--- id: "VOLT-HOME-WP-042" title: "When does heat-pump COP overwhelm the electricity-price signal?" slug: "when-does-heat-pump-cop-overwhelm-the-electricity-price-signal" description: "A transparent sensitivity compares the wholesale price signal per unit of delivered heat across four declared heat-pump COP values." published: "2026-08-30" cluster: "Heat pumps and thermal storage" status: "measured" evidence: "/research-data/home-papers/when-does-heat-pump-cop-overwhelm-the-electricity-price-signal.json" figure: "/research-media/home-papers/when-does-heat-pump-cop-overwhelm-the-electricity-price-signal.webp" figure_alt: "Chart for When does heat-pump COP overwhelm the electricity-price signal?: difference in price signal per delivered heat between COP 1.8 and 4.0, shown as COP 1.8, COP 2.5, COP 3.2, COP 4.0." source_ids: - "energy-buildings-heat-pump" - "nordic-cost-comfort" - "acer-retail-2025" - "ec-retail-flexibility-2026" - "dynamic-tariff-viability" peer_reviewed: false --- # When does heat-pump COP overwhelm the electricity-price signal? ## Abstract Price-aware heat-pump control should compare electricity prices per unit of useful heat, not electricity prices alone. This working paper applies four declared coefficients of performance—1.8, 2.5, 3.2, and 4.0—to the observed mean daily day-ahead price range in a frozen 2,367-observation temperature-price panel. The resulting signals are 194.731427, 140.206628, 109.536428, and 87.629142 EUR/MWh-heat. The difference between the COP 1.8 and COP 4.0 endpoints is 107.102285 EUR/MWh-heat, the registered primary result. This large sensitivity shows that efficiency can dominate a modest price difference, but the evidence does not estimate the outdoor temperature, flow temperature, or price spread at which a particular heat pump should switch operating hours. No uncertainty interval is reported for this estimand. The reduced-order synthetic scenarios are not building simulations, customer telemetry, savings claims, or retail bills. ## Plain-language answer A low electricity price is not automatically the cheapest time to make heat. If a heat pump produces much less useful heat per kilowatt-hour during that interval, its effective price per unit of delivered heat can be higher. In the declared sensitivity, moving from COP 1.8 to COP 4.0 reduces the price-range exposure from 194.73 to 87.63 EUR/MWh-heat—a difference of 107.10. The word “overwhelm” therefore means that the efficiency change is larger than the electricity-price advantage under comparison. This paper does not calculate a universal crossing point because it has no building-specific COP curve and no paired interval-level price-versus-COP schedule. It establishes the scale of the COP effect over four declared values. The [Energy and Buildings source](https://doi.org/10.1016/j.enbuild.2023.113257) supports treating heat-pump flexibility and tariff design together, while the [Nordic cost–comfort source](https://doi.org/10.3390/en18215568) supplies context for keeping comfort in the objective. Neither source turns this sensitivity into a device recommendation. ## Research question The registered question is when heat-pump COP overwhelms the electricity-price signal. We translate it into a reproducible screening calculation: given the observed mean daily wholesale price range, how much does the signal per delivered MWh of heat change between COP 1.8, 2.5, 3.2, and 4.0? This is not an optimisation of a heat pump’s hourly schedule. It does not model an air-source or ground-source unit, a manufacturer performance map, defrost cycles, supply-water temperature, auxiliary resistance heat, or a home’s heat-loss coefficient. The purpose is to stop a common category error: comparing wholesale electrical prices without dividing by the useful heat produced. The registered method family is a transparent reduced-order thermal scenario, and the registered trading-claims field is “none.” ## Data and provenance The evidence has a publication cutoff of 2026-08-30T00:00:00Z. It declares daily-price coverage from 2021-01-01 through 2026-08-29, detailed-interval coverage from 2025-10-01 through 2026-08-29, and long-history coverage from 2015-01-01 through 2026-08-29. The family’s source-table contract is `day_ahead_prices`, `zone_temp_weighted`, `zone_load`, `generation_mix`, and `forecasts`. The COP calculation itself uses 2,367 daily price summaries paired to available population-weighted zone temperature records. Temperature establishes the matched thermal-context panel, but the public statistic does not fit a COP-temperature relationship. The SELECT-only snapshot is identified by SHA-256 `7e97489fc8528c8cc8c38830e05b48d949ce1f67b98575dff26f5d7c321e4c67`. The analysis-code hash is `57c57de79cdab2b5b6d6c54c485cb5162598c5ba0b0bfe995da40d75e6c52ba9`; the protocol hash is `adb36bf6b447af9f96339249b8becaefc20422499cca1977242866347a97bd4b`; the