This article explains why pre–initial public offering (IPO) studies are not a valid basis for determining marketability discounts. They are unsound in concept because the pre-IPO transactions and the subsequent IPO are priced at materially different dates and because the IPO price is not knowable at the earlier date. They are unsound in practice for several reasons, such as selectivity—the data include only companies that subsequently become publicly traded—and the fact that any pre-IPO discount includes not only a marketability discount but also a second discount for the risk that the IPO may not take place.