Two methods to improve the variance of statistical flux-at-a-point estimators over conventional unbounded estimators are developed that are readily implemented in multigroup Monte Carlo radiation transport computer codes. The theory behind the methods is developed, and the procedures for their application to Monte Carlo computer codes are outlined where necessary for clarity. Their application is demonstrated by the solution of a sample problem. These methods do not require a modification of the random walk, are easily implemented in multigroup Monte Carlo computer codes, and provide results that are comparable to other finite variance techniques.