Random Character Chance.

[Hylian]

Ok..so well all know that something programmed can never be 100% random right? It has to have an equation of some sort to work the program out. For example a programming genius figued out a way to predict slot machines 3% of the time. With this info he made a ****load of money. I figure the random space on the character selection screen has to be the same way..is it possible for someone to figure out the %'s of it going to a certian character the first time? Is anyone here REALLY good at math or programming? lol. This just sparked my intrest from no where.

 

[Mars-]

Rofl, I don't know why this would be useful, or you would want to know, but I'm sure it can be done.

 

[Hylian]

Random Character Money Matchs

 

[Wobbles]

Nothing you can do will save us from our money match Hylian.

 

[phanna]

I would imagine the character random is as good as any random number generator for games. The only random select of Smash that is goofy is if you turn on levels, they're like the only ones that will be randomed until they're all played. Which is odd because otherwise the game is good about being random, including giving you the same level maybe two, even three times in a row.

But in retrospect, I bet if you unlock a new character, and random character, you'll get that character

Another random exploit I'm aware of, is if you go to training mode and throw Poké balls, then quit out, and play a vs match, the first pokémon to emerge will be the last one you had in training mode.

 

[Doc_Holiday]

When I get home, I will do it 1000 times, record my results, and post them here. Posting from school right now.

 

[Hylian]

Lol phanna.

And Doc thanks I was wanting for someone to do that >_>. And maybe if like 5 different people did it then we could compare results?

 

[Pasqual]

I'm pretty sure that whatever findings came about would be next to inconsequential. With 26 characters, which should theoretically give each of them a 3.846...% of being selected on any given random, it's likely that the "non-randomness of the randomness" so to speak would only produce a fraction of a percent increase or decrease for a character.

But, I am a stickler for testing things out, so I'll create a random number generator for 26 tomorrow. It won't be set for any specific characters, and it'll be in Java which I assume was not the language used for Smash, but it should still give some kind of idea of how nominal the difference would be.

 

[MuseLives]

So i tried gathering some statistical data. I just went until every character had been selected at least 20 times. This could've occured at a minimum of 520 selections, but it took 680. So, here's how it came out:

Note: I only rounded to 3 significant digits, so this wont add up to exactly 100%

Jigglypuff: 5.29%
Peach: 5.15%
Roy: 4.85%
Doc: 4.71%
Falco: 4.56%
Mario: 4.56%
C Falcon: 4.56%
Ganon: 4.26%
Ness: 4.12%
Bowser: 3.97%
Zelda: 3.97%
Y. Link: 3.97%
Marth: 3.97%
G&W: 3.97%
Luigi: 3.82%
DK: 3.68%
Link: 3.68%
Samus: 3.68%
IC: 3.53%
Pikachu: 3.53%
Mewtwo: 3.53%
Kirby: 3.38%
Fox: 3.24%
Yoshi: 3.09%
Pichu: 2.94%


I doubt this will coincide with other lists that people might come up with. However, I would be intrigued to wonder if any of the vs. mode statistics come into play in calculating the random character?

 

[Hal-Yur-Azn-Pal]

I say its just random.

When you press A on random the game is all:

"I want..."

*puts its hand over its eyes*

"THIS ONE!'

Then boom you have your character.

 

[St. Viers]

some things that you should observe, besides precents:

1. How many in a row/ in close proximity

2. Whether turning off/exiting the screen between hitting random affects the results.

 

[Hylian]

So i tried gathering some statistical data. I just went until every character had been selected at least 20 times. This could've occured at a minimum of 520 selections, but it took 680. So, here's how it came out:

Note: I only rounded to 3 significant digits, so this wont add up to exactly 100%

Jigglypuff: 5.29%
Peach: 5.15%
Roy: 4.85%
Doc: 4.71%
Falco: 4.56%
Mario: 4.56%
C Falcon: 4.56%
Ganon: 4.26%
Ness: 4.12%
Bowser: 3.97%
Zelda: 3.97%
Y. Link: 3.97%
Marth: 3.97%
G&W: 3.97%
Luigi: 3.82%
DK: 3.68%
Link: 3.68%
Samus: 3.68%
IC: 3.53%
Pikachu: 3.53%
Mewtwo: 3.53%
Kirby: 3.38%
Fox: 3.24%
Yoshi: 3.09%
Pichu: 2.94%


I doubt this will coincide with other lists that people might come up with. However, I would be intrigued to wonder if any of the vs. mode statistics come into play in calculating the random character?

I main samus/fox lol!

 

[MuseLives]

some things that you should observe, besides precents:

1. How many in a row/ in close proximity

There were 44 back to back same character selections in my data. Also, there were phases where characters got picked a lot, then very little for a long span in a subtle (not predictable mind you) pattern (i.e. Ness got 13 of the first 200 or so, but didn't get another occurance for the next 137 (approximate) picks). I don't know about leaving the char select screen though, mine was all in one go.

Edit: Aside from two back to back occurances, the other 16 of the pichu picks were spread out very sparsely around the 680 overall picks. Aside from Pichu, the other characters seemed to show up in the pattern I brought up with the ness example.

lol, hylian, like I said my data probably wouldn't match others. If it did, that would seriously suck for you >_> I'll do a second run tomorrow afternoon or maybe later tonight and see if the results are similar.