We consider the relative importance of wind and currents in the determination of optimal routes, commonly referred to as weather routing, for sailboats. Weather routing for sailboats is the process of finding the fastest route fromone place to another based on wind and current information combined with information about the boats performance. In this thesis we also take into account the uncertainty in the wind and current information by making use of ensemble predictions. Our overall conclusion is that currents is a decisive factor in the determination of the fastest (optimal) routes for sailboats, and can be a decisive factor in the routing process.The wind information is extracted from met.no’s ensemble prediction system LAMEPS. Each ensemble forecast is in turn used to force the ocean model ROMS to provide an ensemble of ocean currents as well. Regarding the routing, we study two different cases, one in which the boat is sailing against the wind (the upwind case) and one in which the boat is sailing with the wind from behind (the downwind case). This is repeated using each of the ensemble members as input, giving us an ensemble of possible optimal routes. We then compare the routes made with and without currents as input. For both cases, there is a reduction in sailtime of about 1 − 1.5% following the route based on both wind and current input compared to the route calculated using wind as input only, given that the weather and current would develop according to the forecast. We also calculate what is referred to as the mean and median route by averaging the routes geographically. These routes are simulated for all the different weather and current ensemble members. In three out of four cases, the mean route perform better than the route suggested by the deterministic forecast. This indicates that the use of ensemble predictions may help to improve weather routing for sailboats.