Changing Strategies in Ebay Bidding
Ebay is a popular online platform for auctioned goods, one that many of us are familiar with. Particularly to myself, I have spent the last 5 years selling my father’s old collectible coins, rare figurines, and billiards artifacts on ebay for a significant profit. Because of this, I was extremely interested in the discussion of auctions and game theory. Studying the strategy of buyers on ebay can help to maximize my profits through fully understanding bidding strategies, strategic pricing, and auction lengths.
Bidding in ebay is essentially a second price sealed auction, where, as learned in class, it is one’s dominant strategy to bid your true value. As discussed in this Stanford analysis of ebay auctions, the two factors that are particularly relevant on ebay are “sniping,” or bidding at the last minute to steal the item from someone else. Another key factor is the fact that the same people will bid for one item multiple times, therefore driving up the price of the item. As a seller myself, I have experienced this. I have watched bids I started at $.50 go up incrementally every few days and skyrocket to hundreds of dollars within the last few minutes of the auction period. This is interesting, as in class we did not discuss the timing of ongoing auctions and how this would affect the bids of buyers. Because of these strategies, there have actually been “sniping agents” developed as software to ensure you win the item at the correct time.
Because ebay items are often antique, rare, and unique, another factor comes into play: how do you know what your true value really is? For example, if you value an item at $1000, but the standing bid 1 minute before the auction closes is at $1001, do you submit a bid to win the item? Is the rarity and instant competition factor of this process enough to push people to overbid?
These are topics for more research regarding auctions. Specifically, when a complex ebay auction is involved, with bidding wars, “sniping,” strategic late bidding, and software to ensure your win, I come to wonder whether bidding your true value stays a dominant strategy in this auction. The introduction of these softwares are particularly interesting, because if sniping software becomes more and more used across ebay, the site would transform from a second price auction to a second price sealed bid auction, and therefore, bidding strategy on the site would drastically change.
Another paper, published by the University of Berkley, claims that in real life on ebay, overbidding is the most common strategy. This is interesting, especially from a seller’s perspective, because selling things on ebay could actually result in more net profit as opposed to selling on Amazon or another website, where true value bidding proves to be a dominant strategy. The influence of rare items along with the trend of “sniping,” human pride, and the emotions surrounding bidding wars make it more possible for users to battle with other ebay bidders, ultimately driving up the sale price of the auction. I would like to learn more about the limit of this overbidding- is there a level of price to which the user will drop out of the auction, no matter how rare the item is or how intense a back-and-forth bidding war becomes?
These external factors shed a light on our study of idealized game theory. In the real world, emotional and psychological factors will guide humans at a greater level than logical factors, causing people to sometimes lose in their payoff for ebay auctions by winning the item but paying a larger price than their true value. This also leads to the question of how accurate are one’s “true values” initially, if intense bidding wars and sniping can cause an increase in amount willing to pay from your true value. Additionally, technological and software improvements have the possibility to completely change the way humans and internet auctions interact, and new auction theories could be developed to determine auctions governed by external computer agents rather than humans themselves. This article shed a lot of light on my interaction with ebay as a seller myself, and also has inspired me to research more into “sniping agents,” to see if these could affect the items I purchase though ebay auctions in the future.
Sources:
“The Timing of Bids in Internet Auctions: Market Design, Bidder Behavior, and Artificial Agents” https://web.stanford.edu/~alroth/papers/eBay.ai.pdf
“Study of Auction Theory in ebay data” 10/18/2020; https://blogs.cornell.edu/info2040/2018/10/18/study-of-auction-theory-in-ebay-data/
“Study of Auction Theory in Ebay Data” https://www.stat.berkeley.edu/~aldous/Research/Ugrad/selene_xu.pdf