then you should see an even distribution of all 36 possible dice throws happening for the throws after the 5-2. Including another 5-2.

Most people would not think to check such anomalies.

Here's why this is vital.

The easiest way to rig games is first to use actual random throws. Yes, real random numbers, true numbers that any analysis would show were perfect, all the right amounts of the 36 rolls appearing.

Then BEFORE EVERY GAME actually starts you generate 100 of those totally fair, random dice throws.

Got that? You generate all the dice throws BEFORE the game starts.

So the throws ARE pure random numbers from a genuine RNG.

Now having got those 100 rolls you selectively assign them to whichever player you wish to.

For example there may be in those 100 throws 4 instances of 6-6

You can give those 6-6s to either player A or player B

You could do that according to the Pip count or to some more clever programming that assesses the state of the game.

If Player A is on the bar and can only get back on with a 5 then you can choose from the remaining dice throws whether to give him a throw that has a 5 or one that does not. If there are no such throws left out of the 100 then you just give the next random number.

Designing a system that can pick and choose when to give dice throws to ensure a given player wins a match would be incredibly complex and most developers are simply incapable of doing this. As a result what you find imo is that the places were rigging seems most prevalent are the extremely obvious places that are a doddle to programme. They are:

1. First dice throws in the game (which can give a huge advantage)

2. Allocating doubles and 2-1s when it's just a race

3. Controlling the dice when a player has been knocked off to the bar

4. Controlling the dice when a player has an open blot


So just step back for a moment and consider all this. Imagine that you have 2 people playing a game in front of you and that YOU get to provide the dice throws. You have in front of you 100 dice throws from which you can allocate, no more no less.

Do you think that YOU could swing any BG game by purposely allocating the dice throws from those 100 generated rolls?

I'd say yes, I could do it every time.


So what does this affect?

First please understand that every dice throw is a purely random generated throw using a proper certified RNG. So any analysis of the dice throws will show they are random EXCEPT the specific test I described earlier.

So for example take these 10 dice throws:

1-2, 4-3, 6-1, 2-2, 3-5, 1-5, 4-4, 2-5, 3-1, 6-2

Assuming they are random, what happens if I rearrange them thus:

6-2, 2-1, 3-1, 4-3, 2-5, 6-1, 4-4, 2-2, 5-1, 3-5


It's the same EXACT 10 dice throws just in a different order

Any analysis of the frequency of doubles or specific dice throws will show they are true numbers.

The ONLY anomaly that you might find would be that the number of times any given dice throw is followed by all the other 36 dice throws is not quite right but it would likely take a stonking number of dice throws to detect that.

So, every time there is a 5-2, it should ultimately be followed by dice throws of all 36 throws in equal measure. And that should hold true for every individual dice throw. Every instance of a 6-3 should be followed by all the 36 dice throws in equal measure.

That's where you need to look to see if dice are being rigged because that imo is how it will be being done.


One final point, even if dice throws ARE totally random and are NOT tampered with in any way, that still does not stop the system generating a dice throw before hand and allowing cheating players to see what that next dice throw is going to be so they can adjust their moves accordingly.

I played a game today on a different online BG site and opponent knocked me off and left 2 open blots in his home board in an almost Kamikaze style, but sure enough the dice throw I got prevented me from knocking him off.

In the end, for me personally, no online gaming (involving randomness) is remotely safe and all are open to various forms of cheating. I never put real money into them. Never.
Puzzled is offline   Reply With Quote
Old 10-01-2020, 10:02 AM   #30
Puzzled
stranger
 
Join Date: Sep 2020
Posts: 6
Re: Backgammon Galaxy dice are rigged

Just to save some highlighting some obvious points . . .

Yes I realise that BG games have varying lengths and thus may need more or less than the 100 pre-rolled dice throws. Perhaps any excess left at the end of a game are used as the first of the 100 throws for the next game.
Puzzled is offline   Reply With Quote
Old 10-01-2020, 10:45 AM   #31
Robertie
old hand
 
Join Date: May 2004
Posts: 1,799
Re: Backgammon Galaxy dice are rigged

And what is the motivation for rigging the dice in this (or some other way)? If I were the owner/designer of some new BG site, it seems to me that generating random dice would be the best way to go. What am I missing?
Robertie is offline   Reply With Quote
Old 10-01-2020, 11:43 AM   #32
Puzzled
stranger
 
Join Date: Sep 2020
Posts: 6
Re: Backgammon Galaxy dice are rigged

Quote:
Originally Posted by Robertie View Post
it seems to me that generating random dice would be the best way to go. What am I missing?
Ruthless business sense perhaps?

