Signature campaign stats: users, bans, posts, deleted posts - page 2.

suchmoon

legendary

Activity: 3654

Merit: 8909

https://bpip.org

I have updated the lists through September 30 so this now includes our new favorite campaign.

I have also made a couple of other changes:

Quote from: DdmrDdmr on September 26, 2019, 12:56:09 PM

The exercise is very interesting, although I do find the PPD rather low in general. For example Chipmixer has a PPD of 2.4, but going over the last two weeks of the spread sheet it is more in the 4.4 to 4.6 area (some random lower week tolled 3.87). It may be related to dividing by the 110 distinct participants over a 365 day period. If we do it by a rough 60 concurrent participants, we get a 4.4 PPD, which is in line with the spreadsheet. Ofcourse we know the number of cuncurrent users in Chipmixer, but not off-hand for all the other campaigns.

I have adjusted PPD to count only "active" days and users who posted on those days (as per the definition in the OP). So this is going to be even higher for Chipmixer. It may discard some low-activity days but I think it better reflects the overall activity.

Quote from: o_e_l_e_o on September 26, 2019, 05:18:56 PM

Any chance of a "posts deleted divided by number of active days" column? I feel that might be a useful metric for both comparing campaigns and finding the worst offenders.

Done - see "Del/Day".

And last but not least - I have to say I'm very impressed by the number of deleted posts in the CryptoTalk campaign. I don't have time to report that crap but my hat's off to whoever is doing it, and please don't forget to report serial offenders to yahoo62278.

suchmoon

legendary

Activity: 3654

Merit: 8909

https://bpip.org

Quote from: Little Mouse on September 27, 2019, 08:31:29 AM

Just a question-
Imagine, an user has participated in Chipmixer campaign, later he switched to yobit (Noone will do in reality Grin

), and got banned. Which one you are counting? Current signature?
If you are counting current signature, what if they have removed the signature?

I'm counting the ban against the last known signature within 7 days before the ban. There might be some inaccuracies but it goes both ways - some bans may get uncounted, some may get double-counted - e.g. if someone within 7 days switched campaigns twice AND got banned. This should be very rare though.

Quote from: Little Mouse on September 27, 2019, 08:31:29 AM

It's now 4000, I have shared it on behalf of you Cheesy

Thank you

Little Mouse

legendary

Activity: 2268

Merit: 2327

Marketing Campaign Manager |Telegram ID- @LT_Mouse

Just a question-
Imagine, an user has participated in Chipmixer campaign, later he switched to yobit (Noone will do in reality Grin

), and got banned. Which one you are counting? Current signature?
If you are counting current signature, what if they have removed the signature?

Quote from: TheBeardedBaby on September 27, 2019, 08:11:58 AM

Oh, lovely thread. Pity I'm out of smerit to give you your 4000th one.. Sad

There you open some new work for me to browse and see how many spammers I can report in each campaign and to get in contact with the managers.

It's now 4000, I have shared it on behalf of you Cheesy

Most problems are with altcoin campaign, where 99% campaign manager never think to check the post quality, they only thinks about the post counts.

TheBeardedBaby

legendary

Activity: 2240

Merit: 3150

₿uy / $ell ..oeleo ;(

Oh, lovely thread. Pity I'm out of smerit to give you your 4000th one.. Sad

There you open some new work for me to browse and see how many spammers I can report in each campaign and to get in contact with the managers.

Quote from: Little Mouse on September 27, 2019, 08:31:29 AM

~
It's now 4000, I have shared it on behalf of you Cheesy

Thank you

suchmoon

legendary

Activity: 3654

Merit: 8909

https://bpip.org

Quote from: eddie13 on September 26, 2019, 07:28:33 PM

Is this any good? What is the ban for spamming?

First offence - typcally a tempban, and it takes a lot of spamming to get that. Some extreme serial offenders could eventually get a permaban but that'd be very rare.

Quote from: eddie13 on September 26, 2019, 07:28:33 PM

Aren't straight permabans mostly for only plagiarism, ban evasion, and malicious links?

Pretty much.

Quote from: eddie13 on September 26, 2019, 07:28:33 PM

Could you easily redo it to count all bans?

Sadly no. modlog contains only permanent bans (and not all of them). However as a comparative metric it's still quite good I think. A campaign that had 10% of its participants permabanned is very likely not managed as well as a campaign that had 1 or 2%.

eddie13

legendary

Activity: 2296

Merit: 2262

BTC or BUST

Yeah, now this is what I was talking about!

