FIBS Board backgammon forum

Quote from: Yvon on June 02, 2008, 02:19:08 PM
Hi this is my first post here. I agree with the previous posters (Zorba et al) that the 1-1 roll was just bad luck, it proves nothing really.

HOWEVER this is not what usually happens at Fibs. What happens is a SERIES of bad rolls for one player combined with either neutral or a series of best rolls for the other. This causes unbelievably high impropabilities that cannot be attributed to random dice. ****

OK in the above example we had the one and only bad roll 11 i.e with a probability 1:36.
Let's suppose the black player then rolls one of his best rolls 12,14,16, 21,23,24,25,32,34,36,41,43,54,56,61,63,65. The chance for this is 17:36 almost 50%.Not bad at all. Then white rolls one of the worst again 55, or 66. This is a chance of 2:36.

We have already reached a propability of this happening of 1/36*17/36*2/36 of 34:46656 or 1:1372
I won't be surprised if black then rolls some neutral rolls and then when ready, a very usefull 66 while white either rolls neutral or more catastrophic doubles.

*** So if this cannot be atrributed to random dice on what the hell can be atrributed? In my oppinion all games at FIBS are monitored by a computer program which at some stage (usually at critical positions) interferes and decides who the winner will be. Don't you dare double in such situations, you will end up crying.

I am going to open up a forum for discussing specifically this issue. I don't expect Fibs to ever admit this  happening, so my only hope is for people to submit their files, the number of games they played, and see if such high improbabilities fall within range.

I personally have played about 1000 games there, i have lost more than a 200 under such bizare conditions.

NB. Has anybody noticed how frequent is your rolling 66 at the BEGINING of the game (say within the first 7 rolls) when they HIT you? Is the chance for this happening a plain 1:36??? No it is not! Look at the conditions (within the first 7 rolls + hit you)

Wellcome Yvon,

I'm very eager to participate in your forum.

BTW, I think more proof is needed to conclude that whether the dices are random or not, 1000 games is not sufficient. There's also a page on Patti's website related to this issue :

http://www.pattib.org/fibs_dice/

But it would be nice to post some odd positions that a lucky roll decides the game, so let me know when you'll create your forum.

Regards

Thanks for the welcome PersianLord. I have sent an e-mail to Patti regarding this matter and although I don’t really expect a reply, just out of courtesy I will wait for a week or so before opening up the forum. Most propably it will be on MSN as is the only place I know of opening up a forum easily.

Thanks for reminding me of that article on whether on the overall the fibs rolls are random or not. I ‘ve read it a long time ago it  just examines whether on the overall the rolls or doubles fall within range of probabilities.For example whether a 45 (or 54) appear 2 times every 36 or 5.5% over a million or billion-I don’t know- of fibs rolls.

This however has nothing to do with my suspicion that the games at Fibs are monitored by a computer which at some stage interferes by giving one of the players a SERIES of best rolls. The article just examines if the dice generator at fibs is reliable or not. I have never claimed it is not, all I claim is that is often bypassed by some computer program. You are very right, this interference often takes place at the bearing off with lots of doubles, all we need is just records to prove it mathematically. Furthermore imo it targets only specific players….

This Fibs computer interference takes many forms but is always a series of 3-5 rolls in critical positions. Suppose the computer wants to destroy an almost quaranteed win of player A. It may start giving  a Series of worst rolls for A + Series of best rolls for B, or a series of worse rolls for A +neutral rolls for B or just a series of best rolls for B+neutral for A. More often than not it spreads this SERIES of let’s say 4 rolls over a bigger range of say  6 rolls and to complicate things even more it may even throw some smoke in, by braking the series and give player A a "sweatener" roll when it’s all too late.

We will analyze all these if and when we open that forum. We will really need good mathematicians to calculate the odds of posted files. Things get terribly messy sometimes considering that for every position there is a set of best rolls, worst rolls, a bit good, a bit bad, and totally neutral ones….

You told me in email that you had proof.  I asked to see it, and also inquired as to your hypothesis, testing method, sample size, and confidence interval.  You dodged those questions.

However, please feel free to post any statistical evidence you would like.  FIBS is honest, and I have nothing to hide.  If you have convincing evidence that there is a problem then I would be happy to see it.  If your evidenced is biased and your collection methods are dodgy, then that too will speak for itself.

By the way, there is no way for a computer program to bypass the FIBS dice roller.  You can use the telnet interface and see for yourself exactly how the client and server communicate-- everything is out in the open and is plainly human-readable.

Quote from: Yvon on June 02, 2008, 08:18:17 PM
opening up the forum. Most propably it will be on MSN as is the only place I know of opening up a forum easily.

http://excoboard.com/

Excoboard - Free, easy and doesn't require any connection with Microsoft, something which gives pause to many folk. You can set up your forum to allow registration without email confirmation which will make it easier to get people to join. You'd have full control over almost every aspect of your forum.

