Located in Domain Marketplace Reviews started by prague7, Nov 4, 2009.
Oh Yeah. That will boost buyer confidence.
Mailing in my form tomorrow morning (still shaking my head). High XXX. I will never use Snap again.
Now that the bidding history has been put back online, I know alot of people are going through their bidding history to track down additional fraud accounts.
Halavrez is not the only fake account. Contrary to what Snapnames is saying, it is fairly obvious to me that more than one shill account was being used to bid against customers. If you spent more than a few hundred dollars at Snapnames, I advise you to review their bidding history for additional shill accounts and post the usernames on namepros. No doubt every big buyer has been doing that for the last few days.
Here is some advice for tracking down fraud:
1) Bidder frequently bids against you.
2) Bidder frequently loses more than 50% of the time.
If you have are seeing a bidder lose 90% of the time, that would be a major red flag that it is another halvarez account. This is something that Michael Berkins pointed out in his thedomains.com article. I have found additional accounts that are suspicious after reviewing my records.
If you find a shill account, please post it online so we all know. Snapnames is sitting on its hands and will not help you. In fact, they have not even charged Nelson Brady with fraud. Just shows how evil they are. We have to crowdsource ourselves the fake accounts.
---------- Post added at 04:52 AM ---------- Previous post was at 04:36 AM ----------
H/T to Michael Berkin for his post on Snapnames fraud.
Repost for how to find additional shill bid accounts ( in comments from thedomains.com post)
Let us define “Auction Value to Bidder” (AVB) as the sum of dollars paid by a user for auctions won divided by the number of auctions entered for any given sample of n auctions.
AVB = Sum(Completed Bid) / Count(Initial Bid).
If you can calculate AVB for all users in your sample with n > Halvarez’s that may float a few interesting things… And no, I’m not the one with the same ID, but in the sample I’m working on has “someone” with very low AVB.
Very interesting dumbluck. Brady has not been charged yet by anyone? Surely the police are involved by now? Has he even been questioned?
does anyone know if the username 'overbid' is real
lol....now people will find everything fishy
350,000 bids by 8,900 bidders in 12,000 auctions the results
We just published our conclusions from a detailed analysis of all our snapnames auctions
Plenty of info
Check it out:
12,000 Auctions, 8,900 Bidders and 350,000 Bids: We Conclude Halvarez Was The Only Shill Bidder At Snapnames | The Domains
That is simply BREATHTAKING analysis.
Fab analysis, numbers rarely lie. While I'm not directly involved in any of these auctions, small fry compared to the numbers above, I do have a solution here. Snapnames and Oversee won't like it, but its the only 'fair' one I see... refund 'halvarez's top bid for each auction he was involved in, irrespective if he was #3 or #20.
I see the problem this might cause snapnames, specially since its 50k auctions, but its the only way, imho, that they can claim that they didn't profit from it, nor would like to be seen profiting from it.
Oh and signing away any rights against a repeat of this incidence in the future, as the agreement seems to convey, will give snapnames access to a list of the most stupid domainers on this planet. Unless those signing only do it to take the money and run and leave snapnames forever, which might not be feasible in the long run.
Can someone please explain to me the following sentence from the worldwide report linked to above:
"In terms of comparing all bidders, he was the 5th most frequent second bidder, winding up in 2nd place a whopping 58% of the time, while only winning 5.7%."
This sentence makes absolutely no sence to me at all! How can the fifth most frequent second bidder be so on 58% of the time? That would mean that the 1st/2nd/3rd/4th frequent highest second bidders made up only 42% of the time between them!
I think I will go through this report, with such a blaring question mark on one of the basic pieces of information to be analysed it makes one wonder what else 'may' have been misinterpreted.
I can explain this further
If you look at the raw number of times each bidder was the 2nd highest bidder Halvarez was 5th, being 2nd, 339 times.
Yes there are 4 people, including myself who had more second place finishes than Halvarez.
I had 1,081 auctions I came in 2nd.
If you now take the 339 number and look at how many auctions Halvarez actually placed a bid in other than the minimum, 590, the percentage of the time he was 2nd compared to the number of auctions he actually bid on is 58%.
The next highest % of 2nd place finishes was around 35%.
My percentage for 2nd place finishes was around 13%, but then again I won almost 65% of the auctions I bid on, while Halvarez only won 5% of his.
Hope that clears it up
Michael B, thanks for your explanation, but that now leads to another question (probably due to my ignorance), but it would seem that you are saying you were the second highest bidder in 1,081 auctions against yourself or the company you work for/represent ("We just published our conclusions from a detailed analysis of all our snapnames auctions" from your post above #616)?
As I say this may be my ignorance of domain groups and companies, but it does seem rather strange to say the least.
Also as you point out in your report you only looked for patterns similar to the 'Halverez' bidding. It is not unusual for fraudsters to modify their fraud subsequently and each modification is generally a complimentary system rather than a duplicate. An example would be that where 'Fraud A' had inserted a bid then 'Fraud B' would not be permitted to bid, also the bidding pattern would be different so as to reduce the risk of discovery. If you have say 10 schill bidding accounts then the likelyhood is that at first only one existed, later perhaps two, then this number would grow, indeed a simple modification of cancelling one account and substituting it with another account would not have shown up in your analysis from what I have read of it. (Though I think you attacked the subject well, perhaps another sorting of the data with a much more jaundiced and skeptical eye might be needed.)
