A system and method for valuing a portfolio in terms of its performance relative to a specified benchmark under a range of future scenarios is disclosed. In particular, a portfolio is taken and two values related to the portfolio are calculated: the first value corresponding to an amount by which the value of the portfolio is expected to fall below the value of a benchmark over a given time horizon, and a second value corresponding to an amount by which the value of the portfolio is expected to exceed the value of a benchmark over a given time horizon, in view of the range of different future scenarios. Means for determining the portfolio which optimally trades-off these two values, and to evaluate risk/reward performance measures using these two values which can be used to rank instruments, securities or portfolios are disclosed. Means for pricing portfolio insurance for optimal portfolios are also disclosed.