Multiple regression is used to construct a prediction equation relating bidder competitiveness (the depen-dent variable) to the independent variables of bidder, contract type and contract size. The regression model shows that differences in contractor competitiveness are greater for different contract sizes than for different contract types. The most competitive contractors appear to be those with a preferred contract size range. Such a model can be used as part of a more systematic approach in prequalifying contractors. It may also be used by contractors as a basis for assessing bidding performance.