Quote from: suchmoon on September 26, 2019, 12:51:28 PM

I count only permabans.

Is this any good? What is the ban for spamming?
Aren't straight permabans mostly for only plagiarism, ban evasion, and malicious links?
Could you easily redo it to count all bans?

Quote from: o_e_l_e_o on September 26, 2019, 06:40:18 PM

If any of these campaigns is having excessive numbers of their posts deleted for spamming, then that bounty manager should either do their job properly, or the campaign should be banned.

I'd say ban the bounty/campaign manager a couple times before you go banning the signature URL..
If they have to replace the BM a couple times then maybe ban the URL, or if managed offsite then just ban the URL..

Then their will be risk to the companies if they agree to low quality with low quality managers, or risk for 1 strike and your out if they choose to manage offsite..
This will create incentive to hire quality managers and to not manage offsite for the companies..

I wanna see the bad BMs/CMs accounts get WREKT and push the companies to hire managers that do a good job and will only agree to do a good job.. So this would be good for the economy of the quality managers..

o_e_l_e_o

legendary

Activity: 2268

Merit: 18771

Quote from: hosseinimr93 on September 26, 2019, 06:59:35 PM

Unfortunately, managers of altcoins signature campaigns accept any user applies. Just check some of their spreadsheets. I just wanted to say that even if a campaign has lower number of deleted posts, it doesn't mean that its manager is doing his job properly. The lower number of deleted posts is only due to his luck. Because he has accepted all the applications.

Ahh, I misunderstood your tone. Such are the perils of communicating via a forum.

I agree with your statement that pretty much all altcoin bounty managers are not doing the job they are supposed to be doing and are accepting any old trash to their campaigns, and as you say, number of posts deleted is not an absolute measure of quality. It is a useful indicator, though, and it is no surprise that in general the altcoin paying campaigns have many, many more posts deleted than the bitcoin paying campaigns. There are obvious a couple of exception to this, most notable the recently reactivated Yobit spam.

hosseinimr93

legendary

Activity: 2380

Merit: 5213

Quote from: o_e_l_e_o on September 26, 2019, 06:40:18 PM

If you are talking about bounties in relation to signature campaigns which pay in a(n) token/altcoin, as opposed to signature campaigns which pay in bitcoin, then you are correct to say that bitcoin paying campaigns generally have much higher quality posts than token/altcoin paying campaigns.

I was talking about signature campaigns that pay participants with altcoins.

Quote from: o_e_l_e_o on September 26, 2019, 06:40:18 PM

This, however, does not mean bounty managers have free reign to let their participants spam low quality trash across the forum. In fact, the vast majority of spam comes from campaigns which do not pay in bitcoin; since they can create their payment token out of nothing, it costs them nothing to "pay" for this advertising and so they don't have to be picky and look for high quality posters. Instead, the go for quantity over quality and recruit hundreds of spammers.

If any of these campaigns is having excessive numbers of their posts deleted for spamming, then that bounty manager should either do their job properly, or the campaign should be banned.

You are completely right.
Unfortunately, managers of altcoins signature campaigns accept any user applies. Just check some of their spreadsheets. I just wanted to say that even if a campaign has lower number of deleted posts, it doesn't mean that its manager is doing his job properly. The lower number of deleted posts is only due to his luck. Because he has accepted all the applications.
We cannot compare the quality of mangers work based on deleted posts. Because all of them are doing a same thing. They accept every one.
If campaign "A" has less deleted posts than "B", it doesn't mean that the manager of campaign "A" is better than manager of campaign "B". Because both managers have accepted every one. Manager "A" is only more lucky than manager "B".

Check the spreadsheets. Only some users with red trust are not accepted. It is same in almost all of altcoins signature campaigns.

o_e_l_e_o

legendary

Activity: 2268

Merit: 18771

Quote from: hosseinimr93 on September 26, 2019, 06:06:09 PM

There is a big difference between bounties and bitcoin signature campaigns.

If you are talking about bounties in relation to Facebook, Twitter, Medium, etc., promotion, then this is irrelevant to these data, since they are "based on URLs found in user signatures".

