Question regarding a bit of a javascript mystery

[Grognard]

Hi all,
I have a curiosity I wonder if someone can divine for me:

In RPG Maker MV, I have an event with a conditional set up using 2 variables.
In separate event I define Var 0001 = 20
Then I have Var 0002 = a random integer 0 - 100.
The conditional is set up so that if Var 0001 > Var 0002, the event fails.

I noticed the event was succeeding a lot more often than 20% of the time, so I tested the event with Var 0001 set as 1, and then again set as 99. (Var 0002 remains random).

It seems that when I had Var 0001 = 1, I had a 99% chance of success, and at 99 I had a 1% chance.
It's not a problem, because I just reversed the sign to get the probability I want, I'm just really curious why "If 20 > Random(0-100)" gives me an 80% chance rather than 20%.

 

[Doktor_Q]

For clarification, your best bet is to take a screenshot of the eventing in question, including the structure of your if statement, because that sounds more like you have something set up oddly.

 

[CaRa_CrAzY]

Are you sure you are comparing "If 20 > Random(0-100)" instead of "If Random(0-100) > 20"?

 

[Grognard]

Yeah, I'm sure. I was kind of just wondering if there was some quick of js that caused it that I didn't know about. I don't use those variables, but there must be something in the code somewhere. Thanks for the responses~

 

[Engr. Adiktuzmiko]

Welcome to randomess and RNG... Unless you're doing an inifitissimal amount of trials, chances/random can work really randomly.

Unless of course its really a case of you actually using the reverse of what you intended to do. So yeah, a screenshot of the events would be good

 

[caethyril]

In separate event I define Var 0001 = 20
Then I have Var 0002 = a random integer 0 - 100.
The conditional is set up so that if Var 0001 > Var 0002, the event fails.

As written, this means a 20/101 chance of failure, i.e. ~80% chance of success.

I.e. when var 2 is 0~19 the condition is met and "the event fails".

 

[Zeriab]

When dealing with random numbers I find it helpful doing a bunch of test runs to see if the statistics were correctly setup: