The design and implementation of online auctions has given rise to a unique set of bidding strategies that has stimulated a growing body of research. We make use of a theoretically grounded, well-understood, and empirically observable bidder behavior-the winner's curse adjustment for the expected number of bidders in an auction-to examine the relationships between bidder experience, bidding patterns, and the winner's curse adjustment in rare coin online auctions. We also examine the impact of uncertainty on the winner's curse adjustment, both by using precise measures of uncertainty and by considering seller and bidder strategies for reducing that uncertainty. We analyze a complete record of all auctions in a three-month period for rare U. S. coins, examining 284,681 bids from 62,625 auctions hosted by eBay, the market leader in online auctions. One of the main contributions of this paper is to demonstrate that the bidding patterns associated with different bidders are strongly related to whether they calculate their bids to take into account the number of competing bidders, as predicted for common-value auctions. This is a substantial extension and empirical confirmation of prior work that has explored the implications of different observed patterns of bidding. We also explore new territory by examining the relationships between bidder experience, bidding patterns observed, and the economic outcomes for bidders. We are able to show that bidders with more domain-specific experience (rather than general auction experience) make better adjustments for the winner's curse, that experience has an effect on the type of bidding strategy, and that the type of bidding strategy has a significant effect on the economic outcomes for the bidders.