If you are talking about bounties in relation to signature campaigns which pay in a(n) token/altcoin, as opposed to signature campaigns which pay in bitcoin, then you are correct to say that bitcoin paying campaigns generally have much higher quality posts than token/altcoin paying campaigns. This, however, does not mean bounty managers have free reign to let their participants spam low quality trash across the forum. In fact, the vast majority of spam comes from campaigns which do not pay in bitcoin; since they can create their payment token out of nothing, it costs them nothing to "pay" for this advertising and so they don't have to be picky and look for high quality posters. Instead, the go for quantity over quality and recruit hundreds of spammers.

If any of these campaigns is having excessive numbers of their posts deleted for spamming, then that bounty manager should either do their job properly, or the campaign should be banned.

hosseinimr93

legendary

Activity: 2380

Merit: 5213

There is a big difference between bounties and bitcoin signature campaigns.
- In bounties, the acceptance of your application does not depend on your post history, merit, spamming, etc.
- In bitcoin signature campaigns, your post history is always reviewed by managers.

So, if the number of deleted posts of participants of campaign "A" is higher than "B", we cannot say the bounty manager of "A" is worse than "B". Almost all bounty managers accept any user unless the user has red trust.

In bitcoin signature campaigns, the quality of posts of participants depends on manager. The manager remove any one spamming and try to accept high quality posters. In these campaigns the quality of posts depends on the rates too. It's normal that the quality of posts made by Chipmixer's participants is higher than posts made by Bitvest participants.

o_e_l_e_o

legendary

Activity: 2268

Merit: 18771

Any chance of a "posts deleted divided by number of active days" column? I feel that might be a useful metric for both comparing campaigns and finding the worst offenders.

I've certainly noticed (and have been reporting) a flood of "Minter" spam over the last week. This makes more sense now that I see they have recruited 300 users. Their total number of deleted posts doesn't look that high, but when looking at posts deleted per day, they are currently coming in at 23.5, which is the highest of the top 10 recently active campaigns in your second post. I'd be interested to see how it compares against all campaigns.

I appreciate the limitation in what I'm asking for as it does not take in to account total number of posts made in the campaign.

suchmoon

legendary

Activity: 3654

Merit: 8909

https://bpip.org

Quote from: khaled0111 on September 26, 2019, 04:47:32 PM

Excuse me suchmoon, but anyone who wants to jump to conclusions, must be sure he read the number of days for which the campaign have been running during the last year and if the campaign is still active or not.

Sure, that's why there is a list of current campaigns in the second post for a more apples-to-apples comparison but even that is not quite fair. Recently started campaigns haven't had a chance to get as many bans or deleted posts as the older ones, although they are trying for sure... see minter: 14 days, 4 bans.

Historic data can be quite interesting for other reasons, particularly since the totally-not-a-yobit campaign is making a comeback. I'm almost tempted to start taking bets on that one - will it be able to exceed the crappiness of the previous one before getting blacklisted.

khaled0111

legendary

Activity: 2744

Merit: 3097

Top Crypto Casino

Excuse me suchmoon, but anyone who wants to jump to conclusions, must be sure he read the number of days for which the campaign have been running during the last year and if the campaign is still active or not.

Impressed by ChipMixer stats! (DS is doing a great job,)

suchmoon

legendary

Activity: 3654

Merit: 8909

https://bpip.org

Quote from: bob123 on September 26, 2019, 12:53:18 PM

Quote from: suchmoon on September 26, 2019, 12:51:28 PM

bill gator was never permabanned and I count only permabans.

I am pretty sure he was. Because of plagiarism.

Later it turned out that it was not he who did the plagiarisms but the person he bought the account from.

He was temp-banned and sig-banned and got it overturned. No permaban.

Quote from: stompix on September 26, 2019, 12:49:47 PM

Cellard? Although it was also a temp ban!

cellard was first permabanned, then the ban was revised to temp+sig. That's why he counts and bill gator doesn't. It's just how modlog works.

Quote from: DdmrDdmr on September 26, 2019, 12:56:09 PM

I tried to make analyse signatures way back, but found that it was rather much a cul-de-sac to interpret (or at least it was for me).

The exercise is very interesting, although I do find the PPD rather low in general. For example Chipmixer has a PPD of 2.4, but going over the last two weeks of the spread sheet it is more in the 4.4 to 4.6 area (some random lower week tolled 3.87). It may be related to dividing by the 110 distinct participants over a 365 day period. If we do it by a rough 60 concurrent participants, we get a 4.4 PPD, which is in line with the spreadsheet. Ofcourse we know the number of cuncurrent users in Chipmixer, but not off-hand for all the other campaigns.