I commend you generally on your report as a first quick analysis, but several of your disclaimers are so important that they identify the fact that more analysis needs to be done. My concern is that Oversee.com has done this or suspects this that is why they are not answering such straight forward questions as to charges being laid against Nelson Brady and/or others.
I was not bidding against myself I was in an auction against other people 12,000 times. I won a ton of these, lost a ton finishing in second, lost a ton more finishing way down and other I never bid higher than my $60 initial bid.
""the fact that more analysis needs to be done""
I agree with you.
I wish I had access to the 98.8% of the 1,000,000 snapnames auctions I wasn't in so I can run data on all of the auctions and give you a compressive report.
I can only report on the info I have.
Great work, Mike....
The waiver clause on future claims...
I have no doubt Oversee (and, Rust?) have done an exhaustive analysis on 'halvarez' bidding patterns, too. Been at it for quite some time before they went public, I'd surmise....The decision they have made to offer compensation based on halvarez coming second in auctions, appears to me to be the minimum base position Oversee has adjudged they may get away with, to fix this, now. But, its thin, imo...
It'd be a fair guess, too, that they have been running models seeking patterns that might show up other false bidding identities, as well. Why would they not, once the 'halvarez' situation became apparent...?...And, intuitively, it'd be amazing if this bloke, Brady, had NOT run several false identities.
Whatever Oversee have concluded about multiple fraud identities, I suspect their view is: "If someone out there thinks there is more to this....then, prove it!"....They won't volunteer it, imo.
Its possible, they know more than they are disclosing, imo - and, hope settlements re halvarez, today - & waivers re all other fraud activity - will protect them.
We need to get to the bottom of this right away! Sounds like a job for Captain Obvious. Lol.
Worldwide - great analysis! Are you going to post the scripts you used to scrape snapnames.com for your bid histories? I suppose it wouldn't be that hard to write, but I'm a typical lazy programmer . I would like to analyze the 3,581 auctions in my history. Snapnames said they would "try" to get me the raw bid history data, but they haven't produced anything.
I'm glad you emphasized the outrageous liability release waiver they want us to sign. If there is going to be a waiver, it should only cover what they are actually compensating us for, which is the bidding activity of halvarez in the auctions where he was took 2nd place. Not every fraud, known or unknown, that Snapnames has ever or will ever engage in.
Any news on Brady? Or has he gone on a very long holiday overseas?
It is fairly obvious from Michael's analysis that Brady wrote a script to shill bid for him. Brady used a computer program to proxy bid for him. I have to say that there was more than one shill bid account. Take it from an expert.
Any programmer knows how easy it is to automate things. If you write a program to shill bid, why stop at 40 auctions a day? For someone as greedy as Brady, who was willing to shill on $60 auctions, it is clear he used more than one account.
What is outrageous is that Oversee has all the data and could run the analysis themselves. They could press charges against Brady and have him testify in a court what he did. Why is he not being charged? Why is he not being questioned?
Oftentimes, what is not being said is more important than what is being said. Look for the gaps. :bingo: Coverup.
Does anybody think that the same explosive revelation can happen to the parking industry ?
"No comment officer".
I am reviewing Michael's analysis and the extent of the fraud is huge.
If it is accurate that Michael participated in 12,000 auctions worth $13 million, then the value of the fraudulent auctions is easily $52 million. This is based on Snapname's admission of 50,000 affected auctions.
However, we know several things. 50,000 is the number of auctions tied to halvarez. Snapnames has not released additional details of additional fraud accounts.
We know that Snapnames has auctioned 1 million domains. Snap says only 5% of auctions were fraud. If the fraud number is 20%, that means $200 million of auctions could be affected.
Another point is that Snap only offers rebates to cases where halvarez is the 2nd highest bidder. But halvarez bidding activity clearly drove up the price of auctions where he was not the 2nd highest bidder.
Snapnames is in much bigger trouble than they are admitting. Oversee is in big trouble. $200 million in revenues and $100 million in profits every year makes one big plump class action target.
We will be posting an open source program in a day or two so that anyone can use it (not sure how much programming skills will be required to install and run, since I have none of these skills) but our programmer is making the changes to it that will allow anyone to plug in their account info and get the results I have.
As soon as its ready we will blog about it on thedomains.com and post here.
It would be a big plump class action target if it were not for the fact that most of those figures are guesswork. You are working under the assumption that Michael is an average snapnames member. If he participated in many of the higher value auctions then the figures you are quoting would be skewed quite badly.
Until Snapnames quote either the average value of the auctions you are not going to truly know how much extra revenue Snapnames made from this over time.
....wondering if Peter works at Snapnames. Carry on.
Because I am trying to apply some common sense into the mix? Some people in this thread are getting themselves hyped up over things that they either have no control over or because of what other people are claiming with no proof.
And to answer your question no I do not. It really is not that difficult to do research on me. I do not do whois privacy and I do not hide who I am. And to be honest you have been here long enough and should have seen me about a fair bit and know that I am not an employee for Snapnames.
Separate names with a comma.