Strategy to the Casino Event! Don't lose your money!

a % chance of something occurring is a fixed value agree .09% is 9/10000

how you generate the check against that value is random
so if you generate a number from 1-10,000 if it is less than 10 you "win a rancor" if it is between 11 and 5000 you get your coin back if it is between 5001 and 8000 you lose if between 8001 and 9,000 you get a buff between 9000 and 9500 you get 2x tokens between 9500 and 10,000 you win a kingpin toekn

do you see the difference? you will have a 50% chance to win your coin back a .09% chance to win a rancor a 30% chance to lose a 10% chance to get a buff and 10% chance

over 1 million rolls you will on average have something like 9 rancors

Duly knows all and shows all

with .009 is about a 1111 clicks

It reminds me when I was a newbie and did

and wondered why I got streaks of the same randomnumber

Good! Those are the percentages! Now I'm not saying you will win a rancor! Exactly what you said the probability percentage is .09%. I'm talking to the people who have not won anything. It's cheaper to buy 50 smuggler tokens because the average KP tokens you will get based on the experiment I did with guild mate and what I saw on the Reddit post is 7-10. If someone buys the 50k KP token, they will go broke, because the ratio of 50 smugglers = 7-10 KP tokens is greater.

one just gets you broke faster, if you have average luck, it will be at a fairly constant rate. if your luck is higher/lower it i will vary. so if it takes you 1111 clicks on a kingpin that will cost you XXX amount, you can get that amount either directly purchasing it, or trying to win them from the other machine at a 10% rate.

dulfy does the math one just gets you there faster, if you get lucky it coudl be cheaper but "on average" it will be that amount. so if that amount isn't worth it dont' spend it.

Well it's hard to base it off of luck after we just said all those things

its still all luck..hence "random"

person 1: buys 1 token walks to smugglers machine hits a kingpin token. goes to kingpin machine and hits a jackpot dropping a rancor walks away happy.

person 2: buys 10 stacks of 99 smugglers tokens blows them all on a smugglers machine and gets 500 kingpin tokens. goes to kingpin machine and loses every one of them, not even a certificate

random is random just because something has a 5/10 chance of happening doesn't mean you get 5 if you do it 5 times. each event is independent of what happens before it, and what happens after it.

I'm getting the feeling you still don't understand what "random" means.

I don't think you get how statistics work. If an event has 0.1% (1/1000) probability of happening, let's say the drop rate of the rancor, it doesn't mean that 1000 rolls will get you one.

Set percentage doesn't mean set chance of winning after a given amount of rolls.

Yes, but those percentages are exactly what you use to implement your random event.

This is very basic software game design. The stereotypical RNG generates a floating point number between 0 and 1. If you want something to happen 40% of the time, you hit the RNG and check if the number is less than 0.4 (simple math: 40% of the floating point numbers between 0 and 1 are less than 0.4). If you want something to happen 10% of the time, you hit the RNG and compare against 0.1. You can produce fractional outcomes using the same techniques:

Event A : 50% chance : Check for RNG < 0.5
Event B : 30% chance : Check that A didn't happen and RNG < 0.8 (0.5 + 0.3)
Event C : 15% chance : Check that B didn't happen and RNG < 0.95 (0.8 + 0.15)
Event D : 5% chance : Check that C didn't happen

The fact that there are defined percentages don't disprove random events. Quite the contrary: They prove that the results are random because they reveal the parameters used to generate the random distribution.

Chance-based is random

Perhaps you're not thinking of the right definition of "random".

There's no evidence to suggest that, and that sort of event-tracking system is actually a very data-intensive and computationally-heavy system. A system with equivalent outcomes over a suitable sample size is obtained simply by generating random numbers and comparing against a partitioned collection of events.

Your test does not test what you think it does. It would behave exactly as you suggest using a random event model with a partitioned result space. Using the random events would also take virtually no memory-per-user, require a trivial amount of CPU time, and be virtually impossible to exploit. Using the event-tracking quota system you are suggesting (getting a result of one type increases the probability of the other results happening, up to a defined proportion) would require a significant amount of per-player memory and constant re-calculation of RNG settings to yield accurate results. Which one of those sounds more likely?

I'm not gonna argue about statistics, once again. I'm saying it will help y'all to follow what me and my guild mates did in my edited post. It worked, maybe it won't get you the rancor, .09 chance, but it will get you something, it's a cheaper way.