Opponent Modelling tries to predict the future actions of opponents, and is\nrequired to perform well in multi-player games. There is a deep literature on\nlearning an opponent model, but much less on how accurate such models must be\nto be useful. We investigate the sensitivity of Monte Carlo Tree Search (MCTS)\nand a Rolling Horizon Evolutionary Algorithm (RHEA) to the accuracy of their\nmodelling of the opponent in a simple Real-Time Strategy game. We find that in\nthis domain RHEA is much more sensitive to the accuracy of an opponent model\nthan MCTS. MCTS generally does better even with an inaccurate model, while this\nwill degrade RHEA's performance. We show that faced with an unknown opponent\nand a low computational budget it is better not to use any explicit model with\nRHEA, and to model the opponent's actions within the tree as part of the MCTS\nalgorithm.\n