Sleuth Home - Message Boards - Sleuth Talk


0 0
judge o' character/success rate weirdness
  <<First Page  |  <Previous  

crunchpatty
crunchpatty
Old Shoe

Apr-18-2006 01:39

no audience, 'judge o' character' is not some irish guy with a gavel and a robe.

What I'm wondering is this (by wondering, I mean its making me nuts..yay subtlety gene):

how is it possible that in the same case you can try to interrogate witness/suspect A successfully when the judge of character skill has reported that the chance of success is 59, while trying the same trick on witness/suspect B can fail even though the chance of success is higher?

Don't get me wrong...it normally works. But in those instances when it doesn't, I'm all 'what the hell?'

Comments? Suggestions? Bueller? Bueller?



Replies

Jullianna
Jullianna

Apr-25-2006 19:54

Nice avatar, by the way. *winks and grins like an idiot*

Eden Zweig
Eden Zweig
Nomad

Jan-26-2012 12:16

"If Judge of Cahracter says you have an Adjusted Interrogation Chance of 59, then that should mean a 59% of success.

The numbers can go over 100"

@Sleuth Admin: That's mathematically a paradox. Chance and the probabilistic (just made it up now don't now what word I should use) statement of chance is always less than or equal to 1 which means you can't have a 100+ % of something as 100 means, all of it.

Unless of course, that 100 % is actually some standard set by a statistical number that corresponds to average success rates of detectives of more or less the same caliber. But this would mean that a there are very complex codes embedded in the game, I guess.

So theoretically 100+ percentages should all round off to 100 and that means you should be able to make the townie talk. Otherwise it's no use when you say 59% strictly means 59/100 chance to succeed. Said another way, 100/100 = 1. 1 is the greatest number in probability. Unless as I said, you are using probability distirbutions rather than probability to calculate the success rate, it is mathematically nonsense to fail with a probability rate that equals to 1. So which is it?

Or is there some conditional that affects you success but is -problematically- not calculated in the success rates stated by the bot during character judgment?

Eden Zweig
Eden Zweig
Nomad

Jan-26-2012 12:18

Sorry for the typos.

Cordelia Falco
Cordelia Falco
Battered Shoe

Jan-28-2012 23:45

Well, as you say, the number can't be a percentage, because you can get a number greater than 100. So yes, it's a number that indicates how likely you are to get your question(s) answered, but no, it doesn't represent the probablity expressed as a percentage. If you stop thinking of it as a percentage, it takes away the problem.

We don't know exactly how the number is used in the programming. One thing we do know, is that it represents not only how likely you are to get one question answered, but also two, or even three. The higher the number, the more likely it is that you get more questions.



R Anstett
R Anstett

Jan-29-2012 15:26

One reason for scores over 100% is that we only see one side of the game mechanic.

50% base chance
+25% skill bonus
+45% equipment bonus

for 120% listed chance

Unseen by the player (using these for example purpose only)

-25% based on toughness of case
-15% added randomly per suspect (Could be +15 or any number)

So total score that the game checks against is +90%

So yes to Eden there are conditional affects that effect your success rate.

Eden Zweig
Eden Zweig
Nomad

Mar-9-2012 11:02

Thank you for the replies. I thought I wouldn't get any.

R Ansttett, that's it, I guess. That explains it. Still, there's no way for us to figure out how much it really affects the case except playing long enough to have an idea, I suppose.

So the max. success rate one comes across is the known lower bound for "unseen conditional factor"(varies with each suspect, I know).

  <<First Page  |  <Previous  

[ You must login to reply ]