Q-Spreads: Measuring Risks and Opportunities of Quarter-Hourly Shape on Power Markets

Defining "Q-Spreads"

The Q-Spread is a quantitative benchmark representing the theoretical upper bound of Day-Ahead arbitrage revenue for an energy storage asset. It measures the price differential between the highest and lowest price intervals.

Figure 1: Q-Spread Designed on Price Curve Example

Q1 and Q2 design based on 2025, 17th June price curve. Q1 is the standard Max-Min spread. Q2 averages the top 2 and bottom 2 extremes.

Note: This analysis is based on Sequence 2 (EXAA, Gate Closure 10:15 CET).

Formula: Daily Q-Spread. For any specific day \( d \), the Q-Spread \( S_{d,Q} \) is calculated as:

$$S_{d,Q} = \frac{1}{Q} \sum_{i=1}^{Q} \left( P_{\text{max}, i} - P_{\text{min}, i}\right)$$

Where:

\( Q \) = Number of 15-min intervals (e.g., for a 2-hour battery in a 15-min market, \( Q=8 \) )

\( P_{\text{max},i} \) = The \( i \)-th highest price on day \( d \).

\( P_{\text{min},i} \) = The \( i \)-th lowest price on day \( d \).‍

Example: The Q-spread \( S_{d,8} \) denotes the maximum value in EUR/MW that can be achieved with a two-hour battery on day \( d \).

From Daily Data to Annual Benchmarks

Volatility is not uniform. To build a robust business case, we aggregate daily data into statistical percentiles. This allows us to separate "business as usual" from high-value outlier events:

• P50 (The Baseline): The median value. On 50% of days, the market spread was higher than this value, and on 50% it was lower. This represents the "Base Case" or average daily revenue for a typical operating day.
• P90 (The Upside Potential): The value exceeded by only the top 10% of days. This represents higher spread scenarios, including events like Dunkelflaute.
• P10 (The Downside Potential): The bottom 10%, i.e. days with lower intraday spreads indicating a stable supply-demand balance throughout the day.

The gaps between P10, P50, and P90 are effective KPIs for hourly price volatility and can be intuitively visualized. Figure 1 illustrates the quarter-hourly price curve and the resulting spread depth. With increasing $Q$ the spread naturally decreases as it absorbs less extreme price spreads. The curve acts as a "speed limit" – a physical asset cannot capture more revenue (on a particular market) than the Q-Spread for the respective duration

Note: To grasp the seasonal pattern of volatility, e.g., due to weather and demand shifts, a Q-spread analysis on quarterly granularity can reveal further insights.

Market Diagnostics: Insights from Historical Data

Structural Volatility vs. Pre-Crisis Baselines. Historical data analysis shows that the (German) power market has shifted into high intraday spreads and higher spread volatility. Disregarding the unprecedented price volatilities of 2022, the power market intraday spreads and spread volatility have increased since 2019, with 2025 (so far) attaining new peak levels.

Revenue Scenarios: The Volatility Risk Premium (P50 vs. P90). BESS revenue potential is characterized by a significant "opportunity premium". Figure 2 compares the P50 and P90 Q-Spreads. The data shows a wide divergence: the P90 spreads are nearly double the P50 values for the same duration. This confirms that a substantial portion of annual value is concentrated in a minority of high-volatility days.

Figure 2: Q-Spread Base Case and Spread Premium

P50 and P90 minus P50 Q-Spread curves for 2019-2025.

Note: 2025 is represented by available data up to 2025, 16th December.

Strategic Implications: Opportunity Windows by Asset Class‍

 The 2025 Q-Spread data indicates distinct performance profiles across different battery durations.

• 1-Hour Systems (Q=4): According to the data, 1-hour assets achieve the highest P50 of around €148/MWh and the largest upside deviation (+€94/MWh in P90 scenario vs. P50). The spread premium confirms that the decay in value is the steepest between Q=1 and Q=4,indicating that the spread value is concentrated in the first hour of duration.
• 2-Hour Systems (Q=8): The P50 moderates to 92% of the 1-hour system's per-MWh revenue potential (~€136/MWh) while reducing the absolute downside risk deviation from €67/MWh to €64/MWh. A 2-hour asset can capture the majority of daily highest spreads, offering a similar downside risk profile to 1-hour assets.
• 4-Hour Systems (Q=16): The P50 spread capture declines to ~€118/MWh, representing approximately 80% of the 1-hour unit potential. However, the risk profile narrows further, with the downside deviation decreasing to ~€56/MWh. At this duration, the asset trades peak spread intensity for a narrower distribution range, capturing less extreme but more consistent price differentials.
• 10-Hour Systems (Q=40): The analysis shows significant diminishing returns, with the P50 falling to ~€76/MWh – roughly 50% of the 1-hour benchmark. The upside spread potential shrinks to ~€39/MWh. The flattening of the spread premium curve in this range indicates that extending duration beyond 4 hours yields minimal additional spread value per MWh, shifting the primary driver from spread maximization to bulk energy shifting.

Figure 3: Different Asset Classes Provide Different Opportunities

Average 2025 spread values by asset class and opportunity curve.

Conclusion

The 2025 power market data demonstrates a persistent volatility premium compared to pre-crisis levels. The historical Q-Spread curves highlight that value is increasingly concentrated in sub-hourly intervals (Q1–Q4) and high-spread tail events (P90). For asset owners, this data underscores the necessity of 15-minute optimization strategies to capture the full depth of the available spread.

‍

At Qmesa, we don't just model these spreads; we help you capture them. From market-based Q-Spreads to grid-constrained Flexible Boundaries, we structure the risk-return profile of your entire flexibility portfolio.

Ready to analyse your asset’s true potential? Contact us at markus@qmesa.eu to model your specific case.