Power economic studies are challenging because of growing system sizes, the advance of smart grids, and economic, technical, and policy unknowns. Consequently, system models must be simplified so that they can be solved with many scenarios of renewable output, load, and long-run uncertainties. Scenario reduction, network aggregation, and DC linearisation are three common simplifications. The authors compare their errors and computation times for optimal power flow (OPF) and stochastic unit commitment (SUC), using the IEEE 14-, 30-, and 118-bus test systems, and also briefly discuss impacts on generation and transmission planning. The authors find that the most appropriate simplification depends on the study type, and there are no consistent results concerning which simplification is most distorting. The following example conclusions apply to these cases, but not universally; nonetheless, the findings provide information about what simplifications can matter, which is a helpful starting point for practicing modellers. The authors find that linearisation's disregarding of losses distorts total costs in OPFs, but it causes relatively little error in SUC. Scenario reduction reduces OPF computational times with little distortion but is less effective for SUC. Network aggregation decreases computation effort more than linearisation in OPF, but causes little error unless there are few scenarios.