Topic: Stats: Who sent merits to Deleted topics/posts the most times (Read 306 times)

Cool, thanks for the offer. I'm good for the moment but in the future I might have a more specific query.

Ok, got it. If you need at some point a fair share of data that you see on the Merit Dashboard, but that would like to have in a more "mass" treatable format (i.e. google sheets), tell me and I’ll retrieve it from the DB itself, hopefully thus avoiding quite a bit of manual work.

Actually I got it from the "Section Subsection" tab and filtered by "Deleted." Then I copied that data to a PDF, then to an Excel spreadsheet, which is why it was a bit messy and not 100% accurate.

Then I subtotaled everybody and cross-referenced it with what was in the Potential Merit Sources tab.

There are a couple of ways to use the Merit Dashboard to derive the data. I’m not exactly sure which one you used, but likely the “Potential Merit Sources” tab, filtering by Section “Deleted”.
-> Doing so we can retrieve how many people we have merited on deleted posts (SentNetworkSize), the amount of sMerits, the amount of TXs involved, and a few other ratios.

I derive “deleted” as the set of posts that:
-   Are in fact deleted, and the scraper sees the “Deleted off-limits” screen.
-   Could not be retrieved by the scraper (few occasional errors).
I believe I overcame the Investigations board issue, since I use an Alt account that logs-in while scraping, and therefore should be able to see this board properly.

Up until two months ago or so, I used to re-download all merited posts on every update of the Merit Dashboard (weekly), so all posts that got deleted retrospectively outside the 120 day merit file window should have been spot-on, marked as deleted if they effectively got deleted.

Nevertheless, to do so meant that I needed to scrape data for 3 days each week just for that, and since the scraper I use is a free one, I’m limited to batched of 10k and 2 parallel processes. This requires to be on it and relaunch a batch every 5 hours or so. Since I’ve come across some availability restrictions, I cannot download all the posts weekly, so over the last few weeks some old posts (outside the 120 day window) may have been deleted and the Dashboard may not reflect it.
My idea is to do the full download once a month, to recap all old deleted posts.

@LoyceV’s current (last Friday) deleted data is as follows: Merit sent to “deleted” posts belonging to 35 distinct people: 41 over 38 sMerit TXs. I do lose the post title in the process as expected.

How recent is the Merit Dashboard's data on deleted posts? Assuming DdmrDdmr doesn't try to download old posts again, the real number could be a bit higher. If I look at my own full Merit history, I count 25 sMerit sent to deleted posts. This data is clearly outdated, but at least I keep the topic titles of posts that have been deleted later.

I usually send only 1 sMerit

Hi @nutildah,
I did not check, but I believe that if you do double check, those two list might partially fit each other.
How many banned users have you merited?
Those banned users have high odds of deleted posts/ topics due to spams, plagiarisms.
Top users sent their sMerits to users who got banned later are most likely appeared in your list of top sMerits senders to deleted topics/ posts.

---

By the way, would you mind to edit the topic title a little bit?
It is better, and more correctly due to only sMerits (sendable merits) can be sent, right?

I should clarify that 37 is the total number of times sent... you may have sent 1000 merits between those 37 instances (but I doubt it).


Thanks, I just played around with tr and td and think I figured it out... Will fix when I have some more time.

That's not that bad, 37 out of 2999 Merit sent Sometimes the entire thread gets nuked, for instance when I merit a post that warns OP is a scammer.

You can make a table by adding columns with "tr" and "/tr" to your spreadsheet:

Using DdmrDdmr's Bitcointalk Merit Dashboard, I did a rough analysis of all merits sent to Deleted topics/replies. It should be noted that this database counts all merits sent to the Investigations board as "Deleted" (are there other reasons why they would be put in the "Deleted" category?).

Out of 3997 users who had Deleted merits, the average amount of times sent per user was 1.91, with the standard deviation being 10.5.

Here is a list of the top 20 users by total number of instances:

Name   Rank   Instances
xandry   Staff   640
OgNasty   Donator   73
suchmoon   Legendary   60
frodocooper   Staff   55
bones261   Legendary   54
dbshck   Staff   46
Vod   Legendary   44
Foxpup   Legendary   41
paxmao   Sr. Member   39
LoyceV   Legendary   37
Vadi2323   Legendary   34
chimk   Sr. Member   29
vapourminer   Legendary   29
LFC_Bitcoin   Legendary   25
o_e_l_e_o   Hero Member   24
The Pharmacist   Legendary   23
DdmrDdmr   Sr. Member   22
deisik   Legendary   22
EFS   Staff   22
DarkStar_   Legendary   20

If you arrange members (with 10 instances or more) by the ratio of instances to Total Sent Merits, a different picture emerges:

Name   Rank   Instances   Tot. sMerit   Ratio
Vangardo   Sr. Member   17   31   0.55
fxpc   Sr. Member   10   21   0.48
xandry   Staff   640   1450   0.44
deisik   Legendary   22   74   0.30
ImHash   Hero Member   14   90   0.16
dumbtool45   Sr. Member   12   81   0.15
Parodium   Sr. Member   10   97   0.10
ace4549   Sr. Member   11   111   0.10
fruit   Legendary   11   118   0.09
foserfox   Legendary   13   141   0.09
Vadi2323   Legendary   34   390   0.09
TimeTeller   Hero Member   12   187   0.06
pawel7777   Legendary   10   193   0.05
richardsNY   Legendary   11   217   0.05
OgNasty   Donator   73   1742   0.04
o_e_l_e_o   Hero Member   24   775   0.03
ebliever   Legendary   10   327   0.03
frodocooper   Staff   55   1878   0.03
vapourminer   Legendary   29   1335   0.02
The Pharmacist   Legendary   23   1215   0.02

If you'd like me to give you the stats or ratio for any given member (based on my data set), let me know and I'll post it in this thread.

I apologize for the ugly formatting; don't know how to fix that except by investing a lot of time into it or else posting images of the spreadsheet.