These forums have been archived and are now read-only.

The new forums are live and can be found at https://forums.eveonline.com/

Science & Industry

 
  • Topic is locked indefinitely.
Previous page123Next page
 

44 fails out of 45 attempts at invention

Author
Rhiil
BlackWater Mining Inc.
#21 - 2012-03-11 15:10:49 UTC
Breaker77 wrote:
I would like to introduce you to the little thing known as a random number generator

If you can not understand that then please go to a game that is much more baby friendly.

PS, I used to run well over 1000 invention jobs PER DAY, so I'm well versed in invention. Call it bad luck, ****** game design, your a moron! No matter what you call it, in the end and with thousands to tens of thousands of invention jobs, you will balance out to the "so-called" statistical chances of invention.



your≠you're
FunGu Arsten
B52 Bombers
#22 - 2012-03-11 16:45:27 UTC
so you're trying to invent paladins rite?!
Verocity
8 Virtues
#23 - 2012-03-12 02:11:14 UTC
You know, it's also mathematically possible to have failed all 45 attempts but you did not. Perhaps be happy that one did succeed?
Aggressive Nutmeg
#24 - 2012-03-12 03:38:23 UTC
Most people struggle with standard deviation theory. 44 fails out of 45 attempts with a 27% chance of success would be well within expectations.

Nothing to see here. Move along.

Never make eye contact with someone while eating a banana.

Nikodiemus
Jokulhlaup
#25 - 2012-03-12 05:11:45 UTC
Jesus Christ, math geeking won't help you at all. Imagine for a moment that programmers are not God and are just as fallible. Did you switch stations//corps with your invention? Did you use the same BPC stats with no meta gear or decryptors? Change up your nonsense or stop posting **** on forums that doesn't help anyone.

Variety is the spice of life.
Bugsy VanHalen
Society of lost Souls
#26 - 2012-03-12 17:06:53 UTC  |  Edited by: Bugsy VanHalen
Liselle D'solos wrote:
(The system seems to have lost my original post. If this is a double post, please remove it.)

I've made 45 invention attempts @ 27% chance of success. Out of these, 1 (one) succeeded. Unless my math fails me, the likelihood of only getting one success in 45 tries @ 27% chance, is somewhere in the region of 9.7*10^(-7), or more explicitly, 0.000097%.

Now, CCP tells me this is normal. I say that 0.000097% likelihood is pseudo zero in terms of probability. It is literally 1 in a million.
To put it another way: If I did a million batches of 45 attempts each, only roughly one of these batches should result in only one success, assuming 27% chance of success and perfect randomness. Since no RNG is perfectly random, there is going to be bias in the system, but this is just ridiculous.

So, who's right? Is this "normal" or is the system broken?

(The funny thing is, before this swathe of fails, I had a fairly normal distribution of fails and successes.)


This is normal, although pushing the envelope of bad luck. But as others have said the odds will even out over time.

look on the bright side. you just took 44 failures off the table. Statistically for every run of bad luck you should get a run of good luck to even it out.

So in theory if say it takes a batch of 1000 to consistently hit your 27% your remaining 955 runs will have a higher chance of success as you have 44 less failures in the 1000 run batch to worry about.

statistically at 27% you should get 12 successes in a 45 run batch. maybe you will hit a batch or two that will give you 15-20 successes out of 45 attempts. Just stick with it. in a large enough sample size the odds will always even out.
Vincent Athena
Photosynth
Just let it happen
#27 - 2012-03-12 17:37:13 UTC
I think your math is off. I looked up the binomial distribution in Wikipedia, put the equations into excel, and for a 27% success rate got the chance of one success in 45 tries as being 0.001%.

Given all the people doing invention in eve, I would expect such bad runs to occur now and then to someone.

Know a Frozen fan? Check this out

Frozen fanfiction

Tsubutai
Perkone
Caldari State
#28 - 2012-03-12 19:58:15 UTC
People win lottery jackpots with far higher odds against. You just got unlucky.
Ford Chicago
Ziz Zag Ziggurat
#29 - 2012-03-12 20:24:33 UTC
Bugsy VanHalen wrote:
look on the bright side. you just took 44 failures off the table. Statistically for every run of bad luck you should get a run of good luck to even it out.

No.

Bugsy VanHalen wrote:

So in theory if say it takes a batch of 1000 to consistently hit your 27% your remaining 955 runs will have a higher chance of success as you have 44 less failures in the 1000 run batch to worry about.

No.

Bugsy VanHalen wrote:
in a large enough sample size the odds will always even out.

