Synth Percents Mean Nothing

[babyeatinbob]

Seriously. Yesterday i realized i had one try left on my hanzo board in my pm from when i had an unpure striking pm. I have since turned it into a pure ranged pm, but i decided wth and put the materials in to make the hanzo despite the 34% chance of synthesis. 3 hours later i took it out and it was 18% dark. Not great but very unexpected i had no confidence that it would work.

Now here's where the story starts to suck for me. Yesterday i found a deva-zashi board, the first S-rank board i have ever seen, and i put it in my lil bro's pure striking pm. It had an 85% chance of synthesis. Unless you're braindead you know where this is going. I failed that dagger with an 85% chance after making the hanzo with 34% chance. WTF. Ugh i dont think im gonna be able to play this damn game for at least a week im so fucking pissed.

[Sekani]

While I agree that the synth percents given are a little... eh, off, there's no rule that says 85% is a guaranteed success or that 34% is a guaranteed failure.

[Retehi]

Everything is 50/50 in this game.

[Mayu]

It's a Yes or No in my opinion

I'm sorry for the lost ><

[Dhylec]

Well, looks like you learned it the hard way. Indeed, synth trend means nothing. It's your luck that counts, not the fortune photon either.

The sooner people realize this, the less painful it is to lose a synth. I lost a few valuable synths, but hey, I expected that, so no big deal.

[Zorafim]

What it means is that, if you synth 100 of them, you should get 34 of them. You won't, but you should.

[DikkyRay]

85% isn't guaranteed.
Sure its high up so i see that you are complaining, but that doesnt mean it is broken. I know people who have failed with a 95% (given, it wasnt a great item).
Its not as much probability as it is luck.

[McLaughlin]

Please show me where it is stated that 34% = 0% and 85% = 100%.

[huntlyon]

singular examples can't be used to dismiss things that really only can be proven over the long haul.

that's the crazy thing about chance... it's completely independent

[Mayu]

I failed a 98% lol

[Sychosis]

I'll be sure to jam that [B] Shi-Kikami into my Pure Tech PM instead of my Pure Ranged PM.

http://www.pso-world.com/images/phpbb/icons/smiles/icon_rolleyes.gif

[Weeaboolits]

I failed a 90% five times in a row. http://www.pso-world.com/images/phpbb/icons/smiles/icon_frown.gif

[DurakkenX]

since some of you are programmers you should know this but...

it's just a randomize code that says if the number that get is above/below this number then you succeed/fail

100 is just a decent number that is easy for everyone to understand with enough variable....

also those percent chances are not really percents so much as ratios...even though they are the same thing. It just means you have 83 possibilities out of 100 possibilities that the results will be favorable. But tha doesn't mean that you will get 83 successes out of 100 attempts.

If I have roulette board and there are 1trillion spaces on it and out of that 1 trillion spaces if the ball lands on any of 999,999,999,999 spaces I get what I want, however if it lands on the 1 space that is a lose i lose. It could happen that the ball will always land on the one losing space, but you have a 99.9999999999% in having a desirable outcome, it just never happens even though according to probability you should eventually not get that. Even if you lose every time you still have a 99.9999999999% chance at getting it...You just have uber bad luck or someone is cheating...or you just suck ^.^

[Mayu]

OWWW MY head>< Dura

[Weeaboolits]

The percentages they give us seem to have little bearing on the actual success rate, though, we can't all be unlucky.

[darthsaber9x9]

All the percentages say really is that if you tried it an infinite number times, every 83 out would be a success.

My $0.02 anyway.

[McLaughlin]

On 2007-05-20 19:39, DurakkenX wrote:
also those percent chances are not really percents so much as ratios...even though they are the same thing. It just means you have 83 possibilities out of 100 possibilities that the results will be favorable. But tha doesn't mean that you will get 83 successes out of 100 attempts.


The Law of Averages dictates you're wrong. 85% means 85 times out of 100 (or 85% of however many trials you run) will have a favorable outcome.

The same law can be applied to random numbers. If the random number comes up as less than or equal to the chance of success, then you "win". 85 times out of 100, you'll get a number lower than or equal to 85.

Of course, the Law of Chance and the Law of Probabilities play a big role too. Basically, the one time you really need it to work it won't.

It's not 50/50.

Edit: How can you "suck" at roulette?



<font size=-1>[ This Message was edited by: Obsidian_Knight on 2007-05-21 08:14 ]</font>

[Zael]

I don't have the video of it anymore, but I synthesized 64 9* guns at a 49% success rate. Exactly 30 out of the 64 succeeded. If you don't belelive me, I'll be more than glad to make another one.

The best way to really test synth percent accuracy is by using a fairly large sample like I did. Synth percents do mean something. You're just unlucky.

[Hucast-Kireek]

This game sucks, will synth rates be better in AoI?

[DurakkenX]

you can only correctly apply chance and possibility if you know all the factors which effect something...

