21 replies · 648 views · thread synced · 5 days ago
· View on torn.com
About this thread
Posts archived:22 / 22 posts (100%) · the total is Torn's reply count + the opening post at the last fetch
Counted by TornLife from the archived posts.
Archived posts
22
Discussion span
→
Authority score
55 / 100
Historical score
36 / 100
Story score
56 / 100
Engagement score
66 / 100
People posting, likes and official posts are not counted for this thread yet: on threads longer than one page they come from a periodic pass over the archive, which has not covered it.
Most-liked replies
TheUltimateSalt[2171375]
· 22 likes · Think this is way too small of a serving size to test.... I leave this to people like Proxima and Andyman
So i decided to do a little actual reserach on it. Im at 30nnb, with 4 CE merits, the Federal Judge Bonus, and the crime enhncer for the crime i chose. (Kidnap Women) i used my a max PI (+LSD bonus), a Ranch at 1200, and my shack. These are my findings.
Kindap Woman Happiness 5333 Success 17 Fail 11
Happiness 1200 Success 15 Fail 13
Hapiness 100 Success 15 Fail 13
Did 28 of each with 1304 nerve, using some stacked points from an AP. Very minimal difference overall, none between Ranch and Shack. Other variables could effect it too. But as far as i can see it does effect it slightly, with higher happy giving more greens. Of course this can be built upon, happy hunting to the proper TORN-Scientists
Andyman did some but his was also a capped success rate NNB. (Post 45) I figured some numbers from someone who may still be affected by lowered natural success rate could lead to a different study.
The biggest issue here(besides the sample size) is that you're not taking other variables like crime experience into account.
Without controlling for that, you'll never be able to tell how much happiness impacted the failures or if they were result of the loss of crime experience when at your highest happiness. (which 11 failures means you've lost somewhere around 220 kidnappings worth of exp)
For each outcome, take the square root of that number to get a crude estimate of the error associated with it. For instance, 100 successes has an error of +/- 10. You'd need to see the number of fails fall outside of the range of 90-110 for your data to hint at a difference in probabilities between the two.
Thank you for taking the time to investigate and share. Regardless of the limited amount of samples used, it may offer clues that even additional bonuses do not take it further perhaps? Maybe the additional success form the happy was over-kill? Read that there was a potential "max" success rate for some crimes...
End of the day, Torn is really a "black-box" and maybe only a few would know how exactly things are calculated... If it is assumed that getting a red gives -20x crime exp.... you would have gotten -693x loss on your crime exp doing this....so who knows. Maybe crimes 2.0 will offer more insight...because as it is, we are either doing 'Arson Warehouses' until the end of time or other crimes that we have no idea if they actually increase the invisible crime exp.. Would be more engaging to have crimes that offer visible risks and opportunities... The balance is way-off imo, and hope it is being looked at in the update..
as others have said, this is very lacking on the sample size, but it's a start. more importantly, it has started some discussion on it.
btw for Kidnap Woman, the success rate it has seems to point to it being more or less the same in CE gain as Arson Warehouse. and there are no blue fails for Kidnap Woman, only red.