Battery spreads
TB2 and TB4: the daily spread between the dearest and the cheapest hours of German day-ahead prices
Gas plants set the price in most of the dear hours, so the cost of gas-fired generation largely sets the level of the spread: through the gas crisis of 2021–2022 the spread rose and fell with it. Since 2024 the spread has been above the gas cost through most of each summer, as growing solar output pushes midday prices towards zero (in 2023 only in July and September). In winter, with little solar output, it has mostly stayed below the gas cost.
| Year | TB2 | TB4 | Average price | Cost of gas-fired generation | TB2 ÷ gas cost |
|---|---|---|---|---|---|
| 2019 | 28 | 25 | 38 | 37 | 0.75 |
| 2020 | 30 | 26 | 30 | 28 | 1.06 |
| 2021 | 76 | 67 | 97 | 114 | 0.66 |
| 2022 | 177 | 159 | 235 | 280 | 0.63 |
| 2023 | 92 | 80 | 95 | 115 | 0.80 |
| 2024 | 104 | 89 | 79 | 94 | 1.10 |
| 2025 | 116 | 100 | 89 | 103 | 1.13 |
| 2026 (303 days) | 152 | 134 | 114 | 138 | 1.10 |
TB2 (top–bottom 2) is, for each delivery day, the average of the two highest minus the average of the two lowest hourly day-ahead prices of the Germany/Luxembourg bidding zone, in EUR/MWh. TB4 uses the four highest and four lowest hours. Since 1 October 2025 the auction sets quarter-hour prices; each hourly price here is the average of its four quarter-hours, which keeps the history comparable. Spreads computed on quarter-hours, as in the EPEX SPOT TBx index, are somewhat higher.
The highest and lowest hours need not be consecutive, and the definition does not require the cheap hours to come before the dear ones. A battery with one MW of power and two (four) hours of storage that charges in the cheapest and discharges in the dearest hours moves two (four) MWh a day; TB2 (TB4) times that energy is what one such cycle could earn on the day-ahead market. It is an upper bound, not a revenue: it assumes prices are known in advance and no energy is lost (real batteries return roughly 85–90 % of what they charge), and it leaves out intraday trading, balancing markets, grid fees and battery wear. TB4 is never higher than TB2, because the third and fourth hours are less extreme; per day a 4-hour battery still earns at least as much on the day-ahead market (4 MWh × TB4 against 2 MWh × TB2).