If you flip a coin...there is 50% chance that it will land on heads, 50% on tails...However you can flip that same coin thousands of times and never land on the one of the sides (and that is a slightly inaccurate percent as it is possible that it lands on it's side as well, but let's not go there) That does not mean the chance to get that other side is not there. It simply means that it hasn't happened.

Furthermore, the probability of getting an outcome can be altered by several factors...For example let us take the coin example again. If I know exactly the amount of force to place on a coin to make it land on my desired side and i am, just this example, 50% dead on with this trick my desired side will come up 75% of the time if all factors are correct, but that's impossible to figure out all the factors as we can not go to the beginning of time and record every last thing that has happened, but if we could we could easily predict the future.

[McLaughlin]

On 2007-05-22 18:49, DurakkenX wrote:
Furthermore, the probability of getting an outcome can be altered by several factors...For example let us take the coin example again. If I know exactly the amount of force to place on a coin to make it land on my desired side and i am, just this example, 50% dead on with this trick my desired side will come up 75% of the time if all factors are correct, but that's impossible to figure out all the factors as we can not go to the beginning of time and record every last thing that has happened, but if we could we could easily predict the future.



That's not probability. That's playing playing with a loaded dice.

[DurakkenX]

loaded dice just weights the opposite side they want to come up...there are still 6 possibilities, it just becomes more likely 1 will come up more than another due to a new factor.

take that wheel game on the price is right...if someone spins the wheel from a full stop and lands on 00...now if the same person uses the same force but pulls the wheel back a short distance, but still applies the same force released from the same point. According to all the factors but one the wheel should land on 00 and the probability that it will come up will be lower than in the first scenario simply because the wheel was pulled back a little.

It's a small factor that would be ignored by most, but it is there.

[Weeaboolits]

But that's not really based on probability, it's based on force exerted. ;]

You could calculate the force required to get it to land wherever you want, but really, who's gonna'?

[DurakkenX]

everything has to do with probability v.v there are just different factors.

[McLaughlin]

On 2007-05-23 06:33, DurakkenX wrote:
everything has to do with probability v.v there are just different factors.



Not if (let's take your coin example) you know the exact force required to make the penny land on one side instead of the other. You're manipulating the situation so that there are no uncontrolled variables, thus removing the probability of failure.

[McLaughlin]

On 2007-05-23 06:15, DurakkenX wrote:
loaded dice just weights the opposite side they want to come up...there are still 6 possibilities, it just becomes more likely 1 will come up more than another due to a new factor.


No, the other sides can't come up because the dice has been loaded so that (being a square) it always tips onto the side opposite the one you want (thus, getting a favorable outcome every time).



take that wheel game on the price is right...if someone spins the wheel from a full stop and lands on 00...now if the same person uses the same force but pulls the wheel back a short distance, but still applies the same force released from the same point. According to all the factors but one the wheel should land on 00 and the probability that it will come up will be lower than in the first scenario simply because the wheel was pulled back a little.


No, there's no chance for the 00 to come up again because we already know how much force must be applied to spin the wheel a full revolution. Moving the start position back, when compounded with using the exact amount of force required to get a complete revolution, provides no possibility for the wheel to land on the 00 again.

And if you're going to take force into consideration, you need to consider the stopping force of the needle hitting every peg on the wheel.

[DurakkenX]

v.v you are either seeing what i am saying and agree but are simply being a dick or you really can't comprehend something simple which well it's pretty strange.

if you have a 6 sided dice and you weight it so that it comes up 6 you are just hoping that you are playing on a flat surface of a certain quality. If you used the same dice on a different surface it would not always result in the same number, nor is it impossible for you to pick up the dice and set it down on a 3. Just because you have changed a factor does not change the possibilities, it changes the ratio of possible outcomes, but never to 0% of all the other possibilities.

as for the second example you are completely missing the concept probably because your being a dick.

[Magician]

I'm still sour after going 1/15 on Muktrand synths with a pure striking pm.

It sucks, but I still march on.

[McLaughlin]

On 2007-05-23 09:01, DurakkenX wrote:
v.v you are either seeing what i am saying and agree but are simply being a dick or you really can't comprehend something simple which well it's pretty strange.

if you have a 6 sided dice and you weight it so that it comes up 6 you are just hoping that you are playing on a flat surface of a certain quality. If you used the same dice on a different surface it would not always result in the same number, nor is it impossible for you to pick up the dice and set it down on a 3. Just because you have changed a factor does not change the possibilities, it changes the ratio of possible outcomes, but never to 0% of all the other possibilities.

as for the second example you are completely missing the concept probably because your being a dick.



I don't think you understand the concept of loading a dice. But in either case, it won't stop rolling (due to it being heavy on one side) until it gets to a flat surface. The quality of the surface (barring it being frictionless) is moot.

Picking up the the dice and setting it on the 3 side is an independent value and has no effect on the trail data or the game being played. Unless you're cheating, which you would be anyway if you were using a loaded dice. However, the scope of your original example was the rolling of the dice.



<font size=-1>[ This Message was edited by: Obsidian_Knight on 2007-05-23 09:10 ]</font>