Soil phosphorous (P) loss and its partition in various pathways may differ depending on water management practices. A study was conducted using large field plots equipped with automatic flow volume measurement and sampling systems over a 5-year period to determine the effectiveness of regular free drainage (RFD) and controlled drainage with sub-irrigation (CDS) for mitigating soil P losses of various forms (dissolved reactive P - DRP, dissolved un-reactive P - DURP, and particulate P - PP) and to identify the relative roles of surface runoff and sub-surface tile drainage in soil P loss. The CDS system produced greater surface runoff but much less sub-surface tile drainage relative to the RFD system. Sub-surface tile drainage accounted for 80 and 97% of total flow volume for the CDS and RFD systems, respectively. Cumulative total DRP, DURP, TDP, PP and TP losses in sub-surface tile drainage were 65 to 71% for the CDS system and 95 to 97% for the RFD system. The CDS system increased FWM of DRP, DURP, and PP concentration in surface runoff and reduced FWM of DURP and PP concentration in sub-surface tile drainage. The CDS system produced greater cumulative DRP, DURP, TDP, PP and TP losses in surface runoff but large reduction on cumulative DRP, DURP, TDP, PP and TP losses in sub-surface tile drainage relative to the RFD system. The cumulative total PP losses accounted for more than 80% of total TP losses for both the CDS and RFD systems. Combined with surface and sub-surface water, the CDS system reduced PP and TP losses by 15 and 12 %, respectively relative to the RFD system. Subsurface tile drainage played predominant role in soil P loss. CDS can be considered as a beneficial management practice to reduce soil P loss under the similar climate and relatively flat field conditions in Southern Ontario.