A regular problem in regression analysis is estimating the comparative importance of the predictors in the model. This work considers the 'net effects', or shares of the predictors in the coefficient of the multiple determination, which is a widely used characteristic of the quality of a regression model. Estimation of the net effects can be a difficult task because multicollinearity among the regressors can produce negative inputs to multiple determination. This paper suggests estimating the incremental net effects as subsequent marginal inputs to the coefficient of multiple determination, and it is shown that the results coincide with estimation by cooperative game theory. This approach guarantees positive and interpretable net effects, which offers a better interpretation of the regression results.