Large-scale heat pumps represent a growing sector of electricity demand with distinct temporal dynamics linked to heating requirements. To reach cost-efficiency and CO₂ reduction, their operation must be aligned with the variable availability of wind and solar power. Fluctuating renewable generation leads to pronounced volatility in electricity prices and system conditions, which creates both risks and opportunities for flexible consumers in the spot market. This volatility makes flexibility in the heating sector a major lever for successful energy system integration. It also introduces numerous constraints, influencing factors, and uncertainties, where operators of district heating networks must balance cost and reliability under dynamic market conditions. Traditional procurement strategies based on stable, centrally dispatched generation are no longer adequate for consumers with flexible and seasonally varying load profiles.

The proposed study addresses this challenge by extending the classical “forward vs. spot” comparison for electricity procurement by adding a third dimension: contract timing. Using the German electricity market shock triggered by the 2021–2023 gas crisis, it asks when a heat pump operator should enter a long-term supply contract, and how that decision compares with active spot-market participation. The analysis combines spot market prices, forward settlement data for annual (CAL), quarterly (QRT), monthly (MON) products, and weather-driven heat demand profiles to quantify the cost of waiting, hedging early, or remaining exposed to spot prices, while highlighting that demand‑side management remains essential to optimize the residual, uncontracted load.

The study is structured in three layers.

First, it identifies the hindsight‑optimal purchase date for each delivery year by locating the minimum observed forward price in the relevant pre‑delivery window and computing the resulting annual heating cost for a fixed share of forward‑covered demand.

It then evaluates dynamic hedging schedules that gradually increase forward exposure as delivery nears, optimizing the forward vs. spot split via a mean‑Conditional Value at Risk (CVaR) formulation. Forward price scenarios stem from a regime‑switching model calibrated to settlement data. To represent the growing reliability of near‑term signals, a horizon‑dependent discount is applied to the CVaR objective: positions taken at the CAL tenor carry higher uncertainty weights than equivalent QRT or MON trades.

Third, it derives rule‑based timing signals, including momentum‑based “buy the dip” rules, seasonal renewal timing, gas‑spread triggers, and volatility thresholds, and evaluates how these rules interact with day-ahead trading by shifting flexible load in response to price deviations.

The study quantifies the economic penalty of buying at the wrong time and the value of locking in at the right moment. Methodologically, it reframes contract procurement as a timing and hedging problem. The resulting framework allows direct comparison of pre-purchased and spot-active strategies under real market conditions - illustrated via representative heat pump use cases - and delivers an interpretable benchmark for procurement decisions. This becomes increasingly relevant as growing shares of wind and solar generation amplify electricity price volatility and strengthen the role of flexible demand in smart energy systems.

The work is part of an ongoing PhD with the project SysMMO.