Widespread prostate-specific antigen (PSA) screening in North America has resulted in a profound stage migration and a marked increase in incidence. One in six men is now diagnosed, many with small-volume, low-grade cancer. This incidence is dramatically higher than the 3% lifetime risk of prostate cancer death that characterized the pre-screening era. This article summarizes the case for active surveillance for "favorable-risk" prostate cancer with selective delayed intervention for rapid biochemical progression, assessed by increasing PSA levels, or grade progression. The results of a large phase II trial using this approach are reviewed. To date, this study has shown that virtually all men with favorable-risk prostate cancer managed in this fashion will die of unrelated causes. Based on the Swedish randomized trial of radical prostatectomy versus watchful waiting, the Connecticut observation series, and the Toronto active surveillance experience, a number needed to treat analysis of the benefit of radical treatment of all newly diagnosed favorable-risk prostate cancer patients, compared with a strategy of active surveillance with selective delayed intervention, is presented. This suggests that approximately 73 patients will require radical treatment for each prostate cancer death averted. This translates into a 3- to 4-week survival benefit, unadjusted for quality of life. This figure is confirmed based on an analysis of the 2004 D'Amico et al. PSA velocity data in favorable-risk disease. The approach of active surveillance with selective delayed intervention based on PSA doubling time and repeat biopsy represents a practical compromise between radical therapy for all patients (which results in overtreatment for patients with indolent disease) and watchful waiting with palliative therapy only (which results in undertreatment for those with aggressive disease).