A disarmingly simple model of ecology does everything well — except for predicting how rapidly nature can change. Can it become more realistic while still avoiding biology’s messy complexities?