One Hour One Life Forums

AmberA

Member

Registered: 2019-07-02

Posts: 168

Re: Playtime is not linearly correlated with curses received

I'm one of the "good" vet players that was sent to donkey town. It's because I have a hard time not going after griefers. Someone kills all the sheep...damn right I'm going to hunt that person until they are cursed and dead. But that makes me a target, that sheep killer is likely to curse me for hunting them down. The random person who didn't see the sheep killing only see me with a bloody knife and also curses me. Taht sheep killer probably had a baby that will also curse me and try to kill me because I'm trying to end their griefer mother. That's 3 curses for trying to save the town. I've gotten better though at getting towns people on my side and letting others initiate the attack so I don't get cursed as often now.

Good thing we have leaders now. Instead of dealing with the problem on your own it's better to just report it to the leader. Leaders can deal with problems like this much quicker and easier and are unlikely to get cursed if they are doing it right.

NOBODY, you are in "big friends" group in Bobo's discord channel, it's good to see that you have hard time trying to go outside of donkey town.
I joined that channel just to check who is there, I left it right after. It's a pathological place...

Yup lately I've been between 2nd place and 10th place so most lives now I end up the leader. It makes things so much easier. I exile problem people and usually it sorts it self out after that. If they continue to be trouble the town people get rid of them or if the problem person behaves then I redeem them.

I only saw one life where the leader murdered everyone, since then I never pass on leadership to anyone, I let the system do it.

Mr.XIX

Member

From: The Netherlands

Registered: 2018-03-12

Posts: 175

Re: Playtime is not linearly correlated with curses received

Jason the first chart you made was this one:

https://i.imgur.com/YZ98xTu.png

Which shows a linear correlation, the more you play the more curses you will get, its a fact.
The more data points you have on the chart the more accurate it is, just look at the first 6000 min, in both charts you will see a linear correlation.
You changed the algorithm that draws the green line because you dident like it, to one that you like more.
It is important to be honest to yourself and to try to see the world how it is, even if the truth can be harsh. (general speaking)

i like the new curse forgiveness mechanic, this gives people who play a lot a chance to redeem some curses and i think it will make things more balanced

@jasonRohrer Please explain, because this makes you look really bad.

You shouldn't base your conclusion on the most extreme data points.

Arcurus

Member

Registered: 2020-04-23

Posts: 1,003

Re: Playtime is not linearly correlated with curses received

...
@jasonRohrer Please explain, because this makes you look really bad.

You shouldn't base your conclusion on the most extreme data points.

Actually the main point what he said / the data showed is that only a small percentage of the players end up in donkey town at all in the last 30 days.

The only left question which is not anwert yet is, how many % of the vet players / the top 10% of playtime ended up in donkey town. So for the most players the current system might be fine, but maybe it hits the vet players? There we need to see the data.

Whatever

Member

Registered: 2019-02-23

Posts: 491

Re: Playtime is not linearly correlated with curses received

This all started from here: https://github.com/jasonrohrer/OneLife/issues/625

But since then Jason added a forgiveness mechanic and fixed a bug that would send people with only a few curses to DT, so things should be better now.

st2019

Member

Registered: 2019-03-04

Posts: 50

Re: Playtime is not linearly correlated with curses received

When I used Rohrer's best fit it puts out a JASON!!!

---

I'm an expert for: Sharp Stones

olooopo

Member

Registered: 2019-02-21

Posts: 28

Re: Playtime is not linearly correlated with curses received

I like Roher's fit very much. Not only is it good scientific practice, but it also makes interpreting things flexible and convenient. It's a very sophisticated and useful tool in statistics for making data supporting your facts.

DestinyCall

Member

Registered: 2018-12-08

Posts: 4,563

Re: Playtime is not linearly correlated with curses received

"There are lies, damned lies, and statistics."

- Mark Twain