paper-registry hash is `7bcb91d7476d0a69fe9fa75a5c7782f8117e0153f82f9112b7e1d307d3943717`; and the source-registry hash is `07949550ac443ff673fda5c0209b99f137544f3ffecf6775f109bb9d09663bd6`. The evidence and figure hashes are `d06190064cc2ddb63c1657d91b798dbe423a86b77587f0d36e86d9cc6d1b728c` and `3b95d68d8eb2ac31da4c5beabe1c649af6785385a37ddff0bd89b8510f577396`. The query contract records a 180-second statement timeout and a read-only transaction. The five assigned sources are contextual, not substituted data. [ACER and CEER](https://www.ceer.eu/wp-content/uploads/2025/11/ACER-CEER-2025-Retail-monitoring.pdf) frame retail contract choice as an enabler of flexibility. The registered [European Commission communication](https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A52026DC0115) concerns retail prices and flexibility. The [dynamic-tariff viability study](https://doi.org/10.1016/j.adapen.2024.100174) places heat pumps among household assets whose economics depend on tariff and enabling costs. This paper does not import any numerical result from those works into the COP calculation. ## Method For every matched zone-day, the analysis obtains the day-ahead maximum price and minimum price. Their difference is the daily price range. The arithmetic mean of those ranges is 350.51656949725395 EUR/MWh. For each declared coefficient of performance \(c\), the reduced-order price signal per delivered heat is the mean electrical price range divided by \(c\). The four calculations are therefore transparent: 350.516569 divided by 1.8, 2.5, 3.2, and 4.0. The registered primary metric subtracts the COP 4.0 result from the COP 1.8 result. No COP is estimated from temperature. No hour is selected. No heating load is moved. No compressor power, thermal store, or room-temperature trajectory is calculated. The assumptions are synthetic heat demand, no customer telemetry, COP sensitivity declared per paper, and wholesale energy only. “Wholesale energy only” means the comparison excludes supplier margin, network charges, taxes, value-added tax, fixed charges, metering, and any retail risk premium. The calculation also excludes heat-distribution losses, cycling, standby consumption, and auxiliary heating. The regenerated evidence reports no uncertainty interval for this estimand and names the interval method as “not reported for this estimand.” Because the primary metric is a deterministic transformation of the overall mean at two endpoint COPs, no interval may be reconstructed from stale spread resampling. Within-family Holm control applies to inferential claims; this descriptive paper makes no unadjusted significance claim. ## Results At COP 1.8, the declared signal is 194.7314274984744 EUR/MWh-heat. At COP 2.5 it is 140.2066277989016; at COP 3.2 it is 109.53642796789185; and at COP 4.0 it is 87.62914237431349 EUR/MWh-heat. The COP 1.8 minus COP 4.0 difference is 107.1022851241609 EUR/MWh-heat. The sequence is monotonic because the same positive mean price range is divided by increasing COP values. Moving from COP 1.8 to COP 4.0 cuts this reduced-order exposure by about 55%, but that percentage is an arithmetic property of the declared endpoint values, not an observed device improvement. The sample size attached to the daily spread distribution is 2,367. The evidence records no bootstrap interval for either the price-range input or the 107.102285 endpoint difference. The defensible result is that COP sensitivity is material at the scale of the observed price ranges. A universal operating threshold is not identified, and the point estimate is not accompanied by a precision claim. ## Robustness and placebo checks The four-point COP grid is the principal sensitivity check. It prevents the comparison from depending only on two extreme endpoints and makes the curve visible at intermediate COP 2.5 and 3.2. The smooth decline also verifies that the implementation follows the declared inverse-COP relationship. There is no empirical placebo heat pump. The evidence contains no resistance-heater control, no randomized operating schedule, and no counterfactual building with identical heat demand. It would be incorrect to claim that the COP relationship was validated against customer measurements. The arithmetic placebo is COP 1.0: under the formula, the per-heat signal would equal the electrical price range. That implication checks dimensional consistency but was not a separately reported evidence result. No bootstrap interval is reported, so sampling variation in the price-range input is not quantified here. Uncertainty in a real COP curve may also vary with outdoor temperature, flow temperature, frosting, part load, and maintenance. Within-family multiplicity governance is retained, but no p-value or causal claim is made. The paper also resists a tempting but invalid robustness shortcut: using the coldest day’s COP and the cheapest day’s