Loads of doubles

[Romain672]

Ok I think I took your last game as the first.
I will read back tomorrow (i think) when I will be fully awake and will have time.

[euklid314]

I did not read everything in this thread, but just some short remark from my side in response to the "13 out of 104" games.

You have included longer games and shorter games. In a short game the variance is higher, so 5(or more) doubles out of 15 rolls is not that extraordinary (1 in 10).

Still 13 out of 104 is above average (but nothing extraordinary) so if you have fun with it you can continue to increase your statistics. We will see if you will get to 130 "fishy games" out of 1000 or rather 80-100 out of 1000.

And you should define a threashhold for "fishiness", from your current example it seems to be around 28%. I would not be surprised if the mathematical chance of such games lies around 8% (difficult to determine, since many doubles influence the length of the game), you are at 13% at the moment.

Have fun!

I have included... all games starting from 104 ago, and I started to count them as a suggestion of some people like you who replied; as well as the continuation.
I seriously doubt... a game with 5 doubles out of 15 rolls (15th. roll before conceding, i assume), is normally as frequent as once in 10 games.
My threshold for "fishiness" is influenced a bit by those 97-98-99 something Romain gave, for the first games.
I tend to take the games with doubles/rolls similar to those.
I might not proceed correctly since he lied in the previous reply about my number of games, as well as the forgetfulness of his "once every 60 games".
You should also define your "extraordinary"; or you could not, I wouldn't mind not seeing it, because some people in this thread said that they would be ok with some games with 100% doubles in 100 games, so for them "extraordinary in backgammon" does not exist - hence the variation of "fishiness limits" and "extraordinary".

And again, I said I will keep posting similar games, and I also said like 10000000 times, that this fishiness is FROM MY POINT OF VIEW.
So I do not understand why you all jump with same replies over and over again, prolonging the discussion, since it is obvious we won't agree; and I also think each of your "fishiness" limits might be different from one another, as well.

Have fun, as well.

I have no problem that you post your "fishy" games - in the contrary, I am interested in your results. Please accept that others want to interpret your data and might get to different results than you.

If you roll a dice 15 times then the probability that you get 5 or more doubles is 8,977%. Sorry that I rounded that to a 1:10 chance, it is rather 1:11.

But more importantly:
You have surely read the thread "High number of doubles explained!" by dschingis27. After having looked at your statistic page I think that this thread might explain more of your data as I initially thought.

Your games last on average 50 rolls (25.01 rolls by you, approx. the same by your opponent). There are on average 8.44 doubles (4.22 by you, approx the same by your opponents). The numbers 25.01 and 4.22 are on your statistic page, so they are exact, not guesses.

All the games that you gave are (far) below average in the total number of rolls. That is surely no coincidence. As I wrote, the shorter the game the higher the variance.
Lets look at the first 30 rolls in any game. There is a 50% chance that there are more doubles than expected and a 50% chance that there are less doubles than expected. The games with many doubles in the first 30 rolls will end sooner and thus they are quite likely to appear as "fishy" games in your statistics. The games with few doubles in the first 30 rolls will last longer and will not appear as special to you.

Summary: We both (hopefully) agree that on average you did not get more or less doubles than one should expect (over your total backgammon career on this site).
You might still think that you have a very crazy game streak with too many doubles at the moment or you might think that the doubles are strangely distributed (too many doubles in some games, too few doubles in other games). In my opinion both of these fears are not justified by your data.

Have fun!

[Ze Monstah]

[Raal]

1. https://www.sitepoint.com/randomness-php-feel-lucky/
There are some interesting things in this article, and the conclusions, in my opinion.

Conclution of this article is we should use random_int, and bga is following it.

2. I found a possible new solution/improvement: to use another special function for generating random numbers, possibly better than random_int.
https://github.com/emonkak/php-random

This function is based on same algorism of php mt_rand, and proved not safe.
Future results have possibilities to be assumed.

3. Another possible solution/improvement... To replace the dice roll piece of algorithm previously mentioned (*), with:

$rand_val = random_int( 1, 36 );
$dice1_value = intval( $rand_val / 6 );
$dice2_value = ( $rand_val % 6 ) + 1;
if( $dice1_value == 0 )
{
$dice1_value = 1;
}

This algorism is totally broken. (most part of 6 is treat as 1)
If it was fixed, it equals old one as mathmatic because it's just mapping.

[Ze Monstah]

[Romain672]

Ok I think I took your last game as the first.
I will read back tomorrow (i think) when I will be fully awake and will have time.

That post was right. I added a disclaimer.
Sorry for that
I didn't do much calculations because the calculations I would have done would have been totally on my favor, because I missed one part. Let's do it now:

So you did 26 more games (so 104 games) and found those results:

Game 165813080 (https://boardgamearena.com/table?table=165813080): 14/47 doubles: 98.7%
Game 165874452 (https://boardgamearena.com/table?table=165874452): 10/35 doubles: 96.1%
Game 165943746 (https://boardgamearena.com/table?table=165943746): 13/43 doubles: 98.6%

Let me add them to my method of calculation (so you won 50+50+25 points, so 100 points)

So you goes from 248 points to 373. For the whole serie.
I did 2400 simulations. 384 were above 373 points, so 16%.

If I isolate the last serie, so 125 points in 26 games (which is a low sample), I find 350 simulations (/2400) which got more than 125 points, so 14.5%.

These kind of number are low, but not too low. So I will not advance much to analyse them
Those number can either show that you got unlucky, or show there is a small difference in the generation with what you expect.

[Bethh Anne]

I didn't know where to ask this question. I have just played a game of backgammon and the opponent said that I was throwing too many doubles and they were going to complain to BGA and get me banned on the message chat of the game. I took a screenshot of the message as evidence. Is there anything I can do?

[Jest Phulin]

I didn't know where to ask this question. I have just played a game of backgammon and the opponent said that I was throwing too many doubles and they were going to complain to BGA and get me banned on the message chat of the game. I took a screenshot of the message as evidence. Is there anything I can do?

On their player profile, there will be a link of "report this player" (or something similar). Threatening behavior is against the terms of service. Threatening because of "throwing too many doubles" is ludicrous, unless people actually have control over the randomness function of the game.

[Tuut Tuut tis Luut]

I didn't know where to ask this question. I have just played a game of backgammon and the opponent said that I was throwing too many doubles and they were going to complain to BGA and get me banned on the message chat of the game. I took a screenshot of the message as evidence. Is there anything I can do?

You definitely have nothing to worry about re: getting banned or any other penalties, if that's what you're asking? Being lucky is not against any BGA rules

Just a sore loser who doesn't understand how probability works.

[NEO56]

Je suis complètement d'accord depuis quelques jours le nombre de doubles a considérablement augmenté !!! cela enlève vraiment du plaisir au jeu !!!