Treus Posted April 21, 2017 Share Posted April 21, 2017 (edited) So I've found 200 chansey in the safari taking notes of how many I caught, the balls I threw to each, and the pokemon in between chanseys. Someone good at estadistics, what can you do with this info :D? Yes I am bored. https://1drv.ms/x/s!AjS7mTVNNRgEgQeg_B4LDmZ7ZQbY That's the excel First topic I make :D sorry if it's in the wrong section or something Edit: Uploaded it to OneDrive, I believe? I don't know the difference so I hope it's okay now. Edited April 21, 2017 by Treus kevola and Jeiseun 2 Link to comment
Desu Posted April 21, 2017 Share Posted April 21, 2017 Don't link raw document files. If you absolutely must link a spreadsheet upload it to an online editor like Google sheets or Onedrive. Diano 1 Link to comment
Gilan Posted April 22, 2017 Share Posted April 22, 2017 what does the Pok no Chansey mean? Does that mean that's the number of pokemon encountered before encountering the next chansey? Link to comment
Gilan Posted April 22, 2017 Share Posted April 22, 2017 (edited) so I went ahead and assumed that was the case. Here are stats that I was able to make: The two unlabled values are part of a risk analysis program. The 3.6085... is a number generated on the normal distribution of 2.13 as the mean and 1.48 as the std. dev.. What does this mean? It is basically a simulation of how many turns you would have in battle with a chansey before it flees (note that instances where you caught the chansey, were not included, as they would skew the data). This number was then used in the generation of the .15654... number. The equation that gives us this number is: 1-((1-J4)^IF(J9>1,J9,1)) . What does this mean? - 1-J4 = the odds of not catching a Chansey on a single toss. - J9 = the generated value of how many turns a chansey lasts until fleeing. - The IF statement takes care of the fact that the minimum number of turns you can have with a chansey is 1. So, when you combine that together you get the aggregate chance that you fail to catch a chansey in those # of turns. But we want to know the chance that we succeed in catching the chansey, so we take 1-(the odds of not catching the chansey). Then with this formula, I ran 10,000 simulations on it (thanks to the risk program) and the output is as follows: This plots # of encounters on the Y-axis. and % chance of catching the Chansey on the X-axis. Things of note: - Each Chansey you encounter, you have about a 10% chance of catching it (on average). Which is in line with the ovbserved catch rate of 20/200 = 10% - You will have about a 5% chance to catch the "unlucky" Chansey's - While you will have about a 17% chance to catch the "lucky" Chansey's - A Chansey will run on the first turn at least 1/5 of the time (it's probably more like 1/4 of the time, but how I did the stats, the numbers are continuous, i.e. a chansey can last for 1.3 turns). Suggesting that a Chansey will flee battle with a 25% chance each turn. so, yeah, have fun with these stats. Though there is really nothing surprising here. Edited April 22, 2017 by Gilan ImFeelingLucky, Summrs, Toupi and 2 others 5 Link to comment
notmudkip0 Posted April 22, 2017 Share Posted April 22, 2017 1 hour ago, Gilan said: nerd Gilan 1 Link to comment
OrangeManiac Posted April 22, 2017 Share Posted April 22, 2017 (edited) 10 hours ago, Gilan said: - You will have about a 5% chance to catch the "unlucky" Chansey's - While you will have about a 17% chance to catch the "lucky" Chansey's - A Chansey will run on the first turn at least 1/5 of the time (it's probably more like 1/4 of the time, but how I did the stats, the numbers are continuous, i.e. a chansey can last for 1.3 turns). Suggesting that a Chansey will flee battle with a 25% chance each turn. What does this "unlucky" and "lucky" Chanseys refer to? The Chanseys holding an item? And what particularly is a "lucky/unlucky" Chansey? Punches and eggs or? Also that "25 % chance of fleeing each turn" doesn't make sense with the data. Fleeing after 1 turn = 77 incidents Staying after 1 turn = 102 incidents This means 77/179 