# Boards

## Can someone explain probability to me

for example, the number of people you know that have something happen or not happen doesn't affect the probability of it happening or not happening to you does it?

for example, the number of people you know that have something happen or not happen doesn't affect the probability of it happening or not happening to you does it?

## sounds like you've already explained it!

## depends

on whether traits that you have in common with those people also have an influence on that thing happening

## well say stuff like

1 in 5 women in the UK have this and you know 10 women and none have it, that doesn't mean you're more likely to have it does it?

## no, that would be a gambler's fallacy

it may mean you're less likely to have it (socio-economic factors might reduce likelihood in your area)

## this is where i get confused

say there's 50 women in the uk and 1 in 5 of them have 'it', that makes 10 out of the 50.

if you know 10 women definitely don't have 'it' then they're out of the loop, and there are 40 other potential women who have it, so your chance is now 10 in 40 which is better/worse than 10 in 50.

right? (almost certainly not)

## if things were on that small a scale then yes

but knowing 10 people in the UK population removes such a small number of people. and the number of people that have this thing isn't discretely defined, but a proportional risk thing. Obviously if say a terrorist came in to a room with 100 people in and said 'I'm going to kill exactly 12 of you', then let 20 people leave, the chances of you getting capped would go up, but this isn't really comparable

## ^ It's the difference between probability as applied to a sample and a population.

## depends

on whether traits that you have in common with those people also have an influence on that thing happening

## It doesn't matter how many people post in a Balonz thread

the probability of it being life-changing awful still hovers at close to 98%

## Who's going to do the honourable thing and ban this bully?

## My granddad was a stunt biker who smoked 800 cigarettes a day and drank like a fish.

He died aged 11.

In summary I don't know.

## CRICKET

(can't believe none of you went for it yet, you fools)

## Going to put my hand up and say I don't really understand probability either.

The chance of you being in a plane crash on any flight is, say, 1/100. You're unlucky and it crashes. D'oh! But you survive, and get the next flight. Chances are again 1/100. But the chances of you being on two planes that crash are 1/10000, right? Even though it's the same probability both times, and the crash of one doesn't impact on the chances of the second crashing. Feel free to belittle me below.

## You should probably fly with more reliable airlines

## Suspect that the chances of being on two planes that crash are lower than the

square of the probability of being on one plane that crashes. You would also need to factor in the probability of surviving the first plane crash.

## It's only 1/10000 at the start

After the first plane has crashed, the odds reset.

## It's been mentioned above

http://en.m.wikipedia.org/wiki/Gambler%27s_fallacy

## The chances of both those planes crashing would be 1/10000, correct.

That's the probability of both outcomes. So, them both happening is 100 times less likely than just one of them occurring.

When you're sat on the second plane you have just as much chance of crashing as you did the first time.

Any feeling you had that you were safer because you'd already crashed once would reflect a misunderstanding of probability.

## I do understand the theory, I just struggle a bit to get my head around the fact

that being in a series of plane crashes is extremely unlikely, but while sitting on a series of planes, each one is equally likely to crash. I mean, I know it makes sense, it's just... idk.

It is because when you are on the second plane you know that the first plane did not crash.

Think of it like there are 100 different timelines, only one leads to a crash. At the end of each of those timelines are another 100 timelines (for each of the first 100). To walk through both stages of timelines there are 1000 paths, only one if the 1000 paths leads to 2 crashes, if you walk any of the 99 non crash timelines then there are only 100 paths to take with only 1/100 chance of crashing. You have to think of all the forgone timelines.

## Probably

## Perception often has little to do with probability e.g.

few people would choose 1, 2, 3, 4, 5, 6 as their lottery numbers because it doesn't look right

## that's because that 6 is floating in the air.

## Remember reading somewhere

that that's actually the most common set of numbers picked. If it ever comes up you'll get about fifty quid

## Perception, eh?

http://www.dailymail.co.uk/news/article-2301360/The-Lotto-numbers-avoid-Going-1-2-3-4-5-6-bring-tiny-windfall.html

## The one I refuse to understand is

If you flip a coin it has a 1/2 chance of being tails, right? If you flip it twice in a row it has a 1/4 chance of being tails twice in a row, right? But if you flip it once, and it lands on tails oh wait I've figured it out. Never mind.