No, that is just the most likely outcome.
Borun Tal
Brutor Tribe
Minmatar Republic
#30 - 2012-03-12 20:33:02 UTC
Overall, I'm getting around 60%-ish success rate. But of the ten jobs I ran yesterday, only one succeeded. 10%. Welcome to Probability Land.
Ajita al Tchar
Doomheim
#31 - 2012-03-12 21:49:52 UTC
Bugsy VanHalen wrote:
[quote=Liselle D'solos]
look on the bright side. you just took 44 failures off the table. Statistically for every run of bad luck you should get a run of good luck to even it out.


This line of thinking is a major contributor to all the drama that surrounds invention failure. What you're saying is roughly equivalent to a famous example of fallacious usage of stats where a man takes a bomb on an airplane with him because he heard that his chance of being on a flight that has a bomb are 1/10,000 (or whatever) and he wanted to get that out of the way to fly safely for many more years. Or people telling their skydiving friends that ohnoes, they had 500 successful jumps, but don't they know that a parachute will malfunction and not open 0.5% of the time, so they are due for a splat any time now. Random variables don't work like that.

See, you're kiiiiind of right saying what you said, but for the wrong reason. Chances are very good that the unlucky inventor will end up with enough successful inventions to bump the numbers to where they are expected to be. But this won't necessarily be the case, and if it happens it has nothing with them getting that crapass 44/45 failure run. Every single event is independent, it has nothing to do with the past or the future, it doesn't care how many people are running invention jobs, etc. It might as well be the only such event ever, but it still has some probability of flipping this way, or flopping that way. So, thinking that this bad run will be made up for by good runs later is reasonable (I console myself like that after double-digit numbers of failures...),but it's not technically accurate because there's always the possibility that it won't happen ever until you give up on invention for good.
Kaaii
Kaaii-Net Research Labs
#32 - 2012-03-12 22:13:04 UTC


"Invention is for people with poor math skills....."


Smile


Steve Ronuken
Fuzzwork Enterprises
Vote Steve Ronuken for CSM
#33 - 2012-03-12 22:40:04 UTC
Ford Chicago wrote:
Bugsy VanHalen wrote:
in a large enough sample size the odds will always even out.

No, that is just the most likely outcome.


Large enough sample = keep going until it does Lol

The numbers are working out for me, for example. Actually, I'm inventing more than I can manufacture, right now. Just waiting for manufacturing alts to spool up.

Woo! CSM XI!

Fuzzwork Enterprises

Twitter: @fuzzysteve on Twitter

Sexorella hotz
SexyCor
#34 - 2012-03-13 03:41:17 UTC  |  Edited by: Sexorella hotz
The probability I obtain is 0.000012 for one or fewer successes out of 45, so one in ~100,000Big smile, which one can I think reasonably argue must happen to someone at some point. However...

The argument that with enough inventions things will average does not work, anyone who suggests it does does not know how to deal with cumulative probability or binomial distributions, but that is neither here nor there, this issue is worked directly into the OPs argument by using the binomial distribution. Maybe it will help if I say I was a math major in undergrad with continued studies at the graduate levelP (except I avoided stats cause its lame and unpleasant), point is, binomial takes this into account, whereas single event probability does not.

quick example of how binomial works:
Probability of single success: 0.27
Probability that one or fewer of 2 trials will succeed: 0.93 (9 in 10)
Probability that one or fewer of 5 trials will succeed: 0.59 (3 in 5)
" " 10 " " : 0.20 (1 in 5)
" " 50 " " : 0.0000028 (3 in 1,000,000)
" " 100 " " : 0.0000000000008 (~ 1 in a trillion)

See how it adjusts for sample size, so if you could kindly dispose of this line (numbers from http://stattrek.com/online-calculator/binomial.aspx)

I invent plenty, and likewise have very improbable strings of both successes and failures, and I see variation by ship type. I get the impression from this that things aren't governed by a random number generator, but also perhaps a weighting system, i.e. market quota and such to improve price stability or something. Given the volume, I feel this shouldn't be necessary, but I think anyone could comb through their results and find incredibly improbably streams of success/failures, of course the success streaks are rarely of note.
Krixtal Icefluxor
INLAND EMPIRE Galactic
#35 - 2012-03-13 11:28:30 UTC
It's called chance and probability. This is completely normal.

I've had several batches of Hulk Inventions, batches of 25 and 30, etc. utterly fail many times the past 2 years.

I've also had 10 of 10 successes upon occasion.

It all works out in the end statistically.

