A model of the rowing stroke is essentially a set of mathematical equations, each representing a different, yet interrelated, component of the stroke. When these equations are combined and solved, the solution (in most cases, this solution is boat speed) can be determined. Rowing models are generally designed to allow the user to change one or more variables (say, stroke rate or oar length) in order to estimate the impact that this change would have on relative boat speed - the change in boat speed compared to a base case. Because of the high amount of variability present in rowing, from stroke-to-stroke, rower-to-rower, race-to-race, and so on, the ability to predict absolute boat speed is impossible. The real utility of such models, then, lies in their ability to predict relative results - where the impact of modifying a variable can be measured while keeping all else constant.
While there have been numerous attempts at designing a full model of the rowing stroke (Atkinson, van Holst, Cabrera et al, to name a few), all are lacking in one regard: in their handling of the force of the water on the blade. As this force on the blade (both the magnitude and direction) plays a large role in determining boat speed, the ability to accurately calculate this value is crucial. The difficulty in determining this blade force lies in the complex interaction between the blade and the water throughout the stroke - something that is not easily observed or measured.
Where the model I have designed differs from others, is in the treatment of the blade-water interaction. While other models treat blade force estimates using rough simplifications (basically assuming that the blade is a stationary, flat rectangle with water moving past it), my model simulates the actual blade motion in the water through the stroke. Using computational fluid dynamics software to replicate the actual geometry of the blade and its trajectory through the water, the model is able to properly resolve the blade force throughout the stroke. For a full description of the model, you can read more here.
With the interaction of the oar blade in the water now modelled, it becomes possible for an in-depth study of this behaviour. The following section, oar blade hydrodynamics, details what is happening at the working end of the oar - from the catch through the finish.