Engineers and biologists have long desired to compare propulsive performance for fishes and underwater vehicles of different sizes, shapes, and modes of propulsion. Ideally, such a comparison would be made on the basis of either propulsive efficiency, total power output or both. However, estimating the efficiency and power output of self-propelled bodies, and particularly fishes, is methodologically challenging because it requires an estimate of thrust. For such systems traveling at a constant velocity, thrust and drag are equal, and can rarely be separated on the basis of flow measured in the wake. This problem is demonstrated using flow fields from swimming American eels, Anguilla rostrata, measured using particle image velocimetry (PIV) and high-speed video. Eels balance thrust and drag quite evenly, resulting in virtually no wake momentum in the swimming (axial) direction. On average, their wakes resemble those of self-propelled jet propulsors, which have been studied extensively. Theoretical studies of such wakes may provide methods for the estimation of thrust separately from drag. These flow fields are compared with those measured in the wakes of rainbow trout, Oncorhynchus mykiss, and bluegill sunfish, Lepomis macrochirus. In contrast to eels, these fishes produce wakes with axial momentum. Although the net momentum flux must be zero on average, it is neither spatially nor temporally homogeneous; the heterogeneity may provide an alternative route for estimating thrust. This review shows examples of wakes and velocity profiles from the three fishes, indicating challenges in estimating efficiency and power output and suggesting several routes for further experiments. Because these estimates will be complicated, a much simpler method for comparing performance is outlined, using as a point of comparison the power lost producing the wake. This wake power, a component of the efficiency and total power, can be estimated in a straightforward way from the flow fields. Although it does not provide complete information about the performance, it can be used to place constraints on the relative efficiency and cost of transport for the fishes.