price from different dates would move information across contexts and manufacture a schedule that was never available. ## Limitations The daily aggregation cannot determine whether high COP and low prices occur in the same operating interval. That alignment is the actual decision problem. A daily range divided by a fixed COP is a screening statistic, not a dispatch simulation. COP values of 1.8, 2.5, 3.2, and 4.0 are declared sensitivity points. They are not attributed to any model, brand, climate, or emitter system. The analysis has no indoor-temperature state and cannot verify comfort, recovery, or heat availability. Daily temperature and price summaries cannot reproduce a building-specific thermal state. The result also omits retail tariff composition. An interval with a lower wholesale price may carry a higher network time-of-use charge. Conversely, fixed charges may dilute both signals. [ACER and CEER](https://www.ceer.eu/wp-content/uploads/2025/11/ACER-CEER-2025-Retail-monitoring.pdf) and the [European Commission communication](https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A52026DC0115) are cited to frame that policy and contract context, not to fill missing household data. Most importantly, the reduced-order synthetic scenarios are not building simulations, customer telemetry, savings claims, or retail bills. The endpoint difference cannot be multiplied by annual heat consumption to claim savings. Such a calculation would require an interval-level load profile, a validated COP map, feasible thermal constraints, device losses, and the complete tariff. ## Practical implication A price-aware controller should rank candidate heating intervals by complete marginal cost per useful unit of heat, not by wholesale electricity price alone. At minimum, that requires an interval-specific COP estimate and all variable tariff components. If the expected COP deterioration is large enough to erase the price discount, preheating in the nominally cheap interval may be counterproductive. The occupant’s comfort and hot-water constraints remain hard boundaries. The Nordic source’s cost–comfort framing is relevant because an optimizer should expose the trade-off instead of silently selecting comfort sacrifice. A conservative controller should fall back to normal thermostatic operation when its COP estimate, weather input, or tariff mapping is stale or uncertain. ## Reproducibility The public evidence is `/research-data/home-papers/when-does-heat-pump-cop-overwhelm-the-electricity-price-signal.json`, and the figure is `/research-media/home-papers/when-does-heat-pump-cop-overwhelm-the-electricity-price-signal.webp`. The JSON freezes the assumptions, table contract, coverage windows, sample size, primary value, explicit null interval, interpretation, limitations, disclosure, and provenance hashes. To reproduce the figure, calculate each matched day’s price range, take the mean across the 2,367 observations, and divide that mean by 1.8, 2.5, 3.2, and 4.0. To reproduce the primary result, subtract the fourth value from the first. Do not infer an interval from an earlier analysis run; the regenerated evidence explicitly reports none for this estimand. A reproduction must use the frozen publication cutoff and snapshot. Adding later days, estimating COP from another weather source, or inserting retail charges creates a new analysis. Licensing and attribution conditions are documented at `/legal/data-licensing` and [Voltcast data licensing and redistribution](https://github.com/ossedk/voltcast/blob/main/docs/voltcast/LICENSING.md). ## Disclosure Analysis and drafting were model-assisted; sources, code, assumptions, and evidence hashes are disclosed. This public working paper is not peer reviewed. It contains no trading claim and is not trading advice. Volt has no live traders or live capital. The scenario is not a manufacturer performance statement, heating-system design, comfort recommendation, or instruction to override safe equipment controls. ## References - Energy and Buildings. “Assessment of the thermal energy flexibility of residential buildings with heat pumps under various electric tariff designs.” https://doi.org/10.1016/j.enbuild.2023.113257 - Energies. “Exploring Cost–Comfort Trade-Off in Implicit Demand Response for Fully Electric Solar-Powered Nordic Households.” https://doi.org/10.3390/en18215568 - ACER and CEER. “Rewarding flexibility: How retail contract choice can help unlock consumer flexibility.” https://www.ceer.eu/wp-content/uploads/2025/11/ACER-CEER-2025-Retail-monitoring.pdf - European Commission. “Communication from the Commission on the Citizens Energy Package.” https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A52026DC0115 - Advances in Applied Energy. “Assessing the conditions for economic viability of dynamic electricity retail tariffs for households.” https://doi.org/10.1016/j.adapen.2024.100174