EPTFrom the 100 games I've launched:
72 close positions
11 close air
17 cross
To me that's pretty conclusive. Does anyone know why there may be a bias towards spawning in close positions?
User was banned for this post.
Metalopolis prone to close positions, why?
| Forum Index > SC2 General |
I apologize to everyone in this thread for taking the OP seriously. My mod senses are definitely off today. -- Chill | ||
|
the p00n
Netherlands615 Posts
From the 100 games I've launched: 72 close positions 11 close air 17 cross To me that's pretty conclusive. Does anyone know why there may be a bias towards spawning in close positions? User was banned for this post. | ||
|
ch33psh33p
7650 Posts
| ||
|
Lonyo
United Kingdom3884 Posts
| ||
|
skirmisheR
Sweden451 Posts
 Or we could count every time we play metalopolis/LT how many close/far positions we spawn on | ||
|
nkr
Sweden5451 Posts
 | ||
|
Moonstruck
20 Posts
try a 1000 to 10k | ||
|
Toast.yum
51 Posts
| ||
|
fabiano
Brazil4644 Posts
No one is ever going to manually launch 1000 to 10000 replays -_- | ||
|
Playguuu
United States926 Posts
Ninja'd^^ | ||
|
magha
Netherlands427 Posts
On February 16 2011 04:50 Toast.yum wrote: nkr: Yes it is Taking sample sizes of 100 would prove that nobody in history has ever won a lottery. | ||
|
Dimagus
United States1004 Posts
- The random number generator is not as random as Blizzard thought - The algorthim to generate starting locations is flawed or it could be both. | ||
|
DamnCats
United States1472 Posts
But no that is quite odd indeed. | ||
|
the p00n
Netherlands615 Posts
| ||
|
TigerKarl
1757 Posts
| ||
NonY
8751 Posts
On February 16 2011 04:52 magha wrote: Taking sample sizes of 100 would prove that nobody in history has ever won a lottery. I hope you are agreeing with Toast.yum and making the additional point that an appropriate sample size is specific to the situation. | ||
|
emythrel
United Kingdom2599 Posts
On February 16 2011 04:48 skirmisheR wrote: I think 100 games is enough if you get that extreme results from it, but really 72? That's sick, could someone else do the same thing? Or we could count every time we play metalopolis/LT how many close/far positions we spawn onThe thing is that while there are 3 possibilities of positions, close ground, close air, cross. Each time you load up a game it has equal chance to roll any of them, completely independent of previous rolls, that means it will never go 33% under any sample size..... it could be 99% close ground positions in theory, thats the wonderful thing about randomness. While its not completely random, no RNG can be completely random, only as random as you can program it to be. If I did a 100 game sample, and someone else did the same, the most likely result would be that our spawn ratios are completely different. I know that for the most part when i play metal I get cross positions, but you might air air positions, and playerX gets a fairly even mix. All of our results are completely independent of the other and therefore under a RNG we should get different results each time we run a test. The sample size needed to get an RNG to even out distribution of events is unfathomable, billions if not trillions or perhaps a google games could have to be loaded before it evens out | ||
|
Fallen33
United States596 Posts
| ||
|
Frozenserpent
United States143 Posts
Even 50 can be sufficient to obtain a p-value < 5%. In this case, p-value is definitely low enough to reject the assumption that it's 1/3rd. Obviously if you want the exact ratios you'd want to expand on the larger sample size, but this is conclusive to determine that it's not 1/3rds. | ||
|
magha
Netherlands427 Posts
On February 16 2011 04:56 Liquid`Tyler wrote: I hope you are agreeing with Toast.yum and making the additional point that an appropriate sample size is specific to the situation. That's exactly what I ment, yes | ||
|
RoyalCheese
Czech Republic745 Posts
![[image loading]](https://blog.startcom.org/wp-content/uploads/2008/05/debian-rng.jpg) ! | ||
| ||
