Smuggler Archive
Thread: Slicing Probabilities...how to almost gaurantee the right slice.
Vevy wrote:
I'm not saying that you can chalk it up to an exact % of success...but I'm saying that if it is a 50/50 chance and you get a run of one slice, it seems to then try to balance itself out to the 50/50 split in the long run
There is nothing "trying" to balance out, it's just your imagination. The overall number of speed slices compared to number of damage slicesis determined frommultipleindependant events. Each event (speed or damage)does not affect each other in any way. Soin the end, the totalnumber of speed vs damageslices you get are completely random.
You may be lucky enough toget to the point, say after70 slices, where the number of speed vs damage slices are exactly 50/50. But that is stilljust random. At 150 slices, there might be more damage slices now, maybe 60/40. And at 300 slices maybe it goes in the opposite direction, ending up with 30/70. And onwards to infinity, the overall balance could be any percentage. The final percentage you end up with will be different for every newslice you add to the total, from 2slices to infinite slices.
Law of averages is much more acurate in RL situations, than mathmatics. Mathmatics points out probability, LOA predicts trends.
In mathmatics its50% probable to flip 6 coins, and have all land heads up. That is unlikely. At the same time, its is also 50% likely to get 3 heads, and 3 tails. The problem is, math only looks at one flip at a time.
LOA takes into account the qty of coins. With only 1 coin, the odds stay 50%. With more your more likely to see the outcome you want.
The law of averages is what makes vegas money. If they went solely on mathmatics, there would be no green on the roullette wheel.
one of the earlier post had a link to a sight on laws of averages that destroys your theory. Just substitute coinswith wpns, tails with speed slice and heads with dmg slice.
- "I just got 5 tails in a row. My chances of getting heads must be very good now." False. Many probabilistic events are independent of one another, which means the result of one event does not in any way influence the outcome of another. Coin flips are independent events. The coin does not "remember" what it has flipped previously and self-adjust to get an overall average result. The coin is not "due" for a heads. The probability remains .5 for each individual flip. A belief in this fallacy can be devastating for amateur gamblers. The thought that "I have to win soon now, because I've been losing and it has to even out" can encourage a gambler to continue to bet more than they can afford.
Now you can get 20 speed slices in a row and the next one is still a 50/50 chance.
All right I'm going to say it again hopefully in a way that those of you that still believe you can predict the next slice based on any trend prior to the slice in question will understand.
The only way to 'almost' guarantee the right slice is to take a set of weapons and slice them all. Your odds of getting the right slice is directly proportional to the amount of weapons you are slicing. However these odds are prior to any slice being done and are in no way affected by any prior slice. Once you slice a weapon and it is not the slice you want your odds of getting the right slice actually goes down and not up.
Lets take the coin flip concept as an example. Lets say that you are going to flip a coin 5 times and are betting that you will get at least 1 heads. Your odds oflosing are 1:32 before you start flipping coins because that is the odds of them all being tails. If the first flip is tails your odds of losing are now 1:16 because that is the new odds of the rest being tails. After the second your odds are 1:8, after the third 1:4, and after the fourth 1:2. The same would hold true for slices. You may be unlucky in your first 4 slices but the last slice still has the same odds as if you never did the first 4.
Someof you have the belief that your odds are better but like I said the coin does not remember what happened in the prior flips so it in no way affects the current flip of the coin. There is no memory in the coin that tells it that it landed tails the last four times so it should start landing on heads to conform to the law of averages. This is because you are looking at the law of averages in the wrong way. The law of averages only applies to what is the most theoretical outcome and does not force an outcome based on past trials.
What many of you call predictions based on trends are simply superstitions. A fact that was pointed out previously in this thread. Some of you have the belief that a trend means that the law of averages will break that trend and you should expect to get the opposite while others believe that the trend is more lasting and you should expect to get more of the same. Many of you will apply these beliefs in the way you play and call it bad luck when it doesnt work the way you believe it would but the truth is that it is pure luck since the odds do not change and is still 50/50.
Nobody is saying that you can't write a computer program that produces nearly-random numbers. It's just that computers don't really have any idea of random. They use a pre-defined algorithm to come up with something that is, for most intents and purposes, random.
WayneInAustin wrote:
I disagree. Many moons ago I wrote my own random-number generator (in C, before C++ was even out) based on a combination of things (clock, date, plus some other things...seems like I had four variables I used in different ways). I wish I still had the algorithm (probably on a FLOPPY disk somewhere in a drawer). The bottom line is it was a fairly short routine and didnt take me long to write. I tested it thoroughly for randomness and it certainly seemed to work. And if I can write one, ANYONE can write one!
Vevy wrote:
Carnage43 wrote:
I know what you mean Vevy, but it is STILL completely random.
Just because you get 4 speed slices in a row, doesn't make the next slice anymore likely to be anything....it's still a 50/50
I think about it in terms of dice...just because you rolled 3 sixes in a row doesn't make the next die roll any less likely to be a six.
yes it does, the law of avgs...the more times you roll a 6 in a row, the higher the probability the next roll WON'T be a 6
However, taken as a 1 off case, (like slicing has to be), then the odds of rolling that 53rd 6 is1 in 6. Since you rolled that previous 6 (continuing the dice analogy) 20 people on your server may have sliced and thus your next 6 is the next in the law of averages.
Someone getting a string of speed slices is offsetting the string of damage slices that someone else is getting. Over the long term... billions of checks, it will average out to a 50/50.
Imaridril wrote:
The notion that previous dice rolls, coin filps, wheel spins, etc., somehow have an affect on the probability of the current dice roll, coin filp, or wheel spin is sometimes known as "The Monte Carlo Syndrome".
That's also how casino's make money. 
Anyway. From a classical computer, you're not going to get true randomness, The better the randomizer algorithm is, the closer to "true random" you will get, but it'll never be quite there. Now, if you rig up some sort of system that generates random numbers based on, for example, detection of radioactive decay from a lump of material, or sampling of atmospheric noise, that will be random. You could also rig up a Quantum computer and have it generate random numbers.
As for this system...well, first, I disagree that it's random. Well...no. I agree that the stat that you slice is as random as it can be, which isn't all that well done...although with true randomness yes, it is possible to get 15, 20, more of the same result in a row, it's rather uncommon....I shouldn't often have times where I'll slice a run of, say,twenty carbines and get eighteen out of the twenty on damage slices (though I wish I could see this for every weapon I make and slice!) However, I don't think that the percentage of the slice is actually random...I think that that works on a weighted probability curve rather than just "pick a random number between 15% and 35% for weapons or 15% and 45% for armor" or whatever. There's a definite tendency, to me, towards "average" (say 22-28%) slices over more "high end" (29-35%) or "low end" (15-21%) slices. That's easy enough to explain as a simple bell curve.