"He has mounted his hind-legs, and blown crass vapidities through the bowel of his neck."  - Ambrose Bierce on Oscar Wilde's Lecture in San Francisco 1882

Ersteen Hofs
Republic University
Minmatar Republic
#36 - 2012-03-13 12:35:46 UTC  |  Edited by: Ersteen Hofs
Sexorella hotz wrote:
The probability I obtain is 0.000012 for one or fewer successes out of 45, so one in ~100,000Big smile, which one can I think reasonably argue must happen to someone at some point. However...

The argument that with enough inventions things will average does not work

yup, the error people keep making here is thinking that long string of failures somehow affects the following trials. in reality each trial is completely independent.

sure you can say that some large sample (like 1000) will always produce exactly specified known percentage but the reality is, you don't KNOW which sample size will produce that percentage - it can be anything including 1e+3 and 1e+3000 with different probabilities.
Quote:

quick example of how binomial works:
Probability of single success: 0.27
Probability that one or fewer of 2 trials will succeed: 0.93 (9 in 10)
Probability that one or fewer of 5 trials will succeed: 0.59 (3 in 5)

I think you used inverse probability here instead of direct.

if probability of success is 0.27, probability that at least 1 of 2 attempts will succeed is the same as 1 minus probability of both attempts failing which is 1-(1-0.27)**2 = 0.467, 5 attempts is 1-(1-0.27)**5 = 0.793, etc.

Quote:

I invent plenty, and likewise have very improbable strings of both successes and failures, and I see variation by ship type. I get the impression from this that things aren't governed by a random number generator, but also perhaps a weighting system, i.e. market quota and such to improve price stability or something. Given the volume, I feel this shouldn't be necessary, but I think anyone could comb through their results and find incredibly improbably streams of success/failures, of course the success streaks are rarely of note.

keeping track of every invention attempt adds even more mindless clicking to already huge clickfest the invention is now, so I stopped doing that a while ago... but from what I have gathered, the overall chance is exactly as expected. I did not do per item stats but I don't think they are really item dependent. Overall score should be more trustworthy since by definition it has larger sample volume.
Ninyania alCladdyth
McLuvin AstroDynamics
#37 - 2012-03-13 16:04:12 UTC  |  Edited by: Ninyania alCladdyth
Ajita al Tchar wrote:
This line of thinking is a major contributor to all the drama that surrounds invention failure. What you're saying is roughly equivalent to a famous example of fallacious usage of stats where a man takes a bomb on an airplane with him because he heard that his chance of being on a flight that has a bomb are 1/10,000 (or whatever) and he wanted to get that out of the way to fly safely for many more years. Or people telling their skydiving friends that ohnoes, they had 500 successful jumps, but don't they know that a parachute will malfunction and not open 0.5% of the time, so they are due for a splat any time now. Random variables don't work like that.


Reminds me of the guy who read the statistic that 1 in 10 children in the USA are Hispanic, so he and his wife stopped after their ninth child, as neither wanted to learn Spanish.
Cyniac
Twilight Star Rangers
#38 - 2012-03-13 16:37:00 UTC
Bugsy VanHalen wrote:
Statistically for every run of bad luck you should get a run of good luck to even it out.


This is a very misleading statement as future events are independent of past events.

The only thing which is true is that you'll have a 27% chance of success on your next attempt. That's it. You do not get a run of good luck for a run of bad luck - ever. You just get a new result.

What people tend to call good luck and bad luck are simply deviations from the expected result. Thing is statistics tells us that deviations from the expected result is what is most likely to happen, not least. When you increase the number of results the deviation of the aggregated results is expected to diminish, but even over a million results you will not have a distribution which perfectly matches the expected result (though it probably would be pretty close).

Having said all that - ouch 44 failures out of 45 runs hurts.

Kraig2
DTOR Squad
#39 - 2012-03-13 18:20:03 UTC
I'm pretty sure everyone already said it, but...

This is normal, this is called chance.

A chance of 1/10 doesn't mean if you do it 10 times you will get one certain success. No, you have 1/10 chance every time.

What is the likelihood of me getting 2 success streak?

1/10*1/10 = 1/100
Taedrin
Virtues Corporation
#40 - 2012-03-13 19:37:02 UTC
Brock Nelson wrote:
This isn't about probability, it's about statistics. Come back when you've conducted 1000 inventions and still have only 97% failure, then you might just have a legit complaint.


Actually, according to statistics the rule of thumb is that you only need a sample size of about 30.

The OP experience a statistically unlikely event. This will eventually happen if you consider EVE's entire population of players. Sorry dude, you won the anti-lottery.
Previous page123Next page