Overconfidence is a significant bias evaluated in many lab settings, but researchers have struggled to examine it in real-world settings, where individuals respond to positive and negative feedback. Studying managers in a competitive setting, Raymond and his co-authors observed psychological functions allowing managers to maintain overconfidence despite poor past performance. This “rosy recall” is also linked to overconfident predictions of future pay and performance. Overconfident managers are likely to save company money by hiring fewer support personnel, but to sacrifice a level of customer satisfaction in doing so.