Is there a stop to Internet Backgammon cheaters?

I recorded the first roll of my opponent for 2221 games.  The opponents rolled a 6  861 times or 39% of the time.

The probability of rolling a 6 is 16.7%

The dice are not random or some opponents have control of their roll.

And, yes, there is a way to stop cheaters - don't play them. 

Here's my way of telling a potential cheat (play experts and make them roll first):

1. Their come out roll is a 6 or you have equal rolls and must roll again.
2. Their first free (2 dice) roll is a double or a 65.
3. There is a long pause before their roll.  (They are busy recording moves)
4. There is a long pause after their roll and especially before an obvious move like a 31, 61, 55, etc.
5. They double before their first foll.  (A win is insured for them)
6. When losing severly, they roll 2 life-saving doubles in a row.

The down side is that you will be playing only about 10-20% of the games to the end, win or lose.  It is very difficult to find a non-cheater based on the above.  But, almost guarenteed, if you play them, you will lose.  So why bother.

---

New study results.  This time I counted (using a program I developed) their come out roll of a 6 or equal rolls.

The expected probability is 33% (1/6 + 1/6)

I got 242/344 or 70%

---

One more thing...try playing as a beginner, not an expert.  They tend not to use the cheat programs against beginners.

 

The odds of rolling a 6 is not 6 to 1, it is 3 to 1, ie: 33%. This the the same with any number.

 

Many people mistakingly say that the odds of any particular throw is 36 to 1, wheras it is in fact 21 to 1

You are showing you ignorance.  I dare you to roll one die 1000 time and get around 333 sizes!  If you do, then the die is loaded!  I guaranty that you will get close to 166.

Look again, I said the first roll which is ONE die.

I ran this program in BASIC 3 times and out of 1 million rolls got 166,462, 166,596, 166,445 respectively.

randomize

let tot = 0

for i=1 to 1000000

   if int(6*rnd)+1 = 6 then  ! rolled a 6

      let tot = tot + 1

   end if

next i

print tot  !  will be around 166000

end

Read a book!  Then come back to this forum.

I recorded the first roll of my opponent for 2221 games.  The opponents rolled a 6  861 times or 39% of the time.

The probability of rolling a 6 is 16.7%

The dice are not random or some opponents have control of their roll.

And, yes, there is a way to stop cheaters - don't play them. 

Here's my way of telling a potential cheat (play experts and make them roll first):

1. Their come out roll is a 6 or you have equal rolls and must roll again.
2. Their first free (2 dice) roll is a double or a 65.
3. There is a long pause before their roll.  (They are busy recording moves)
4. There is a long pause after their roll and especially before an obvious move like a 31, 61, 55, etc.
5. They double before their first foll.  (A win is insured for them)
6. When losing severly, they roll 2 life-saving doubles in a row.

The down side is that you will be playing only about 10-20% of the games to the end, win or lose.  It is very difficult to find a non-cheater based on the above.  But, almost guarenteed, if you play them, you will lose.  So why bother.

---

New study results.  This time I counted (using a program I developed) their come out roll of a 6 or equal rolls.

The expected probability is 33% (1/6 + 1/6)

I got 242/344 or 70%

---

One more thing...try playing as a beginner, not an expert.  They tend not to use the cheat programs against beginners.

 

The odds of rolling a 6 is not 6 to 1, it is 3 to 1, ie: 33%. This the the same with any number.

Many people mistakingly say that the odds of any particular throw is 36 to 1, wheras it is in fact 21 to 1

There are not 36 different dice combinations, there are 21. there is a 21-1 chance of any given throw

I do believe it is possible for cheating to occur with MS Internet Backgammon. I have played both the XP and Windows 7 versions. Both versions do have their own "bugs" and some of the problems described by people are nothing but these bugs.

There is, however, a program (called Cheat Engine) that, so far, I have used to find the address of the two dice (die) and view the results of the roll. This indicates that the numbers rolled are generated by the computer of the roller. This is a random number generator that the program uses. Several years ago (back in the DOS days), I wrote a program for generating random numbers. The main problem with RNGs is that they can sometimes "stick" and repeatedly generate the same numbers. The fix is that the RNG is reseeded each time by something (check out http://romhack.wikia.com/wiki/Random_number_generator for one explanation). Once these numbers are generated, they are sent to the game server, which passes the numbers on to the opponent (at least I think this how it works).

The reason that I think this how it works is that with Cheat Engine, I tried to change the results. Once you have the address and the numbers rolled, you can change the number of one of the die. On the board, the move indicator shows the correct move for the either of the numbers. When I move my first piece, then the indicator changes to the number that I changed. It doesn't matter which die I change (I can enter numbers into both of them), whichever number is the remaining move, that indicator changes to the number I entered in. The problem is that when I move the piece, I get kicked out of the game with an error code.

My thought is that others who are much better at hacking than me can easily figure this out. I didn't try to find out this out in order to cheat, but I wanted to find out if it was possible. Perhaps if I did suspect someone of cheating, it might be nice to turn the tables, but then again, they would just quit.

As mentioned before, it is just a @)%*#ing game and for those of you who don't understand this, keep politics out of it. There is really no difference between Republicans or Democrats ... they are all @)%*#ing politicians.

Good Luck