Optimistic beliefs are a source of nonpecuniary benefits for entrepreneurs that can explain the "Private Equity Puzzle." This paper looks at the effects of entrepreneurial optimism on financial contracting. When the contract space is restricted to debt, we show the existence of a separating equilibrium in which optimists self-select into short-term debt and realists into long-term debt. Long-term debt is optimal for a realist entrepreneur as it smooths payoffs across states of nature. Short-term debt is optimal for optimists for two reasons: (i) "bridging the gap in beliefs" by letting the entrepreneur take a bet on his project?s success, and (ii) letting the investor impose adaptation decisions in bad states. We test our theory on a large data set of French entrepreneurs. First, in agreement with the psychology literature, we find that biases in beliefs may be (partly) explained by individual characteristics and tend to persist over time. Second, as predicted by our model, we find that short-term debt is robustly correlated with "optimistic" expectation errors, even controlling for firm risk and other potential determinants of short-term leverage.