Demand Response is an attractive option to solve the problem of balancing supply and demand while providing flexibility to the grid. Buildings have been identified as highly valuable flexibility resources due to the increase of thermoelectric devices such as heat pumps for space heating and Domestic Hot Water (DHW). However, current methods for estimating the flexibility potential for residential buildings tend to rely on simplified models and often overlook the limitation of on-site control systems, especially for complex thermoelectric devices such as heat pumps. This contribution aims at accounting for this in assessing the flexibility offered by heat pumps for several market-based indicators using a simulation model and Model Predictive Control (MPC). Simulations are carried out to evaluate the proposed formulations and metrics, especially in relation to maximum flexibility potential fluctuations over a day or between seasons. Our results show that flexibility varies throughout the year, with maximum values reached at mid-season. Our results also emphasize the importance of the choice of metrics to assess flexibility.