What is 'shill-bidding'?

According to eBay's shill-bidding FAQ: "Shill bidding is the deliberate placing of bids to artificially raise the price of an item and is not allowed. To avoid the appearance of being involved in this activity, family members, friends, and individuals living together, working together or sharing a computer should not bid on each other's items."
"...bidding on your own item is considered shill bidding and is not allowed."

To report a suspected shill-bidder to eBay, go to: the Shill Bidder Form-mail

What is 'shill-busting'?

Whatever you want it to be. general civic-mindedness, a chance to slap a shill who's cost you money or an engrossing hobby - it's your choice :-)

Two strikes and they're out!

According to eBay's shill-bidding FAQ: "We believe in educating members and providing them with a second chance. At times, a temporary suspension is made. Any suspension is serious, and sellers take it as such.

What happens if the seller shill bids again after the temporary suspension?

If the shill bidding continues at any time after the temporary suspension has ended, the shill bidder will be suspended from eBay indefinitely."

So get eBay to accept your report and they'll be suspended. Get eBay to accept a second report and it's a notch on your keyboard. They may come back as another member-name, but that one's dead. And if they do, you'll be there waiting for them anyway :-)

How can you spot shill-bidding?

According to eBay's shill-bidding FAQ:

"How do Bid History and Bidder Search help?

The Bid History page provides a view of the bidding and retractions that take place while the listing is active. Bidding patterns that show no advantage to the bidder but significantly increase the bidding price of the listing may suggest shill bidding. Questionable patterns include multiple bids by a bidder in short, deliberate intervals or bidding several times in small amounts even when other bids have not been placed on the item.

Sample Scenarios:

A member bids several times just under the highest bidder towards the end of a listing, incrementing the final sale price by a dollar and retracting if he/she inadvertently bids more than the high bidder.

A member bids 30 times or more even when there are no other bidders on the listing. The Bidder Search page allows you to view the Sellers of the items that a member has bid on. Sorting this list by the Seller ID allows unusual patterns to be seen., For example, a member who bids exclusively on one seller's items. Keep in mind, of course, that it is common for sellers to have loyal, repeat customers who frequently buy from them."

If you believe that another member is artificially raising the price of an item by shill bidding or has artificially inflated his feedback score through shill feedback please report it to eBay. Please be sure to provide the user ID's and the item numbers in question."

Which is fine as far as it goes. But to maximise the impact of your report you really need to PROVE your case. And with the information archived online via eBay search, you can make a compelling case with a little effort.

If you can't be bothered to put in an hour's work to prove your case, you're unlikely to convince eBay to suspend offending members. They're unlikely to put in more than a few minutes examining any case you refer because they're very busy people. They get thousands of assorted reports every week, many plain wrong or maliciously made to settle personal scores. And if you don't put the effort in, you're unlikely in all honesty to PROVE the case beyond reasonable doubt even to yourself. Hunches are a good start, but if you don't gather as much evidence as possible and carefully analyse it, your suspicions are merely OPINION - not PROOF.

So what evidence do we need?

in a word (well, two actually,) bidding histories. And for a convincing case, lots of them. As many as possible in fact. But before you start the time-consuming task of collecting them (which we will examine in further detail in part 2 of this introduction,) take a moment to assess the situation. You may save yourself much wasted effort collecting data on a bidder who isn't actually a shill, or collecting the WRONG data.

Go to http://pages.ebay.com/search/items/search_seller.html and enter the suspect seller's member name. Make sure you tick "Yes" for all completed items and hit search.

Make a note of how many auctions are listed. In another window, open http://pages.ebay.com/search/items/search_bidder.html and enter the suspect bidder's member name. Make sure you tick "Yes" for all completed items, and hit search.

Make a note of how many of the suspect seller's auctions the shill-bidder(s) has bid on, and how many other auctions he's bid on.

Firstly - the most important pattern you can reveal is what proportion of the seller's auctions the suspected shill(s) is/are bidding on, and you already have the figures to swiftly calculate this. This is your first piece of evidence to support your opinion. At this stage, that's still all it is.

Now seems a good time to introduce a fact which is pretty critical to the whole process.

There are two entirely different types of shill bidder!

• 1. a 'real' shill
• 2. an 'imaginary' shill

You can think of these like real or imaginary numbers if you're mathematically inclined - in essence a 'real' shill is an eBay member who is shilling for a friend in addition to his regular eBay activities. These are a lot rarer than imaginary shills, who account for probably 80-90% of all shills.

An 'imaginary' shill is not, despite its name, non-existent. It is merely an eBay member who is shilling on his own auctions via a secondary member-name. We are in the fortunate position that both types have more advantages than disadvantages - it's a WIN-WIN situation :-)

BUT - you've got to decide which type you're dealing with before you proceed with the investigation.

