The standard definition of the breeding ratio is subject to some suspicion, since in many cases of practical interest it gives an incorrect measure of the fissile fuel doubling time and erroneous trends of the fuel cycle reactivity variation. The reasons for this behavior are given in this study, along with methods for removing the difficulties. In addition to the standard definition, the method of η weighting due to Ott and the method of reactivity weighting due to Baker and Ross (British definition) are examined. It is shown that the η weighting procedure is not much better than the standard definition, whereas the British definition can be used to give very good results for the doubling time and fuel cycle reactivity variation. The standard for comparison is a detailed explicit fuel cycle analysis of both the startup cycle and equilibrium cycle of an oxide fast reactor. With the methods given it is shown that the important quantities needed from a single fuel cycle analysis can be obtained just from a statics calculation, for the known composition at the beginning of the fuel cycle of interest (e.g., first fuel cycle). This result has significant importance for conceptual design studies and for optimization studies where many reactor calculations must be performed, precluding the use of an explicit fuel cycle depletion analysis. However, it is also shown that the equilibrium fuel cycle performance cannot be adequately predicted solely from start-of-life statics analysis. An approximate procedure is formulated to predict the equilibrium reactor composition, from which the equilibrium breeding gain, doubling time, and reactivity variation may be determined. The impetus for this approach is the need for an extremely rapid computational technique in optimization studies based on the equilibrium fuel cycle performance.