Novaya_Russia
Smash Cadet
- Joined
- Aug 19, 2007
- Messages
- 34
Let's talk about what makes a pattern "solid" or "plausible".
Simply put, a pattern is plausible if there's a low chance that the pattern you're describing could arise by chance. Therefore, to test if this theory is plausible, we must look at all the different orderings of the games, and look at what percentage could fit our pattern.
To simplify maters, I'll just look at the ten original franchises. Later, I might come back around and look at how they're segregated by Smash Bros. entry (SSB to SSBM to SSBB).
How many ways are there to order ten things? Well, in the first slot, you could put any of the ten items. In the second, you could put one of the nine items not chosen for the first slot, and so on and so forth. Therefore, there are 10*9*8*7*6*5*4*3*2*1 = 10! = 3,628,800 different orderings.
How many of these orderings could fit our pattern? Well, lets assume that if we change from the series name to a character name, we change the ordering of the games. This is not strictly true - Mother and Ness are interchangeable without changing the ordering of the games - but that does mean whatever percentage we get from our analysis will be greater than the true percentage.
There are two ways of naming each franchise - by franchise name and character name. This means there are 2*2*2*2*2*2*2*2*2*2 = 2 ^ 10 = 1,024 orderings that could fit our pattern.
That means the percentage of orderings that fit our pattern is 1024/3628800 = .02822%, or roughhly three in every ten thousand orderings.
Sure, there's no rhyme or reason as to why they switch haphazardly between the two, and intuitively, you're thinking that we arbitrarily extended our pattern to the point where its statistically invalid. But look at that number. .02822%. Even less than that, as I noted above. Tell me how you get around the impossibly low odds of this pattern emerging randomly.
Simply put, a pattern is plausible if there's a low chance that the pattern you're describing could arise by chance. Therefore, to test if this theory is plausible, we must look at all the different orderings of the games, and look at what percentage could fit our pattern.
To simplify maters, I'll just look at the ten original franchises. Later, I might come back around and look at how they're segregated by Smash Bros. entry (SSB to SSBM to SSBB).
How many ways are there to order ten things? Well, in the first slot, you could put any of the ten items. In the second, you could put one of the nine items not chosen for the first slot, and so on and so forth. Therefore, there are 10*9*8*7*6*5*4*3*2*1 = 10! = 3,628,800 different orderings.
How many of these orderings could fit our pattern? Well, lets assume that if we change from the series name to a character name, we change the ordering of the games. This is not strictly true - Mother and Ness are interchangeable without changing the ordering of the games - but that does mean whatever percentage we get from our analysis will be greater than the true percentage.
There are two ways of naming each franchise - by franchise name and character name. This means there are 2*2*2*2*2*2*2*2*2*2 = 2 ^ 10 = 1,024 orderings that could fit our pattern.
That means the percentage of orderings that fit our pattern is 1024/3628800 = .02822%, or roughhly three in every ten thousand orderings.
Sure, there's no rhyme or reason as to why they switch haphazardly between the two, and intuitively, you're thinking that we arbitrarily extended our pattern to the point where its statistically invalid. But look at that number. .02822%. Even less than that, as I noted above. Tell me how you get around the impossibly low odds of this pattern emerging randomly.
opo (Aisu Kuraima) (Ice climbers)