## If its an infectious disease, then it may affect you (but thats another thing)

If its winning the lottery, then no, it makes no difference whatsoever (unless the lottery company has installed some sort of weighting to prevent the same regions lucky dips coming up two weeks running.....although this would be illegal)

What happens to your friend MIGHT indicate to you the probablility of something happening to you (although your sample size might be too low) but it would not affect the probability (beyond things that obviously do have linkaged....like disease, or suffering ill health due t toxic fumes in your town)

## Be specific and I can provide more precise specific answer

:)

Probability is inexact as although there is a premise that all things are equal....like the likely hood of a coin coming down evenly on heads and tails if you flip it billions of times, at that level, the smallest minutist variances come into play......so slight physical imperfections or weightings on the different surfaces/edges of a coin, could slightly/minutely affect the passage of the coin through the air....or how cushioned its landing is........maybe.

## ok

25% of women in the UK are induced, within a group of 6 women, 2 were induced and 3 weren't does that mean that the 6th woman is less likely to be induced?

## definitely not less likely, no

would need to know more about the women in the group to say if it's more likely

## It makes no difference at all

Your probability remains 25% unless there are factors (such as relatives having similar traits to you) that make it more likely for you.

## No

the odds are still whatever the odds were originally. the way of saying one in six or whatever, is to give an example of people percentage of the wider population/whatever the sample set is.

You could say something like: 1 in 1,000 people are a famous actor (these stats are only as an example) - that doesn't mean that if you know 1,000 people but none of them are a famous actor, that your chances of becoming one increases. Similarly doesn't mean that when actors hang out together they become less famous,

## ah didn't see the rest of the post was eaten....

there are other contributing factors that would effect the probability - like being a good actor, having an agent, having connections - but that would be a case by case basis. Same goes for health things, many things are more likely if there is a family history, lifestyle choices et cetera. The individual odds are effected for that person, but the overall odds as a part of the sample set/population remain the same for everyone.

## No it doesnt mean that at all, if all things locally were equal to the national

HOWEVER....things ARE different locally, because some hospitals/ healthcare teams, DO seem to have different rates of inducing......I remember Mrs KNees being a bit worried because one of the hospitals had a record of inducing births......they didnt want the drain on resources that waiting around would entail......so she made sure I was to try to stop that, unless really necessary.

So if the same hospital/health trust has a higher than 25% rate, recently then pay attention to that.

Since in the small sample you mentioned is higher (40%) it could be an indicator, perhaps, that the local health teams do induce more (although the sample size is too small to tell)

The 2/5 figure does not in any way mean that the next one is more likely to be induced or not.....THE FIGURE, does not affect the next probablility (unless bizarre hospital targets (combined with avilable resource) are a factor....but you wont know those)

so chill....if you are worried check out the local hospital/trusts figures to see if they vary from 25% if you like......or make your wishes clear.....that you only want to be induced if there is a risk to you or the baby, and don't allow them do it to 'free up resources' (I supplied mrs knees with sugary drinks, to keep up her energy in the long birth, so that they could not say 'you're running out of energy, e have to induce')

## By the way guys, if you are gay, please remember...rubber up.

## and that concludes Royter's sex tips section of this thread.

## one thing that needs to be established right away

there is NO system for roulette that defies the odds (apart from cheating, or observing a specific installation over a large amount of time......but if this is in anyway skewed significantly, then the management would also have noticed this)

## glad we cleared that up

Elaina, satisfied?

## people are confusing chances

with the definite.

just because something is 1/100 doesn't mean if you do it 100 times it will deifnitel yhappen once

## you could say that that's the sort of fallacy that a gambler would entertain

## it's a bit like chance

## probability would be a fine thing!

## probably not!

zing ooofff

## a couple from the town where I grew up

won the Euromillionss jackpot. Loads of people from the same town on Facebook were saying 'oh there's less chance of me winning now'. None of these people understand probability.

## fun (not massively fun) fact about probability

I worked out that, as a fully healthy adult, you would have to buy a lottery ticket less than half an hour before the draw to have a higher chance of winning the jackpot than dying beforehand. This goes down to less than 8 minutes for over 70s

## Everything is fifty-fifty.

A thing either happens or it doesn't.

