What are the chances of getting 8 daughters in a row?

[Busulon]

Just got the game on steam from ~500 hrs on a pirated version and somehow my 1st character had 5 daughters (1st was attractive so that was great) but then my heir had 3 daughters aswell before having a male heir... Just wanted to know the wacky probability of this.

[Onlineguy]

Most programmers, who write games, use the built-in Pseudo Random number generator on a computer. The word "pseudo" means that it is not a true random number. This will spoil some of the fun of strategy games, where you have unrealistic random number results like what you experienced.

There are ways to get almost "true" random numbers in a game or anything else for that matter, and there has been a lot of research on the subject. There are a number of algorithms that generate more accurate random numbers. But unfortunately very few game programmers bother to try to use anything even resembling a more accurate random number generator in their games.

[Varainger]

I recommend to delete the p-word and just not mention it.

5 daughters in a row is not that unusual.

I've recently had a character whose wifes (they died) i impregnated with help of a cheat mod
and he got something like ten daughters before the first son....who then died on an infection
at the age of ten.

[lefty1117]

is your character's name Henry?

[broartwar]

1/16 I think? I'm terrible at math but 50% to get a girl.... 8 girls... Thats my guess... It's definitely happened to me before.

Someone who's good at math could figure this out. My guess is 6% chance... Though I'm not sure if getting the same outcome changes the percentage

[Onlineguy]

Probability is calculated by multiplying the mathematical chance of something happening against each other, each time.

For example, if it is a 1 in 2 chance of getting something the first time, then you get it twice in a row, it is a 1 in 4 chance of that happening, if it happens 3 times in a row, then its 1 in 8, if it happens 4 times in a row, then its 1 in 16 (so the person's math above is wrong). Each time you multiple the number two against the previous calculation to get the new odds.

5 times becomes 1 in 32, 6 times becomes 1 in 64, 7 times becomes 1 in 128 and then the 8th time becomes a 1 in 256 chance of happening. Which if you divide 1 by 256, it is actually mathematically .00390625 chance of it happening, or you can display it as .390625 % chance of it happening. Yes, that low.

The odds of something happening are based on the number of possible mathematical permutations. By permutations, I mean the number of possible outcomes. For example, in the case of repeating it twice, you have 4 possible combinations of outcomes, 1 girl 1 boy, 1 girl 1 girl, 1 boy 1 boy, 1 boy 1 girl. Thats why it would be a 1 in 4 chance when you repeat it twice. Then if you repeat it 3 times, you have 1 in 8 chances, with 8 possible outcomes.
You calculate the number of possible permutations to determine the odds (1 in X number of permutations). When probability is based on a 50 / 50 chance of something occuring, you are actually dealing with binary numbers when calculating the permutations. So when calculating whether you can have 8 girls in a row with a 50 / 50 chance of it happening, you are looking at an 8 digit binary number when calculating the mathematical permutations which as I said, is 256 permutations of possible outcomes. (An 8 bit binary number is actually what they call a byte, or 2 hexadecimal numbers, with 0 to 255 values). So a 1 in 256 odds of getting 8 girls in a row.

I went to college for both Computer Engineering and Computer Programming, understanding probability and learning how to write computer programs was my major. This is why I said this kind of stuff and having these kind of results, is only because of the fact that programmers use Pseudo random number generators. Otherwise you will almost never see something that has a 1 in 256 odds (.00390625) of happening. Computer Engineering was 2 classes short of a Minor in a Math Degree in the college I attended.

[Halfinger]

The chance of something happening is 100%. The fact the result is something you find odd is not relevant. While humans love to identify patterns the maths does not care one bit they all happen to be female. There will be numerous times this does not happen but no one posts about that do they? In fact given the number of plays this is likely to happen a lot.

You always see something that has 1 in 256 odds as there are 256 possible results. The winner of the lottery does not have special numbers either no matter what they chose.

Also it is nothing to do with psuedo number generation. They work just fine for games and a lot faster than the alternatives.

[Onlineguy]

I disagree about Pseudo Random Numbers working just fine in games. And it isn't that hard to create a pretty random pattern. High quality disk wiping programs, will ask you to type random characters on the keyboard or move the mouse randomly to generate a random seed, when used together with a well written random number algorithm, will generate almost perfect random numbers.

But as far as disagreeing with the Pseudo Random Number generator use in games. In real life, it is possible to have the exact same 8 random numbers occur in a row. But it happens RARELY. Key word, is rarely. Since it is 1 in 256 chances of it happening.

The problem with Pseudo Random Number generators, is this kind of thing, happens very often and far more then the probability of it happening, if they instead used true or nearly true random numbers.

And since I see you edited your post, and mentioned the lottery. There is nothing unrealistic in terms of probability theory with the results of the lottery. Only a fraction of the money spent on lottery tickets goes to the winnings. The money is used by the government for various things. So maybe billions of people buy a lottery ticket, and I seen in the United States (my country), there will be no winners for multiple days in a row, until the prize gets up to over a hundred million dollars (and as I said, only a tiny fraction of the number of people who buy the ticket, goes to the winnings). So the number of people who buy the ticket matches the odds of winning.

[Halfinger]

It is NOT rare at all. Multiply the number of times players have 8 children by 1/256 and you will get a lot more than 100%. You forgot to multiply. In addition, as I said, ALL results are 'rare' which does not make them special. There's no pixie maliciously teasing us humans with patterns we can easily identify.

The odds of someone winning the lottery is high despite the actual result being extremely low.