Patti, I've been running out of good rolls recently. Could you give me the account number for Marvin's slush fund again? And the amount .. is it still 10 times the Fibs membership fee for a month of good dice?

Quote from: Yvon on June 02, 2008, 08:18:17 PM
We will analyze all these if and when we open that forum. We will really need good mathematicians to calculate the odds of posted files.

Yes, let's think about this. I roll ten times two dices. Then I get some exact serie of rolls. But wait a minute. What is chances to roll just that exact serie? Something like 1 in billions. And still it happened! I must be cheating.

I just rolled the following sequence of numbers:

4-6 2-1 2-2 2-2 5-1 5-5 5-4 6-1 2-2 4-1

The odds of that are one in 57127475625984, assuming I've counted and done the math right.  Man, this site must be totally rigged.

Don't forget the required reading for all newbies: 

http://www.alef.co.uk/fibs/archive/dysfunction.html



socksey



Make yourself familiar with the angels and behold them frequently in spirit;  for without being seen, they are present with you. - St. Francis De Sales

QuoteMore often than not it spreads this SERIES of let’s say 4 rolls over a bigger range of say  6 rolls and to complicate things even more it may even throw some smoke in, by braking the series and give player A a "sweatener" roll when it’s all too late.

I must admit that I've noticed these "sweeteners" many times, such as needing 6-6 to get a man in off the bar and zoom round the board to even up the race. I do get the 6-6 but only after dancing long enough that it doesn't matter. Time after time similar things have happened to me. I don't know why I've been targeted by the servers at Vog, BrainKing, GoldToken, DailyGammon and FIBS but it's there, plain as day. GnuBg does it to me too. Even my real dice do it to me. I think they're all linked somehow.

Quote from: Patti on June 03, 2008, 12:45:35 AM
You told me in email that you had proof.  I asked to see it, and also inquired as to your hypothesis, testing method, sample size, and confidence interval.  You dodged those questions.

However, please feel free to post any statistical evidence you would like.  FIBS is honest, and I have nothing to hide.  If you have convincing evidence that there is a problem then I would be happy to see it.  If your evidenced is biased and your collection methods are dodgy, then that too will speak for itself.

By the way, there is no way for a computer program to bypass the FIBS dice roller.  You can use the telnet interface and see for yourself exactly how the client and server communicate-- everything is out in the open and is plainly human-readable.

I already told you by e-mail the evidence and the evaluation methods will be posted in a new forum. I already explained it will examine specifically a series of rolls in crystal clear situations. I also told you each poster, will be required to tell his sample size.
Maybe the evidence will not be sufficient to substantiate what I am saying. Maybe it will.
So, patience please....

I also told you by e-mail that this suspicion (either valid or not) could just vanish if you were giving your members the option to play using rolls from a password protected file that your server would send to both players just before the game starts. At the end of the game they get the password and check the rolls themselves... Simple and effective.

Why do you avoiding even mentioning my suggestion? I don't think that would be too difficult for you. I 've read somewhere in this forum you are an expert programmer yourself. Well?

Listen Patti, my purpose is not to undermine the Fibs server.
Instead of trying to attack me, you should have been more logical and asked me "well suppose what you are saying is true, why should Fibs do that, is there anything for it to gain"?

Answer this question to yourself dear Patti.

PS. As I said in my e-mail i will keep on waiting your reply to my suggestion to play matches from a password protected file. It will save us all a lot of time and energy.

....

OK, since you mentioned it... why would FIBS do that?  Where is the incentive?  What does it have to gain?

There are myriad problems with sending rolls ahead of time:

- How long is the game?  How many do you send?
- The server doesn't have email addresses for most users
- Andreas is not currently developing the FIBS software

And the most important is the one that I sent to you in email yesterday.  Allow me to quote:

QuoteIf the server used predetermined rolls, then the complaint
would just change to "my opponent figured out the rolls".
People who are convinced the site is rigged just won't use
logic or reason no matter what you do.

In fact, sending encrypted rolls is actually less secure than generating them on-the-fly from the server.

Oh, and you said this to me in your first message:

QuoteOr I will use my record of games which ALREADY proves everything mathematically and statistically

However, it appears that you are unwilling to share this proof with me, or even describe *how* you created such proof.  I can only conclude from this that you are bluffing.

Quote from: Patti on June 03, 2008, 07:37:42 PM
OK, since you mentioned it... why would FIBS do that?  Where is the incentive?  What does it have to gain?

There are myriad problems with sending rolls ahead of time:

- How long is the game?  How many do you send?
- The server doesn't have email addresses for most users
- Andreas is not currently developing the FIBS software

And the most important is the one that I sent to you in email yesterday.  Allow me to quote:

In fact, sending encrypted rolls is actually less secure than generating them on-the-fly from the server.