One does not need to be Sherlock Holmes to realise that gaming and gambling sites would have very little traffic if they were constantly frequented by BG experts (or any other game) who were pawning all the games. You'd end up with a little clique of top players and few others.

Fixing games, however you choose to do it, ensures that a balance is maintained and that the suckers do actually have an even break. Traffic is the key, keeping people coming back over and over and over. That's the most basic business principle. Then you have the instances where unscrupulous people use special software or admin abilities to cheat and thereby fleece lots of people out of their money. That was seen with online poker many years ago.

In essence any online gaming site needs to be programmed and set up like a slot machine, a controlled environment that delivers a form of entertainment that keeps people putting the money in even though they are inevitably losing it.
Puzzled is offline   Reply With Quote
Old 10-03-2020, 08:53 AM   #33
Robertie
old hand
 
Join Date: May 2004
Posts: 1,799
Re: Backgammon Galaxy dice are rigged

I know lots of good players who play regularly on Galaxy. According to you the dice should be rigged against us. Yet none of us have noticed that the dice are in any way unusual or that it's harder for us to win matches than it would be over the board. What are we missing?
Robertie is offline   Reply With Quote
Old 10-11-2020, 12:08 PM   #34
TweedleDum
stranger
 
Join Date: Sep 2020
Posts: 2
Re: Backgammon Galaxy dice are rigged

Robertie, I like your posts and arguments, and for the most part I agree with your arguments (who benefits from rigged dice? How is the algorithm skewed, and why?), but I, too, have played thousands of games in BGG, and there are definitely disturbing problems with their "random dice" implementation.

First, I am not arguing that the dice favor any single player, but that they are programmed to be "dramatic" in a way that is totally different from real life backgammon games. One instance, that I'm sure you've noticed...when on the bar, you throw a truly extraordinary number of 6-6 or double rolls, to the point that my father and I joke about it. BGG games are "wilder" than real-life games, in that rescues and perfect rolls are commonplace, in a way that non-virtual games rarely proceed.

Reassure me, Robertie, that you really have never noticed the doubles thrown with regularity while on the bar, or the fantastic drama of the end games.

Having said this, I truly believe that it's not directed at any single player or players, and moreover, that good players still win with a regularity that is predictable and reflected in their ratings. But it's different, in my experience, from face-to-face backgammon. Am I being clear? The dice aren't random, but they're not skewed. I will continue to play on BGG, but will smile wryly at the fantastic luck I (and others) have.

-- TweedleDum
TweedleDum is offline   Reply With Quote
Old 10-11-2020, 06:32 PM   #35
Alpha Fish
veteran
 
Alpha Fish's Avatar
 
Join Date: Aug 2008
Posts: 3,320
Re: Backgammon Galaxy dice are rigged

yeah I love how every time someone is an expert on rigged dice they are all universally terrible players but somehow know better than world class players who played 100x more matches and seen every rollout under the sun

Dunning Kruger effect in action
Alpha Fish is offline   Reply With Quote
Old 10-12-2020, 02:29 AM   #36
RolldUpTrips
adept
 
Join Date: Oct 2006
Location: Las Vegas
Posts: 1,084
Re: Backgammon Galaxy dice are rigged

Quote:
Originally Posted by Alpha Fish View Post
yeah I love how every time someone is an expert on rigged dice they are all universally terrible players but somehow know better than world class players who played 100x more matches and seen every rollout under the sun

Dunning Kruger effect in action
This comment should be the end of all rig threads.
RolldUpTrips is offline   Reply With Quote
Old 10-12-2020, 02:23 PM   #37
stremba70
newbie
 
Join Date: Aug 2020
Posts: 21
Re: Backgammon Galaxy dice are rigged

Quote:
Originally Posted by Puzzled View Post
I'd like to give the frustrated people who instinctively know the dice are rigged (in just about every online BG game) something to think about. I make no assertions about any BG site being rigged. I shall just provide a theory on how it might be done.

When people analyse numbers they usually just look in the obvious places.

Were there the right number of doubles?
Did each of the 36 possible throws come up the expected number of times
and so on

To imagine that a multi-billion dollar gambling company would not have such obvious bases covered is just idiocy.

Here's where I believe you need to conduct your analysis.

You should look at every instance of a SPECIFIC dice throw, say every instance of 5-2 and then look at the NEXT roll that occurred after that. If the dice are true, then you should see an even distribution of all 36 possible dice throws happening for the throws after the 5-2. Including another 5-2.

Most people would not think to check such anomalies.

Here's why this is vital.