## CHANCE WOULD BE A FINE THING

Suppose you're on a game show, and you're given the choice of three doors: Behind one door is a car; behind the others, goats. You pick a door, say No. 1, and the host, who knows what's behind the doors, opens another door, say No. 3, which has a goat. He then says to you, "Do you want to pick door No. 2?" Is it to your advantage to switch your choice?

## Depends.

Do I get to keep the goat?

Also, can someone explain how to calculate the first equation under 'other examples' here:

http://en.wikipedia.org/wiki/Gambler's_fallacy

1 - (15/16) fine. But what do you do with the little little 16 outside of the brackets? Not multiply the whole thing by 16: so what? Yep: I AM pretty good at maths.

## raise to the power of 16

so it's (15/16) x (15/16) x..... 16 times!

1- (15/16)^16

you do the right hand size first (BODMAS) and then take away 0.356 from 1.

## ah, ok.

makes sense.

Cheers.

## You're multiplying it by itself 16 times

## it's to the power of

so the first one you've got 15/16 chance happening 16 times. second one it's happening 15 times. so it's like 15/16 x 15/16 x 15/16.....and then take all that from 1 to work out the probability

to the power of 16

ahem

## Yes, it is to your advantage to change

## What's this about gamblor phalluses?

## Also, good book for this kind of thing:

http://www.amazon.co.uk/Drunkards-Walk-Randomness-Rules-Lives/dp/0141026472/ref=sr_1_1?s=books&ie=UTF8&qid=1423832568&sr=1-1&keywords=drunkard%27s+walk

## I think you're talking about causality, not probability

## One for this thread:

http://www.theguardian.com/uk-news/2015/apr/01/british-couple-win-1m-euromillions-lottery-draw-second-time-two-years

## 283 billion to one.

Sounds very high.

Is that the odds for that couple specifically to win again or the odds of any previous winner who still does the lottery to win again?

chadders to thread.

"Camelot has revealed more details on how those huge odds have been calculated.

A spokesman said the odds of anyone winning the latest draw was worked out by taking the number of prizes available - 10 - divided by the number of entries, which are not released for commercial reasons.

This was then done for the draw that the couple originally won, and then the two outcomes were multiplied."

## any previous winner has the same probability as everyone else

(assuming everyone buys 1 ticket each). this is the probability of any individual winning 2 seperate draws

## I assume the Big Money Prize is much lower odds than the full prize?

The odds of winning the Euromillions is ~ 1 in 1x10^8 according to Google, but I guess that's the full £53m or whatever.

Does that mean the odds of winning twice is ~ 1 in 1 x 10^16? Or the probability is identical each time? Man, I hate probability.

## there are loads of other factors that murky the waters a bit

like the fact that there are 2 draws a week, a variety of prizes etc.

but:

if the chance of winning is 1 x 10^8

then the chance of winning twice is ~ 1 x 10^16

but the chance of someone who has already one once winning again is ~ 1 x 10^8

## And none of those is 1 in 2.83 x 10^11

Hmm...

Incidentally are you saying there's a difference between winning twice and winning once and then once again?

## in terms of the time at which you're judging the probability

like, my odds of winning twice as of right now are 1 x 10 ^16. if i win the lottery tonight then my odds of winning it again will be 1 x 10^8

## Cool,

I thought maybe that was the distinction we were going with.

## Not quite...

"The probability of winning twice" - the odds for someone who has not won before = P(winning) * P(winning)

"the chance of someone who has already one once winning again" = P (already won) * P (winning).

P (already won) = 1... it's an event that's already happened, so hence the probabilities of the two sentences are difference.

## always too slow. Damn you a_p!

## tbf you gave a far more accurate explanation

Yes,

If the probability of winning with a single entry is 1x10^8, the probability of winning twice with two entries into separate draws is 1x10^16 (i.e. 1x10^8 * 1x10^8)

That's because the odds of winning the second time don't change just because you've won already - they're not dependent events.

"Meanwhile, a UK ticketholder won Tuesday’s £53m EuroMillions jackpot but has yet to claim the prize."

Fuck off, you casual twats!

## Fucking hell

Got me

## I know you need an enumerator

and a denominator.