Oh, and you said this to me in your first message:

However, it appears that you are unwilling to share this proof with me, or even describe *how* you created such proof.  I can only conclude from this that you are bluffing.

For lots of reasons. Maybe it assumes that some members need  "help" otherwise they will abandon the place. Maybe it assumes that it teaches one of the 2 players (by punishing him) to be less risky, more careful etc. Maybe to add some "spice" to the game. In fact Microsoft admits that their dice generator at zone, accelerates bearing off of already lost games.

OK I 've read your reasons as to why you cannot set the OPTION to your users to play from file. Fair enough, i simply disagree on what the most important reason is. You basically assume that I am illogical and nothing would convince me... well everybody can distinguish from what you said that the most important reason is the owner of the software is not developing it anymore.
Fine, I hope someone else has heard this suggestion and do it at last.

Finally, no I am not bluffing. It just takes a lot of time to prepare a written analysis and present a figure (concerning the odds) which cannot be disputed. So far i was just doing it for my own use, and to tell you the truth i was satisfied with an approximation. I never kept any records, so i have to dig up my files and do everything from the begining so that it can be presented in a public forum. This is the reason I haven't sent you anything yet.

Frankly I want to avoid it. I know how much time that will cost me, and I know how much time I will have to waste presenting and discussing that in a forum.

Btw I almost laughed at you example of 10 rolls and your estimated chance to happen. This is NOT what I am talking about.
See my post to PersianLord regarding his example. When you analyse games it is extremely rare to have 1 and only 1 best roll, or 1 and only 1 worst. Usually there are more, as there is a set of rather good rolls (but not best), rather bad (but not worst) as well as a set of neutral ones.

Quote from: playBunny on June 03, 2008, 06:28:11 AM

http://excoboard.com/

Excoboard - Free, easy and doesn't require any connection with Microsoft, something which gives pause to many folk. You can set up your forum to allow registration without email confirmation which will make it easier to get people to join. You'd have full control over almost every aspect of your forum.

Thanks my friend i will check that out.From a quick look it doesn't take pictures in, and doesn't allow uploading of files does it? 

So the scientific method basically goes like this:

Construct a hypothesis
Design a method for testing that hypothesis
Test the hypothesis
Analyze the data to see if your hypothesis is true or false
Report the results

The trick is that you need to know what it is you're looking for before you start looking.  You also need to gather data in an unbiased manner.

Here's a trivial example.

Hypothesis:  FIBS cheats in favor of the user with the first username alphabetically.

Design:  Write a bot that monitors the output of "toggle report" and logs 10,000 matches.  Analyze the collected data set to determine the frequency with which the lower-username player wins.  If that result is within one standard deviation of expectation, conclude that no cheating takes place.  If it's between one and three standard deviations, record 100,000 matches and retest.  If it's outside of three standard deviations, conclude that such cheating takes place

Test: Write the bot, gather the data. 

Analyze, Report: self-explanatory

Yes, testing an analysis takes a lot of work.  However, if you're already testing the data then presumably you have a hypothesis and have designed the test.  How about sharing that much?

But truly, I have no fear of anything you might publish.  I'm highly confident that no such intentional manipulation is happening on FIBS.  If you design a valid test and prove otherwise, then I would be more than happy to hear it.  If you design a bogus test, or fail to gather data in an unbiased manner, then that will speak for itself.

QuoteFrom a quick look it doesn't take pictures in, and doesn't allow uploading of files does it? 

I'm not sure. I've only had mine for a very short time and it's just for messages between myself and one other person. We hardly stretch the feature list at all and have had no requirement for anything more than text so far. ;-)

Oh, and if you really want to set something up, Yahoo groups probably meets your requirements.

It's perfectly obvious, and scientifically verifiable, that Fibs hates ME far more than anyone else.  I keep losing games that I should win!  Sorry, I just don't have time to post the data.  Trust me! 

C'mon people, if you want a frustrating exercise, play several rounds with gnubg set to grandmaster level, with dice generated by random.org.  Gnubg will kick your keester from here to kingdom come, and you can't blame gnubg, random.org, Fibs, Patti, or anyone else. 

Give it a rest.  The dice aren't rigged -- backgammon is just a frustrating game.  Read the "Dysfunctional" article referenced above, then read it again.  Now THERE'S a conspiracy you can sink your teeth into! 



Bones

Here's the new forum with its first evidence.

http://groups.msn.com/IsFIBSinterferingingames/general.msnw?action=get_message&mview=0&ID_Message=2&all_topics=0

Thanks for the laugh, Yvon.

I won't post on MSN, but tell me this:  What are the odds of encountering a 1-in-5000 sequence in 1000 games?