Stop blaming number generators when maths explains it just fine. The forums are full of 'the rng is broken' with no evidence that is true within the limitations of a game. People have done tests on them numerous times with no indication of bias within the tolerable limits. If you want to do your own test though please feel free.

Of course they are useless if you are running an online casino or doing a scientific experiment but that is not relevant here.

[Onlineguy]

Let me try putting it another way.

Since it is 1 in 256 odds of getting 8 children in a row. Any one person, should have to play on average 256 games, to get 8 children. Now how many people, have played 256 games of this, before getting 8 children?

I know what you are doing, you are going by the number of players. And yes, a game like this, may have 200,000 or more players.

But, you can decide how you want to do the math, and that in no way, changes the accuracy of the math. You can do this on an individual basis. How many people, on average, played 256 games, before getting 8 children?

Now in the case of true random, there will be situations where someone will get it on the very first try, or the exact opposite, they may play over 500 games without getting 8 children. But probability theory is that, and they key word, is "on average" it will be 256 games that each individual has to play to reach that number. It doesn't change the accuracy of the math, if you do this on an individual basis.

You can experiment with probability theory, by just rolling dice. If you use six sided dice, it would be 1 in 216 odds of getting 3 sixes in a row. On average, most people will have to roll these dice around 216 times to get the 3 sixes in a row. Now because it is random, it might occur at 100 rolls for one person or 300 rolls for another, but it will average out to 216 rolls. And it is a set of rolls that I am talking about in this scenario, rolling the dice 3 times in a row, for each try.

And my comments on Pseudo Random Number generators, is based on studying the results for a lot of my life. I seen plenty of Pseudo Random Number generators have trouble generating certain range of numbers, they tend to throw out numbers in one range more often then any other range. For example, I seen some cases, where they tend to throw out high numbers at such a ridiculous percentage compared to low numbers, and I am talking about spending hundreds or thousands of hours playing a game, where this was the pattern for the entire hundreds or even thousands of hours of play time. They don't behave like true random number generators.

Also I been writing computer programs since the 1980s, starting out programming on the Atari 800XL and the Commodore 64. Then I learned object oriented programming in college, like C++ and Java. You often seed a random number generator, and they tend to have trouble throwing out zeros. As I stated in the last paragraph, sometimes they tend "stuck" on the higher number range. If you want the random number generator to give you from 0 to 6. It will almost never give you 0, but give you most often from 2 to 6. And the number 1, a fair amount of times, but not at much as the range 2 to 6. And the number zero will almost never be generated by the random number generator. I actually done studies and wrote programs, that spit out billions of random numbers, and analyzed how many of each number it generated. So I actually have some knowledge of what I speak.

[Halfinger]

Not 8 games. And even if it was that is not relevant. You are mixing a specific result with any result. There is a 100% chance of getting any result and you can make your argument for any of those results. I could say 'I got f,m,ff,mmm' and according to you be as equally upset that the ring was broken. The only reason there was a post at all was because the result was something us humans see as a pattern. The maths does not care one bit.

There have been numerous and extensive tests on psuedo generators. There is only a few used by all games. No bias has been shown at all outside tolerable limits. If you can find them and support it with solid maths and repeatability then please do so, it is always good to spot these things.

Even if there was bias there is no reason at all it would produce results that humans recognise. Confirmation bias is a major issue that proper testing tries to remove.

[Onlineguy]

Yes, you are correct, and nothing I said, nor the math I stated, disputed that it is an EQUAL chance to get any result.

But if you were trying, the key word is "trying", trying to get one specific result, such as the 8 females in a row. Then if you try to get just that one specific result, then you have a 1 in 256 chance of it happening.

Or if you went to a casino, and a specific Roulette wheel had 50 numbers on it (and sorry, I really don't know how many numbers a standard Roulette wheel has on it), and you were trying to get one specific number and win money on that one specific number, then again, you would be "trying" to get that one result, and have 1 in 50 odds of it happening. On average, if you try for that result, it would take 50 tries.

The same with the 8 females, on average it would take 1 in 256 tries to get that specific result. And I never said, that the quote "math" cares about the results, one way or another, so I have no idea why you keep saying something, I never stated. The fact that the math doesn't care, doesn't change whether the math is accurate or not, and nothing I said was not correct. Your interpretation of what I am saying, may be off.

[Halfinger]

Originally posted by Onlineguy:

Yes, you are correct, and nothing I said, nor the math I stated, disputed that it is an EQUAL chance to get any result.

But if you were trying, the key word is "trying", trying to get one specific result, such as the 8 females in a row. Then if you try to get just that one specific result, then you have a 1 in 256 chance of it happening.

Sure. Just we get these sorts of posts so often on numerous forums and they get a bit tedious. Even if games ended up with a perfect number generator they will never go away.

Having said that, as an x-com player I've caught myself blaming 'bad' rng plenty of times when i get myself killed.

[Varainger]

Besides....
YOU DID NOT GET EIGHT FEMALES IN A (single) ROW.

That were two different characters you mentioned. Father and son.
That's two rows.

[Onlineguy]

This is hopefully my last response (because as you said, this is starting to go around in a circle now), but I did enjoy talking with you, to Halfinger (I always enjoy intelligent conversations):

My final response: It is the nature of humans to be stupid and complain about everything. You didn't figure that out yet?

[Halfinger]

Originally posted by Onlineguy:

It is the nature of humans to be stupid and complain about everything. You didn't figure that out yet?

I like complaining about complaining.