The easiest way to rig games is first to use actual random throws. Yes, real random numbers, true numbers that any analysis would show were perfect, all the right amounts of the 36 rolls appearing.

Then BEFORE EVERY GAME actually starts you generate 100 of those totally fair, random dice throws.

Got that? You generate all the dice throws BEFORE the game starts.

So the throws ARE pure random numbers from a genuine RNG.

Now having got those 100 rolls you selectively assign them to whichever player you wish to.

For example there may be in those 100 throws 4 instances of 6-6

You can give those 6-6s to either player A or player B

You could do that according to the Pip count or to some more clever programming that assesses the state of the game.

If Player A is on the bar and can only get back on with a 5 then you can choose from the remaining dice throws whether to give him a throw that has a 5 or one that does not. If there are no such throws left out of the 100 then you just give the next random number.

Designing a system that can pick and choose when to give dice throws to ensure a given player wins a match would be incredibly complex and most developers are simply incapable of doing this. As a result what you find imo is that the places were rigging seems most prevalent are the extremely obvious places that are a doddle to programme. They are:

1. First dice throws in the game (which can give a huge advantage)

2. Allocating doubles and 2-1s when it's just a race

3. Controlling the dice when a player has been knocked off to the bar

4. Controlling the dice when a player has an open blot


So just step back for a moment and consider all this. Imagine that you have 2 people playing a game in front of you and that YOU get to provide the dice throws. You have in front of you 100 dice throws from which you can allocate, no more no less.

Do you think that YOU could swing any BG game by purposely allocating the dice throws from those 100 generated rolls?

I'd say yes, I could do it every time.


So what does this affect?

First please understand that every dice throw is a purely random generated throw using a proper certified RNG. So any analysis of the dice throws will show they are random EXCEPT the specific test I described earlier.

So for example take these 10 dice throws:

1-2, 4-3, 6-1, 2-2, 3-5, 1-5, 4-4, 2-5, 3-1, 6-2

Assuming they are random, what happens if I rearrange them thus:

6-2, 2-1, 3-1, 4-3, 2-5, 6-1, 4-4, 2-2, 5-1, 3-5


It's the same EXACT 10 dice throws just in a different order

Any analysis of the frequency of doubles or specific dice throws will show they are true numbers.

The ONLY anomaly that you might find would be that the number of times any given dice throw is followed by all the other 36 dice throws is not quite right but it would likely take a stonking number of dice throws to detect that.

So, every time there is a 5-2, it should ultimately be followed by dice throws of all 36 throws in equal measure. And that should hold true for every individual dice throw. Every instance of a 6-3 should be followed by all the 36 dice throws in equal measure.

That's where you need to look to see if dice are being rigged because that imo is how it will be being done.


One final point, even if dice throws ARE totally random and are NOT tampered with in any way, that still does not stop the system generating a dice throw before hand and allowing cheating players to see what that next dice throw is going to be so they can adjust their moves accordingly.

I played a game today on a different online BG site and opponent knocked me off and left 2 open blots in his home board in an almost Kamikaze style, but sure enough the dice throw I got prevented me from knocking him off.

In the end, for me personally, no online gaming (involving randomness) is remotely safe and all are open to various forms of cheating. I never put real money into them. Never.
You do know, right, that the “kamikazi-style” play you discussed might very well be a perfectly correct checker play, depending on the roll and position. Just a simple example, player A wins the opening roll with a 43 and plays (incorrectly) the European split, moving his runners to the opponent’s 5 and 4 points. Player B responds with a 21. Hitting both checkers with this roll is absolutely the correct play (At least that’s what I recall from reading Mr Robertie’s books — please correct me if I’m wrong). The probability of a return hit from the bar is 20/36 or 55.6%. Thus 44.4% of the time, you’re expected to miss both blots. The fact that you missed in your scenario is indicative of nothing other than one mildly unlucky roll.
stremba70 is offline   Reply With Quote
Old 10-14-2020, 01:23 PM   #38
Robot9999
newbie
 
Join Date: May 2017
Posts: 49
Re: Backgammon Galaxy dice are rigged

Quote:
Originally Posted by Robertie View Post
I know lots of good players who play regularly on Galaxy. According to you the dice should be rigged against us. Yet none of us have noticed that the dice are in any way unusual or that it's harder for us to win matches than it would be over the board. What are we missing?
What is your screen name on backgammon galaxy?
I would like to play you if i happen to be on there at the same time.
Robot9999 is offline   Reply With Quote
Old 10-16-2020, 03:51 PM   #39
kruidenbuiltje
journeyman
 
Join Date: Mar 2010
Location: Netherlands
Posts: 265
Re: Backgammon Galaxy dice are rigged

I love this discussion about rigged dice.

