Evolution by natural selection is commonly described as ‘survival of the fittest,’ and this at first sounds simple enough. However, in natural populations, it can be entirely unclear who ‘the fittest’ really are because that depends on a multitude of factors and influences, many of which are constantly changing and some of which depend on the size and properties of the population. Mathematical models of evolution tend to neglect this complexity, making it difficult to judge whether they give an accurate account of evolution. In my thesis, I used a more versatile simulation approach that considers the members of a population individually, along with the landscape they inhabit. Specifically, I constructed models for the evolution of movement and competition strategies in various ecological scenarios. These models incorporate many more factors than are usually considered and almost develop a life of their own, as the rules of interaction can, to a certain extent, evolve themselves. My simulations reveal that evolution does often not converge to a steady state but rather can lead to oscillations or rapid switches between alternative states. Ecological and evolutionary processes act on similar timescales, making it impossible to understand them in isolation. In addition to these simulation studies, my thesis also contains work on mathematical models. In particular, I demonstrate that small changes in the models on the evolution of senescence can produce surprising new insights and explain the high early-life mortality of many organisms.