Chanseys fled after first turn. Making the chance of flee roughly 43 %. Fleeing after 2nd turn = 52 incidents Staying after 2nd turn = 50 incidents This means the Chanseys fled 2nd turn with 52/102 chance, which is roughly 50 %. Fleeing after 3rd turn = 24 incidents Staying after 3rd turn = 26 incidents This means the Chanseys that fled on 3rd turn was 24/50, making it 48 %. Fleeing after 4th turn = 13 incidents Staying after 4th turn = 13 incidents 50 % of the Chanseys fled on 4th turn. Fleeing after 5th turn = 7 incidents Staying after 5th turn = 6 incidents 54 % of the Chanseys fled on 5th turn. You can see where I'm going with this. First of all, it seems as the chance of fleeing every turn is standard. It also shows that the standard chance of fleeing each turn is somewhere close to 50 % (possibly even 50% precisely), rather than 25 % which means somewhere in your processes with that statistics program you went wrong. Also the statistics behind the lucky/unlucky seem really weird, unless you calculated the numbers for this particular person but meanwhile you assume in your graph that the chance of finding each item is 5 %. Simply put if you catch a Chansey every 10 % of the time you find specific item every 20th catch making getting egg chance 0,5 %. But yeah. Regardless, good work. Edited April 22, 2017 by OrangeManiac Link to comment
Treus Posted April 22, 2017 Author Share Posted April 22, 2017 (edited) 12 hours ago, Gilan said: what does the Pok no Chansey mean? Does that mean that's the number of pokemon encountered before encountering the next chansey? Yeah pok no chansey are the pokemon in between for example the first says 33, so the pokemon number 34 was chansey, and when it says 0 it means a chansey right after the last :O to be honest all I expected was "You find 1 chansey ever X pokemon, the average balls per chansey is X and there is not enough information to do anything else". In layman's terms, what does all that mean? Also should I continue with this endeavour? I'm guessing I'd have to actually catch like 200 chanseys to get better data rather than just find them :D Edited April 22, 2017 by Treus Link to comment
OrangeManiac Posted April 22, 2017 Share Posted April 22, 2017 (edited) Also with these aforementioned statistics where I made the assumption the flee rate is about 50 % / turn (which is based on by the raw data), you can calculate the chances to catch the Chansey each turn taking consideration the possibility of Chansey fleeing each turn. 4,6 % (1st turn catch) 95,2 % * 50 % * 4,6 % = 2,2 % (2nd turn catch) 95,2 % * 50 % ^2 * 4,6 % = 1,1 % (3rd turn catch) 95,2 % * 50 % ^3 * 4,6 % = 0,5 % (4th turn catch) 95,2 % * 50 % ^4 * 4,6 % = 0,3 % (5th turn catch) 95,2 % * 50 % ^4 * 4,6 % = 0,015 % (6th turn catch) Rest is getting so ridiculously small numbers. So these added together: 4,6 % + 2,2 % + 1,1 % + 0,5 % + 0,3 % + 0,2 % = roughly 9 % chance to catch a Chansey. Which is what the statistics here say, but just another way to approach this mathematically. Edited April 22, 2017 by OrangeManiac Link to comment
Gilan Posted April 22, 2017 Share Posted April 22, 2017 "unlucky" Chansey refers to a Chansey in the bottome 25% of cases in the distribution. So it's basically saying: "At worst, your chance to catch a Chansey is about 5%" (i.e. the catch rate, cause unlucky chanseys flee after 1 turn). "lucky" Chansey refers to a Chansey in the upper 25% of cases... I didn't factor in holding items into the program, cause yeah, they hold an item about 1/10 of them time (1/20 for either item). In regards to chance to flee. Yeah you're right. My program is not wrong, I just interpreted it in a poor way last night. The chance to flee is around 40-60% (probably is 50%). My mistake was only looking at 5% chance (i.e. 1 turn), which is about 25% of the time, and ignored the contiuous turns from 1-2 (i.e. I considered a turn of 1.345 as 2). Hope that makes sense. tl;dr The program is not wrong, my interpretation was just slightly off. OrangeManiac 1 Link to comment
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now