How was this page built?
Head-to-Head Backtest
Ran the same battery dispatch simulator for both 2-hour and 4-hour configurations across all 2,557 days. Same charge/discharge windows, same efficiency assumptions, only duration differs.
Revenue Comparison
4-hour earns EUR 4.6M/yr vs 2-hour EUR 2.85M/yr (+61%). The extra revenue comes from capturing more of each spread cycle (4 hours of charge at low prices, 4 hours of discharge at high prices).
CAPEX Scaling
4-hour battery costs ~2x the cells but only ~1.3x the BOP (inverters, transformers, grid connection are shared). Net: 4-hour at EUR 34.0M vs 2-hour at EUR 21.1M (+61% cost for +61% revenue).
CRM Advantage
4-hour duration qualifies for higher CRM de-rating (better “equivalent firm capacity” credit). This gives an additional revenue boost of EUR 8-15k/MW/yr over the 2-hour configuration.
Limitations: The analysis assumes identical charge/discharge windows for both durations. In practice, a 2-hour battery might optimise for shorter, sharper peaks (e.g., catching price spikes). The 4-hour advantage is robust across all 7 years tested, but the magnitude varies with spread structure.
Nerd level:
Show figures as
01
The Core Argument
Why the answer is unambiguous
Revenue ratio vs cost ratio
62%
of 4h revenue
2h earns 62% of 4h
EUR 56,619 vs EUR 91,246 /MW/yrEUR 2.83M vs EUR 4.56M /yr
<
must exceed
75%
of 4h CAPEX
2h costs 75% of 4h
EUR 25.5M vs EUR 33.9M
Revenue ratio (62%) < Cost ratio (75%) — the 2h battery saves on CAPEX, but loses proportionally more revenue. The gap is 13 percentage points, too large for any plausible market shift to close.
For 2h to win, it would need to earn >75% of 4h revenue per MW. It earns only 62%.
02
Revenue Per MW (2019–2025)
Arbitrage only, EUR/MW/year C1
Annual Revenue by Duration & StrategyAnnual Revenue by Duration & Strategy (50 MW) SEM backtest
4h consistently earns 60–66% more than 2h, every single year. The ratio is remarkably stable across both crisis years (2021–22) and normal years. Perfect Foresight (PF) is the LP upper bound; Fixed Schedule (FS) is a zero-intelligence lower bound.
03
CAPEX Breakdown
Where the cost difference comes from C2
Stacked CAPEX Comparison
MW-linked costs (grid, transformer, EPC, development) are identical for both durations. The only difference is MWh-linked cell and container costs.
CAPEX Detail
| Component | 2-Hour | 4-Hour |
|---|---|---|
| MW-linked (same) | €15.6M | €15.6M |
| Grid connection | €4.5M | €4.5M |
| Transformer | €2.8M | €2.8M |
| EPC / installation | €5.1M | €5.1M |
| Other (PCS, dev, legal, bond) | €3.2M | €3.2M |
| MWh-linked (differs) | €7.6M | €15.3M |
| LFP cells | €3.4M | €6.8M |
| Pack, containers, BMS, other | €4.2M | €8.5M |
| Contingency (10%) | €2.3M | €3.1M |
| Total CAPEX | €25.5M | €33.9M |
| Per MW installed | €511k | €679k |
| Per kWh installed | €256 | €170 |
Key insight: Grid connection, transformer, EPC base costs, development, and legal fees are the SAME regardless of duration. Doubling from 2h to 4h adds only €8.4M (25%) in cell-related costs — but unlocks 61% more revenue.
04
Why 4h Works in Ireland
Daily price shape analysis C1
Cheap & Expensive Hours per Day SEM daily average
Irish SEM has 8–9 cheap hours and 7 expensive hours daily — plenty of room for a 4-hour battery to fill and empty without hitting the steep part of the price curve.
Spread Concentration Best N-hour spreads, EUR/MWh
Hours 3–4 capture ~85% of the per-hour spread of hours 1–2. The extra hours are NOT marginal — they earn almost as much per hour.
Why the Irish Price Shape Favours 4-Hour
→
Wind-driven troughs persist for 6–10 hours overnight. A 4h battery can charge at low prices for the full window. The trend is toward wider troughs (8.0 cheap hours in 2020, 9.1 in 2025) as wind penetration increases.
→
Evening peaks span 3–5 hours. The 4h spread per hour (e.g. EUR 93.3/MWh for best-4h in 2025) is still excellent — not dramatically less than the best-2h (EUR 105.2/MWh).
→
The 2h spread is only 13–18% wider than 4h. Each of hours 3 and 4 still captures roughly 85% of the per-hour value of hours 1 and 2. This is the key market feature that makes 4h viable.
05
CRM Capacity Payment Advantage
De-rating widens the gap further C2
40%
4-hour de-rating factor
€59,984/MW/yr CRM revenue€3.0M/yr CRM revenue (at €149,960/MW de-rated clearing price)
25%
2-hour de-rating factor
€37,490/MW/yr CRM revenue€1.87M/yr CRM revenue (at €149,960/MW de-rated clearing price)
€22,494€1.12M
Annual advantage per MW for 4h
This alone covers a significant portion of the extra €8.4M CAPEX over the battery lifetime
The CRM advantage roughly pays for the extra cells. The 4h battery earns €22.5k/MW/yr€1.12M/yr more from CRM than 2h. Over 15 years at 50 MW, that is €16.9M — double the €8.4M extra CAPEX. Even the CRM advantage alone justifies the longer duration, before considering the arbitrage revenue uplift.
06
Full ROI Comparison
Every scenario tested C2
ROI & Payback Across All Scenarios
| Scenario | 2h ROI | 4h ROI | 2h Payback | 4h Payback | Winner |
|---|---|---|---|---|---|
| PF (all years) | 11.1% | 13.5% | 9.0 yrs | 7.4 yrs | 4h |
| PF (ex-crisis) | 9.7% | 11.7% | 10.3 yrs | 8.5 yrs | 4h |
| FS (all years) | 7.1% | 8.5% | 14.1 yrs | 11.8 yrs | 4h |
| FS (ex-crisis) | 6.1% | 7.3% | 16.5 yrs | 13.7 yrs | 4h |
| PF + CRM (all years) | 18.5% | 22.3% | — | — | 4h |
| PF + CRM (ex-crisis) | 17.1% | 20.6% | — | — | 4h |
| FS + CRM (all years) | 14.4% | 17.3% | — | — | 4h |
| FS + CRM (ex-crisis) | 13.4% | 16.2% | — | — | 4h |
PF = Perfect Foresight (LP upper bound). FS = Fixed Schedule (zero-intelligence lower bound). "Ex-crisis" excludes 2021–22 elevated prices. CRM at EUR 149,960/MW de-rated clearing price. ROI = annual revenue / total CAPEX.
ROI Visual Comparison Arbitrage + CRM scenarios
Verdict: 4-hour wins in every scenario tested — all years, ex-crisis, perfect foresight, fixed schedule, with and without CRM. The ROI advantage ranges from 1.2 to 3.8 percentage points across scenarios. This analysis is based on 2,695 backtest days (64,670 hourly SEM prices) from October 2018 to February 2026.
Data sources: SEM hourly DAM prices (Oct 2018 – Feb 2026, 64,670 price points). CAPEX from bottom-up build model (CAPEX-BUILD-UP.md). CRM de-rating from SEM Committee methodology. All revenue figures are per MW installed, before OPEX and degradation. Every number derived from backtest simulation against actual SEM data. C1 = underlying data, C2 = derived metrics.