A friend of mine also believes he always has unlucky dice.

I think there is something else going on here, it is called the Baader-Meinhof phenomenon.

One funny thing is that I have never met (or read) someone who believes the dice are rigged in their advantage.

A sad thing is that spending energy on this subject keeps players from learning from their mistaken plays.

I play on Gridgammon and on Galaxy, sometimes I win, sometimes I lose, sometimes I play good, sometimes I play bad, but I don't think either has rigged dice.
kruidenbuiltje is offline   Reply With Quote
Old 10-19-2020, 01:07 PM   #40
ah_clem
stranger
 
Join Date: Oct 2020
Posts: 4
Re: Backgammon Galaxy dice are rigged

Quote:
Originally Posted by Robertie View Post
And what is the motivation for rigging the dice in this (or some other way)? If I were the owner/designer of some new BG site, it seems to me that generating random dice would be the best way to go. What am I missing?
Not everyone wants "random dice" i.e. computer generated rolls that accurately model real world dice. When Safe Harbor Games first came on line, there was a problem with the dice algorithm. Michael Petch and Will Womack demonstrated it to the site admins and worked with them to fix it. Problem solved, right?

No. A lot of players preferred the less "bouncy" dice (which rolled doubles at about half the rate of real dice) so SHG created a "casual play" room that used the original problematic less accurate dice algorithm.

Now, I have no opinion about BGG; my default is to assume that the dice are accurate models of real dice with the dice generator independent of the rest of the code, since this is the easiest approach for the developers to use. Why would they work harder than they have to do it wrong?

That said, SHG found that there was a market for "bad dice" and many players prefer the unrealistic dice. No idea what BGG's business model is, but until someone presents statistically valid analysis similar to what Petch and Womack provided, it's just the same old dice whining we've heard since the first internet discussion board was created.
ah_clem is offline   Reply With Quote
Old 10-19-2020, 01:16 PM   #41
ah_clem
stranger
 
Join Date: Oct 2020
Posts: 4
Re: Backgammon Galaxy dice are rigged

Quote:
Originally Posted by kruidenbuiltje View Post

One funny thing is that I have never met (or read) someone who believes the dice are rigged in their advantage.
Back when I used to fly on commercial airplanes (pre-pandemic) I would sometimes amuse myself by playing the backgammon program included in the in-flight entertainment. The bot play was so bad that the only way to make it interesting was to attempt to lose on purpose. Now, I don't have any stats to back this up, but it sure seemed that it gave me "good" dice that effectively thwarted my attempts to lose.

In this scenario, the developers probably knew that most of the audience had no idea how to play well and tailored their product accordingly. The poor checker play was obvious (and I'm sure would be easy to demonstrate). The dice may or may not have been "fair" but my limited perception was that it was biased against the bot.

Apologies for the gratuitous dice whining. (c:
ah_clem is offline   Reply With Quote
Old 10-19-2020, 01:34 PM   #42
ah_clem
stranger
 
Join Date: Oct 2020
Posts: 4
Re: Backgammon Galaxy dice are rigged

Quote:
Originally Posted by Robertie View Post
And what is the motivation for rigging the dice in this (or some other way)? If I were the owner/designer of some new BG site, it seems to me that generating random dice would be the best way to go. What am I missing?
The Story of Mel seems appropriate here.

Excerpt:

After he finished the blackjack program...
he got a Change Request from the sales department.
The program used an elegant (optimized)
random number generator
to shuffle the "cards" and deal from the "deck",
and some of the salesmen felt it was too fair,
since sometimes the customers lost.
They wanted Mel to modify the program
so, at the setting of a sense switch on the console,
they could change the odds and let the customer win.

Mel balked.
He felt this was patently dishonest,
which it was,
and that it impinged on his personal integrity as a programmer,
which it did,
so he refused to do it.
The Head Salesman talked to Mel,
as did the Big Boss and, at the boss's urging,
a few Fellow Programmers.
Mel finally gave in and wrote the code,
but he got the test backwards,
and, when the sense switch was turned on,
the program would cheat, winning every time.
Mel was delighted with this,
claiming his subconscious was uncontrollably ethical,
and adamantly refused to fix it.

https://www.cs.utah.edu/~elb/folklore/mel.html
ah_clem is offline   Reply With Quote

Reply
      

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off


Forum Jump


All times are GMT -4. The time now is 01:54 PM.


Powered by vBulletin®
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.
Copyright © 2008-2020, Two Plus Two Interactive
 
 
Poker Players - Streaming Live Online