Would people report a different valuation for a 50 dollars Amazon gift card that is certain to be available, and for the same gift card that has a 10% chance of being available? In violation of expected utility theory, we show that people anticipate less utility from uncertain outcomes than from certain outcomes, even conditional on their realization. We show that it isn't driven by beliefs about the quality the good (we always use gift cards, which are of unambiguous quality), by a misunderstanding of the instructions, or by differences in “weirdness“ (Mislavsky and Simonsohn 2018) between the transactions.