Page 2 of 3

Re: Act I with random choices - Probabilities

Posted: Thu Jan 08, 2015 12:18 am
by Atario
Mirage_GSM wrote:
Atario wrote:However, the devs' love of subverting genre meant that by trying to be nice to everyone, I insured my own Kenji ending those first two runs…
Uh, no.
Uh, yes. I'm telling you I tried to be nice to everyone, and got Kenji, twice. If you want to call me a liar, please do so explicitly.

Re: Act I with random choices - Probabilities

Posted: Thu Jan 08, 2015 1:15 am
by brythain
Atario wrote:
Mirage_GSM wrote:
Atario wrote:However, the devs' love of subverting genre meant created the situation that, by trying to be nice to everyone, I insured obtained my own Kenji ending those first two runs…
Uh, no.
Uh, yes. I'm telling you I tried to be nice to everyone, and got Kenji, twice. If you want to call me a liar, please do so explicitly.
Infelicitous word choice perhaps? :D Fixed it for you. I don't doubt the material facts of your case.

Re: Act I with random choices - Probabilities

Posted: Thu Jan 08, 2015 8:08 am
by Mirage_GSM
Atario wrote:
Mirage_GSM wrote:
Atario wrote:However, the devs' love of subverting genre meant that by trying to be nice to everyone, I insured my own Kenji ending those first two runs…
Uh, no.
Uh, yes. I'm telling you I tried to be nice to everyone, and got Kenji, twice. If you want to call me a liar, please do so explicitly.
I never called you a liar whether ex- or implicitly.
You seemed to be under the impression that you got to Kenji's ending because you tried to be nice to all the girls, and I explained in detail why you were mistaken in that assumption.

Re: Act I with random choices - Probabilities

Posted: Thu Jan 08, 2015 11:00 pm
by Atario
Mirage_GSM wrote:You seemed to be under the impression that you got to Kenji's ending because you tried to be nice to all the girls, and I explained in detail why you were mistaken in that assumption.
What you did was tell me I got a speeding ticket because my foot was pressing too hard on the gas pedal, and I'm telling you it's because I had to piss. What you said about those four choices was technically correct, but entirely missing the point.

Re: Act I with random choices - Probabilities

Posted: Thu Jan 08, 2015 11:18 pm
by ogorhan
Does it really all matter though? Like the thread says, its choices that are made randomly that give us these numbers. I doubt anyone started playing KS and just closed their eyes and picked a random option from the list. If you randomize everything that the high percentage of getting the manly picnic might seem correct but in reality it should be the one with the least % of getting it, since you usually have a some sort of goal on which girl you want to pursue, unless you go totally blind, though Rin and maybe to a lesser extend Shizune, are the ones that may be harder to get, which of the former I had to look up a flowchart to even get to her route.

In the end knowing what % of getting each route with random choices, serves little to no point to the average reader imo.

Re: Act I with random choices - Probabilities

Posted: Fri Jan 09, 2015 11:11 am
by BMFJack
ogorhan wrote:I had to look up a flowchart to even get to her route.
Seriously? I got Rin the first time I ever played Katawa Shoujo, and I wasn't even going for her.
ogorhan wrote:In the end knowing what % of getting each route with random choices, serves little to no point to the average reader imo.
Sure, but it's interesting to think about.

Re: Act I with random choices - Probabilities

Posted: Fri Jan 09, 2015 10:27 pm
by Notguest
Rin was the first girl I got, but I got the manly picnic twice first.

Mainly because I tried to go after Lilly while avoiding everyone else (including Hanako), which put me on the path to Rin by default.

Re: Act I with random choices - Probabilities

Posted: Sat Jan 10, 2015 9:20 am
by AlexKidd
Mirage_GSM wrote: Please keep in mind that few - if any - readers picked totally random choices.
For example if you don't tell Kenji that Hanako is "not cute" - which no sane reader who is trying for her path would do - you greatly increase the chance of getting her ending. Not pissing off Lilly during Cold War and going to the library instead of to town in Sip (Part 2) are also no-brainers.

I don't really agree with you on that.
I did not play completly randomly the first I've ever tried KS but I did not target one path, I just answered like I thought I would. So even though it's not entirely random, I played the game with no ideas of I would like to get.

