Quote from: Hueristic on August 27, 2023, 11:44:49 PM

Quote from: philipma1957 on August 27, 2023, 11:38:18 PM

Quote from: Hueristic on August 26, 2023, 10:26:23 PM

Quote from: LoyceV on August 26, 2023, 08:33:55 PM

Quote from: fillippone on August 26, 2023, 01:02:39 PM

Quote from: LoyceV on August 26, 2023, 11:49:28 AM

Should I update How many banned users have you merited??

Please do.

Quote from: Hueristic on August 26, 2023, 03:37:54 PM

I'd like to see that list updated, I'd be interested to see how many I've fucked up on.
I have a notoriously fast ignore trigger so...

Also I think you should order that list (in the thread) by percentage rather than absolute which seems to me to be a much better metric as it can be incorrectly interpreted as displayed. Obviously those who distribute larger amounts will hit the scum more often then those that don't distribute at all.

Now if I wanted to start a witchunt, I'd start at those with the highest percentage.

Here's the update: How many banned users have you merited?

Quote from: LoyceV on August 26, 2023, 08:08:08 PM

(click for the full version)
1. 231 (8.66%) banned users merited by OgNasty (details)
2. 125 (4.66%) banned users merited by JayJuanGee (details)
3. 115 (5.33%) banned users merited by Ratimov (details)
4. 112 (4.51%) banned users merited by DdmrDdmr (details)
5. 108 (5.14%) banned users merited by hugeblack (details)
...
(click for the full version)

I don't think sorting by percentage is useful: thousands of users who sent only a few Merits have a 100% score to banned users.
Many users with a very high percentage have already been banned.

Disclaimer

Quote from: LoyceV on March 06, 2019, 12:10:37 PM

banned users can still have good posts. Meriting them isn't necessarily a bad thing.

Of course it is, outliers are easily identified and culled from the dataset.

Looking at the data as a percentage rather than an absolute is a much better way to get an accurate picture of trends over a long period of time and the longer the trend line the more accurate that metric becomes.

The issue with sorting in absolute order is the fact that not all users are dealt an even amount of cards to begin with.

Also highlighting sources in that list would be meaningful as well.

If you look at that list one guy or girl was over 60% and is now banned.

and the next worst is at 12%

hell theymos is over 6% lets ban him he does not make the cut. please note that is a joke.

I am not on the list because I am slower with my merits and really take time to not give merits easily to a newbie.

Why is that because back in 2019 5 or more people I merited were banned so I decided to be very deliberate with merits to anyone under 400 merits.

I reall never give to a guy with 100 posts and 110 merits as I suspect them as a merit alt being feed by a few legends.

Thus I give a lot to heros and legends.

By absolute:

33. 38 (3.42%) banned users merited by philipma1957 (details)

132. 13 (2.68%) banned users merited by Hueristic (details)

By percentage:

2794. 3.42% (38) banned users merited by philipma1957 (details)
2965. 2.68% (13) banned users merited by Hueristic (details)

You can easily see how it looks way worse by absolute for those that disseminate large volumes.

Tongue

Funny I was surprised to see Noic was banned when checking my list.

I think I remember his account getting hacked but couldn't find out anything on that after a search.

Good guy really liked him.

we gave out 5769 merits between us and 51 banned users which is 51/5769 = 0.88 percent and does not match the 3.42% or the 2.68%

because in my case a very rarely give a 5 merit to a newbie I give a 1 merit.

this means my % is very likely to be higher due to lots of 1 merit posts to a newbie

2794 and 38 is 38/2794 = 1.36% not 3.42%. so it is not absolute number it is the % most likely found by different people merited .

that would mean I merited 1111 different people a total of 2794 merits and 3.42% were banned giving 38 losers

is this the correct stat for me