A 'real' shill has an eBay life outside the shilling. His feedback is likely to appear perfectly normal in relation to the amount of time he's spent on eBay, and will continually increase at about the same rate as his auctions won or sold. Spend a little time checking out his feedback (no need to keep it - unlike the auctions and bid histories, eBay archives all feedback forever):

Does it come from a wide assortment of other members with varying feedback levels themselves?

Is the feedback that's been left in the first half of the last month roughly consistent with the auctions he's won/sold in that time? (newer feedback may not have been given if the transaction and delivery proceeds slowly, but come back in a fortnight and check out the second half of the month you've saved data for.)

Is there evidence of 'reciprocal' shilling? Most real shills will work in pairs, bidding on each other's auctions, and this is their weak point that gives them away. It's twice the work to investigate 2 bidders, but the patterns are twice as visible and for twice the gain.

In my experience 'imaginary' shills are rather commoner than 'real' shills, at least in the corner of eBay I habitually frequent, so we will deal further with 'real' shills in part 2, as their detection is only reliably possible with bid-history analysis. For now we'll finish off by rounding up the 'imaginaries' to avoid this lesson suffering excessive data-bloat.

Imaginary shills

If he's been a member for over a year and has won 20 auctions in the last month, mostly or exclusively from one seller, yet has a feedback rating of 12, with nothing left for 3 months, then you may decide he could be an 'imaginary' shill.

The weak-point of an imaginary shill is fairly self-evident. He has no eBay life outside the shilling operation. He's just a cardboard cutout, and if you push him he'll fall over.

Check out when he joined eBay and compare it to the seller. If he joined a month after the seller and got his first feedback from him, that's a little evidence to support your opinion.

If the suspected shill has bid only or almost only on the suspect seller's auctions, this is a positive sign that you may be right. But remember that "it is common for sellers to have loyal, repeat customers who frequently buy from them." It is not unheard of for some customers to buy EXCLUSIVELY from a single seller. This proportion is not PROOF - it is, in conjunction with your observations regarding real/imaginary behaviour, merely fairly strong EVIDENCE that he may be.

If the suspect shill has bid on lots of different seller's auctions, and has bid on more of some other seller's auctions that on the suspect seller's, this is conversely not PROOF that he's not a shill, but it's pretty strong EVIDENCE that he's not an imaginary shill. If your hunch is still nagging you, turn the problem around: try to find evidence of shilling on this suspect's auctions - you may have misidentified what is actually a 'real' shill. If he's not a seller as well as a buyer, then you're probably wasting your time. Only you know how much investigation you're prepared to put in to uncover a shill, and individual cases do vary, but it really isn't looking very hopeful at this stage.

If you're really committed (or just enjoy the practice,) check other sellers' auctions he's bid on. Move on to the more scientific analysis techniques outlined in subsequent sections of this course.

CONCLUSION

Features of 'imaginary' shills

Imaginary shills do not have much of a life of their own. They may make a few small value purchases from other sellers to gather a bit of feedback history but there are tell-tale signs (ranked roughly in order of evidential strength):

They MAY have appeared on eBay at around the same time as the suspect seller.

The first feedback given is NOT UNCOMMONLY given by the suspect seller, and USUALLY the suspect seller has given a glowing feedback very early on.

They OFTEN bid on one seller's auctions only

They OFTEN bid several times in small increments after a bid has been placed by another bidder, and they MAY retract a bid that accidentally overtakes that bidder's maximum

They USUALLY have a suspiciously low amount of feedback given from the shill-seller member. The first one is useful to boost the feedback rating - subsequent ones aren't, so are often neglected. They may even be stupid enough to think this reduces the visibility of their scam - they're wrong! Absence is far more significant than presence where feedback is the primary fabric of eBay's virtual environment, and deficient feedback will swiftly reveal itself when bidding data is subjected to chi-square or Poisson distibution testing (we will deal with these advanced techniques in subsequent lessons.) Nature abhors a vacuum!

They ALMOST ALWAYS are registered in the same country.

They ALMOST ALWAYS share the same ISP as the seller - you can't easily ascertain this, (we will see in future lessons how you can confirm this with a bit of effort yourself,) but eBay admin can check this at the click of a mouse if you put them on the scent. They can also check the exact IP numbers used, and if there's a match between seller and bidder that's INCONTROVERTIBLE PROOF!

They NEVER snipe. It's not in their interest to risk winning or to reduce the chances of a bidding 'war'. That's probably the only truly definitive statement you can make about shills. If the bidder snipes an auction, he is NOT making a shill bid. If he always snipes, then he is NOT a shill.

A SIGNIFICANCE CALCULATOR for initial assessment of 'imaginary' shills