I have max users per day as well as user count for each day. It's just too complicated to glue it all together as evidenced by the issues above.

I have updated the list. The number of users should now be closer to "who posted at least once within the 1 year period" as stated in the OP. Some no longer active campaigns have dropped off the list.

Next I'll try to make the PPD more relevant.

DdmrDdmr

legendary

Activity: 2338

Merit: 10802

There are lies, damned lies and statistics. MTwain

I tried to make analyse signatures way back, but found that it was rather much a cul-de-sac to interpret (or at least it was for me).

The exercise is very interesting, although I do find the PPD rather low in general. For example Chipmixer has a PPD of 2.4, but going over the last two weeks of the spread sheet it is more in the 4.4 to 4.6 area (some random lower week tolled 3.87). It may be related to dividing by the 110 distinct participants over a 365 day period. If we do it by a rough 60 concurrent participants, we get a 4.4 PPD, which is in line with the spreadsheet. Ofcourse we know the number of cuncurrent users in Chipmixer, but not off-hand for all the other campaigns.

bob123

legendary

Activity: 1624

Merit: 2481

Quote from: suchmoon on September 26, 2019, 12:51:28 PM

bill gator was never permabanned and I count only permabans.

I am pretty sure he was. Because of plagiarism.

Later it turned out that it was not he who did the plagiarisms but the person he bought the account from.

Edit:
Alright.. he got banned for plagiarism, but it was just a temp ban instead of a perm ban. Huh

Sorry for the inconvenience.

suchmoon

legendary

Activity: 3654

Merit: 8909

https://bpip.org

Quote from: bob123 on September 26, 2019, 12:47:52 PM

Quote from: suchmoon on September 26, 2019, 12:18:00 PM

13. chipmixer.com 365 active 110 1 0.9 95613 2.4 327 0.3

Who is that one banned person ?

Do you only count actual lasting bans or generally all bans happened ?
What i mean is.. is the one person banned bill gator whose ban got changed to a sig ban ? Or is there someone else i am not aware of ?

cellard is the one banned user. bill gator was never permabanned and I count only permabans.

Quote from: stompix on September 26, 2019, 12:49:47 PM

Well, I don't think so Tongue

I don't either. There is only 1 actually posting user and a bunch of inactive ones wearing the signature for whatever reason. I'll fix this.

Quote from: DaveF on September 26, 2019, 12:50:30 PM

I think there are a lot of stragglers or some formula you are using is borked.
Mintdice only has ~40 people in it.
windice is ~50

The number will likely be higher than the current number of participants since I'm counting anyone who ever posted with that signature. Might even catch some applicants who didn't get accepted. I'm not too worried about that, it's the big discrepancies that I'll try to eliminate.

DaveF

legendary

Activity: 3500

Merit: 6320

Crypto Swap Exchange

Quote from: suchmoon on September 26, 2019, 12:40:41 PM

Quote from: LoyceV on September 26, 2019, 12:34:52 PM

How active is for instance deeponion's signature? Could it be abandoned, but many inactive accounts still wear the signature and a few of them made some posts in the last year? I haven't seen much of it anymore, while it was the worst spam-campaign at some point during it's peak.

I think I still have some issues with stragglers. Stand by, the number of users in some of those campaigns will likely go down and the list will be reordered.

I think there are a lot of stragglers or some formula you are using is borked.
Mintdice only has ~40 people in it.
windice is ~50

might want to check how you pulled the numbers. A few others I glanced at were also off by a large number.

ETA Or what the last 3 posts above said....

-Dave

stompix

legendary

Activity: 2912

Merit: 6403

Blackjack.fun

Quote from: bob123 on September 26, 2019, 12:47:52 PM

Who is that one banned person ?

Cellard? Although it was also a temp ban!

bob123

legendary

Activity: 1624

Merit: 2481

Quote from: suchmoon on September 26, 2019, 12:18:00 PM

13. chipmixer.com 365 active 110 1 0.9 95613 2.4 327 0.3

Who is that one banned person ?

Do you only count actual lasting bans or generally all bans happened ?
What i mean is.. is the one person banned bill gator ~~whose ban got changed to a sig ban~~ ? Or is there someone else i am not aware of ?

Edit: IIRC his ban got lifted completely, not turned into a sig ban.

Topic: Signature campaign stats: users, bans, posts, deleted posts - page 2. (Read 1341 times)