From Wikipedia, the free encyclopedia - View original article
Auction sniping, is the practice, in a timed online auction, of placing a higher bid than the current highest public bid at the last possible moment (often seconds before the end of the auction), giving the other bidders no time to outbid the sniper. This can be done manually, or by software. The software can run on the bidder's computer or on an online sniping service accessible through a website. Use of an online service is said to decrease the failure rate of the snipe, because the website has more reliable servers and a faster Internet connection with a less variable delay in sending data (jitter), thus enabling the bid to be placed closer to the deadline, even arriving within the last second.
A bid sniper is a person, or software agent, which performs auction sniping.
While auction sniping is frowned upon by some people, it may not always break the rules established by the auction site. For example it is permitted by eBay. eBay Germany banned automated sniping services in 2002, but the ban was declared illegal by Berlin's County Court, and revoked.
A few other auction sites (such as iGavel, the New Zealand auction site TradeMe, and the real estate auction site Bid on the City) automatically extend the bid deadline by a few minutes if a bid is placed in the last moments of the auction, to give other buyers time to react. Another way to prevent sniping is to end the auction at a random time (e.g., at any instant within a one-hour interval).
Experienced bidders of online auctions with fixed ending times often prefer entering bids late in the auction to avoid bidding wars (multiple rounds of bidders each increasing their maximum bid to temporarily regain "current highest bid" status) or bid chasing (where the presence of an existing bid encourages others to bid on the same item).
Economic analysis of sniping (Roth and Ockenfels, 2000) suggests that sniping is a rational gain-maximizing (i.e., price-minimizing) strategy for bidders in auctions which fulfill two criteria: 1) the end time is rigidly fixed, and 2) it is possible to gain additional information about the "true" value of the item by inspecting previous bids. For example, a novice antiques buyer may prefer to bid in auctions which already have bids placed by more experienced antiques buyers, on the grounds that the items which the experienced buyers are interested in are more likely to be valuable. In this case, more informed buyers may delay bidding until the last minutes of the auction to avoid creating competition for their bids, leading to a lower winning bid. The highest bid 10 minutes before the end of a contested auction is often not at all representative of the eventual selling price.
Analysis of actual winning bids on eBay (Yang and Kahng, 2006) suggests that winning bidders are more likely to have placed a single bid late in the auction, rather than placing multiple incremental bids as the auction progresses.
Many online auctions use proxy bidding, an iterative sealed bid auction where winners pay a fixed increment over the second highest bid. The auctioner does not disclose the current maximum bid, but the second highest bid is always public.
In proxy bidding, the wise bidder must know in advance the "true" value of an item as a basis for their secret bidding limit. The fact that the maximum bid is revealed when it is outbid introduces the possibility of maximum bid fishing. Bidders unsure of the value of an item may incrementally increase their bid until they narrowly exceed the previously hidden maximum, thus placing themselves in a winning position without placing a very high bid.
Sniping eliminates this possibility and effectively converts the auction to a Vickrey auction, the same as a proxy bidding auction except that all bids are kept secret until the close of bidding.
This opens a loop-hole to fraudulent practice by a shill which allows an agent for the seller, which may be another account of the seller to raise the bid to the maximum. They then hope the original bidder will increase their maximum bid even by a small amount to win the auction. The danger to the seller in this case is that the original bidder may not choose to increase their bid, leaving the seller with a futile transaction (selling the item to themselves) which will still incur a fee from the auction service. However, it is possible for the seller to avoid this fee by submitting a mutually agreed cancellation from the buyer and seller application which will result in the waiving of the auction fee as well as a 'free relist' credit.
Non-sniping bidders object to sniping, claiming that it is unfair to place bids at a point when it is impossible or unfeasible for other bidders to evaluate and possibly counter the bid, causing them to lose auctions even though they would have been willing to meet the winning bid amount. Bidders sometimes object to sniping, when multiple, identical items are listed as a separate lot or on breakup auctions, where items that constitute a set are broken down and sold separately, as they must wait until the last minute to find whether their maximum bid on one lot has been exceeded before being in a position to bid on another.
However, online auction sites, unlike live auctions, usually have an automatic bidding system which allows a bidder to enter their maximum acceptable bid. This is a hidden or proxy bid, known to the system, but not any other bidders; during the auction the actual bid is incremented only enough to beat the existing highest bid. For example, if an item's current maximum high bid is 57 and someone is prepared to pay 100 and bids accordingly, the displayed bid will be 58, with the hidden maximum of 100. Proxy bidding also discourages "bidding wars" and can encourage "opportunistic" bidding on low-priced items.
The failure of a maximum acceptable bid beaten by a sniper prepared to pay more is not due to the act of sniping, unless the original bidder would have bid higher when they saw the price rise. For this reason, opposition to sniping can be analyzed as more of a subjective reaction to simply losing an auction for the usual reason (not bidding high enough), than a reaction to a "dirty trick". In other words, you would have still beaten the sniper if you had placed your maximum bid higher than the sniper is willing to pay.
However, if the minimum bid increment is very low, the Sorites paradox can come into play, and make it difficult for a person to establish a single maximum bid. For example, if the minimum bid increment on an auction is 10 cents, it can be difficult or impossible for a person to identify a price which they would be willing to pay to win the item, but they would not pay 10 cents more. The fear that 10 cents may make the difference between getting the item and losing it to a sniper can trigger unusual behaviour, such as deliberately bidding strange amounts (such as $274.81) to make it hard for snipers to guess the minimum required to outbid the current high-bidder.
One attempt to defeat automated bid sniping software is requiring bidders to pass a CAPTCHA test prior to entering their bid. This ensures that all bids are entered manually. The counter for this is for the software to re-direct the CAPTCHAs to a CAPTCHAs reading service.
Some online auction systems attempt to discourage sniping (manual or automatic) by automatically extending the auction time if a last-minute bid is placed. This approach leaves all bidding open, and allows any bidders who are watching during the final few minutes to raise the bid. It can also lead to last-minute automated out-of-control bidding wars between bidders, which can extend the bidding time long beyond what the seller desired, greatly raising the final selling price. Any site which implements a limit to the number of time extensions allowed simply causes a final extension snipe.
Some auction systems allow buyers to end an auction early by paying a predetermined final price for the item (generally substantially more than the minimum bid). This may discourage some sniping because another bidder can simply purchase the item outright while the sniper is waiting for the auction end time. Although if they did snipe for the item they could get it for substantially less than the Buy It Now price. In eBay this Buy It Now option disappears after the first bid is placed (or, for auctions which have a reserve price set, once the reserve is reached), and so the deterrent relates only to items that have yet to be bid on or that have not reached their reserve. Provided there is no reserve set, a sniper can switch the sale to an auction suitable for sniping by placing a very low bid on the item early on to make the Buy It Now option disappear, clearing the way for a snipe and preventing any other potential buyers from ending the auction early by buying the item at the Buy It Now price.
Buy It Now items can also be purchased using a similar technique called BIN sniping. BIN sniping takes place on the opposite end of an item's lifecycle where items are purchased immediately after they are posted. Sniping new listings requires more interactive involvement by users. Unlike auctions which can be reviewed at leisure and scheduled in an automated sniping program, new fixed price listings must be evaluated as quickly as possible and purchased as soon as possible in competitive categories. BIN sniping focuses on speed of evaluation and purchase execution rather than timing the final bid.