We integrate into a single optimization problem a risk measure and, either arbitrage free real market quotations, or financial pricing rules generated by an arbitrage free stochastic pricing model. We call a good deal (GD) a sequence of investment strategies such that the couple (expected-return, risk) diverges to (+infinity;-infinity). The existence of such a sequence is equivalent to the existence of an alternative sequence such that the couple (risk, price) goes to (-infinity; - infinity). Moreover, by appropriately adding the riskless asset, every GD may generate a new one composed only of strategies priced at 1. We show how GDs exist in practice, and study how to measure a good deal size. We also provide the minimum relative (per dollar) price modification that prevents the existence of GDs. This is a crucial tool to detect over/under-priced securities or marketed claims. Many classical actuarial and financial optimization problems can generate wrong solutions if the used market quotations or stochastic pricing models do not prevent the existence of GDs. We illustrate this and show how GD indices help overcome this caveat. Numerical illustrations are given.