The dataset accompanies the paper "People use fast and flat simulation to reason about new games" and includes human behavioral data, model implementations, and analysis code for studying intuitive game reasoning. It contains human judgements from "just think" experiments, live human-human play matches, and watch experiments where participants predicted outcomes of observed matches. Model data includes simulations from heuristic search, Monte Carlo Tree Search, and the Intuitive Gamer agent, along with inter-model comparisons and funness evaluations. The stimuli cover 121 board games with varying sizes, winning conditions, and categories, enabling analysis of how people and models evaluate and reason about novel games through fast simulation.