Re: Act I with random choices - Probabilities

Posted: Sat Jan 10, 2015 9:32 am
by brythain
AlexKidd wrote:
Mirage_GSM wrote: Please keep in mind that few - if any - readers picked totally random choices.
For example if you don't tell Kenji that Hanako is "not cute" - which no sane reader who is trying for her path would do - you greatly increase the chance of getting her ending. Not pissing off Lilly during Cold War and going to the library instead of to town in Sip (Part 2) are also no-brainers.
I don't really agree with you on that.
I did not play completly randomly the first I've ever tried KS but I did not target one path, I just answered like I thought I would. So even though it's not entirely random, I played the game with no ideas of I would like to get.
In other words, you were biased in a particular psychological direction, so it wasn't random in any way—you were actively thinking your way through the game whether you had any particular girl in mind or not.

Re: Act I with random choices - Probabilities

Posted: Sat Jan 10, 2015 9:37 am
by BMFJack
AlexKidd wrote:I don't really agree with you on that.
On what, exactly? You agreed with everything you quoted.
Mirage_GSM wrote:Please keep in mind that few - if any - readers picked totally random choices.
AlexKidd wrote:I did not play completly randomly the first I've ever tried KS

Re: Act I with random choices - Probabilities

Posted: Sat Jan 10, 2015 9:47 am
by AlexKidd
Not everything, well not the way I understood what Mirage_GSM said. Especially when he talked about getting Hanako's path. I understood that as if you try to get one girl, then it's not random.

Surely I did not act 100 % random but I didn''t decided which girl I would end up with at the beginning, therefore I would not be influenced by something like "I need to do that to get that path".

But yes I was probably biased, cause I ended with the best girl on my first run (Lilly :mrgreen: )


I'm not sure I made myself understood though.

Re: Act I with random choices - Probabilities

Posted: Sat Jan 10, 2015 2:19 pm
by BMFJack
AlexKidd wrote:I'm not sure I made myself understood though.
Understatement of the year.
AlexKidd wrote:I understood that as if you try to get one girl, then it's not random.
...and that is correct. If you're trying for a specific girl, then you wont make random choices.
AlexKidd wrote:Surely I did not act 100 % random
Correctamundo. You said you acted as if you were in that situation, which is indeed not random.
AlexKidd wrote:I ended with the best girl on my first run (Lilly :mrgreen: )
Now there's a statement I can both easily understand and agree with. Good taste, sir ;)

Re: Act I with random choices - Probabilities

Posted: Mon Jan 12, 2015 12:43 pm
by d2r
I'd be interested in random play notes for all the others. Obviously Shizune's is 50/50 on account of there only being one choice to begin with, but the existence of some of the flags might make the others more interesting.

Hanako's route, for its part, has two endgame-affecting choices: "Beginning of the End", and "Faraway Presence".

If you choose to stay in "Beginning of the End", you automatically get the bad end irrespective of your decision in "Faraway Presence". Thus: 50% bad, 50% other

Assuming you go out, you then have the gate in "Faraway Presence" between good and neutral ends (depending on whether you leave Hanako alone or go to her room)

Thus Hanako's route, played randomly, would yield a bad end 50% of the time, a neutral end 25% of the time, and a good end 25% of the time.

The other routes are more complicated because of the flags, so I'm not going to attempt them.

Re: Act I with random choices - Probabilities

Posted: Mon Jan 12, 2015 6:03 pm
by Atario
The five routes are much simpler than Act 1, sometimes deceptively so.

Shizune:
  • 50% (1/2) bad
  • 50% (1/2) good
…based on comforting Misha or not.

Hanako:
  • 50% (1/2) bad
  • 25% (1/4) neutral
  • 25% (1/4) good
…based on going to town or not and listening to Lilly or not.

Lilly:
  • 87.5% (7/8) bad
  • 12.5% (1/8) good
…based on being open and honest three times in a row for the win.

Emi:
  • 37.5% (3/8) bad
  • 62.5% (5/8) good
…based on listening to advice or not, two or three times between three possible choice points.

Rin:
  • 41.6666…% (5/12) bad
  • 29.1666…% (7/24) neutral
  • 29.1666…% (7/24) good
…based on that whole complex of choices in Act 2 shaking out to fifty-fifty between two paths by the middle of Act 3. After that, one path gives you three choices (1/6 + 1/6 + 1/6 of the time) and the other, two (1/4 + 1/4); "then explain" gets you the bad ending (1/6 + 1/4 of the time = 5/12); the others (1/6 + 1/6 + 1/4 = 7/12) end up splitting fifty-fifty for good vs. neutral.

So, if you're random, Lilly is your toughest row to hoe, and Emi your easiest.

Re: Act I with random choices - Probabilities

Posted: Mon Jan 12, 2015 6:08 pm
by d2r
Image

This means that the random player, assuming they don't get "The Deep End" (the most likely result), will most often get the Emi good ending.