What Realized Value per MWh Actually Measures

Most renewable portfolio reports answer two questions well: how much did we produce, and what was the average price. Almost none of them answer the question the owner actually cares about. After every market interaction was settled, how much money did one megawatt-hour of production turn into?

Realized value per MWh is that number.

Definition

Realized value per MWh is the net financial result of a portfolio over a period, across every revenue and cost channel it touches, divided by the energy produced in that period.

Written out:

realized value per MWh =
  ( day-ahead revenue
  + intraday trading result
  + balancing and settlement outcomes
  + contract settlement effects (PPA, profile obligations)
  + capacity or ancillary service revenue, where applicable
  - imbalance costs
  - market fees and charges )
  / MWh produced

Two definitional choices matter and should always be stated explicitly.

First, the denominator. Dividing by produced energy answers “what did each produced MWh earn.” Dividing by potential production (produced plus curtailed volume) answers a different question, because it also prices the decision to curtail. Both are legitimate. A curtailment during a deeply negative price block can raise realized value on the first definition and needs the second definition to be judged fairly. Pick one, write it down, keep it fixed.

Second, the channel list. If intraday results are booked by a trading desk under a separate P&L, they still belong in the portfolio’s realized value. The whole point of the metric is that it refuses to stop at any single team’s boundary.

How it differs from capture price

Capture price is the production-weighted average of the day-ahead price. It tells you what the market paid for your production profile, and it stops there. Everything that happens after day-ahead gate closure is invisible to it: intraday repositioning, balancing settlement, imbalance costs, contract effects, fees.

Two portfolios can post an identical capture price and end the month with meaningfully different bank balances. Capture price is mostly a statement about market conditions and production shape. Realized value per MWh is a statement about the entire decision chain, which makes it a management metric. When the two diverge, the divergence itself is information: it tells you how much the chain after day-ahead added or destroyed.

How it differs from forecast accuracy

Forecast accuracy measures the quality of an input. Realized value measures an outcome. The mapping between them is weaker and more asymmetric than most owners assume.

A 10 percent production error in an hour where the spread between the day-ahead price and the imbalance price is 5 EUR costs very little. The same error during a price spike or a negative-price block can cost more than a whole week of small errors combined. What matters financially is where the errors land, and how the positions around them were managed.

Working with large renewable portfolios has shown that improvements in headline forecast error often produce no measurable change in financial results. In simulation work we have also seen the reverse pattern as a hypothesis result: a model with better average accuracy producing worse portfolio economics, because its remaining errors concentrated in expensive hours. Forecast accuracy is worth improving, but as a KPI for portfolio optimization it can move in the wrong direction relative to money.

A worked example

The numbers below are invented for illustration. They describe no real portfolio or client.

Take a 100 MW wind portfolio over one month, producing 22,000 MWh.

Channel	Amount
Day-ahead sales (production-weighted 82.00 EUR/MWh)	+1,804,000 EUR
Imbalance costs from forecast errors	-66,000 EUR
Intraday repositioning result	+18,000 EUR
Market fees and settlement charges	-11,000 EUR
Net result	1,745,000 EUR

Capture price: 82.00 EUR/MWh.

Realized value per MWh: 1,745,000 / 22,000 = 79.32 EUR/MWh.

The gap is 2.68 EUR/MWh. On this portfolio that is roughly 59,000 EUR in a single month, and on the order of 700,000 EUR over a year if conditions persisted. A report that shows capture price and availability will never surface it, because each of the component lines lives in a different system and a different team’s spreadsheet.

The gap also runs in both directions. A portfolio that trades intraday well, uses flexibility, and manages its positions through volatile hours can realize more than its capture price. The sign and size of the gap is precisely what portfolio optimization is about.

Why portfolio owners should track it

The practical reasons, in rough order of importance:

• It is the KPI that lines up with the owner’s P&L. Production volume, availability, capture price and forecast error are all partial views. This one is denominated in the same units the annual report is.
• It makes optimization claims testable. Agree a baseline period and a measurement definition, then measure the change in realized value per MWh against it. Without this, every vendor claim and every internal initiative is graded on its own preferred metric.
• Decomposed by channel, it shows where the leakage is. A widening gap driven by imbalance costs points somewhere different than one driven by fees or by weak intraday execution.
• It travels across route-to-market setups. Merchant, PPA-covered and hybrid structures can all be compared, provided the channel list and denominator are defined consistently.

Two caveats. Comparing realized value across months requires some normalization, because spreads and volatility differ between periods; a raw month-on-month change mixes market conditions with management quality. And comparing across bidding zones without adjusting for local price levels mostly measures geography.

None of this is exotic. It is bookkeeping applied where bookkeeping usually stops, at the boundary between production reporting and trading reporting.

If your monthly portfolio report shows capture price and availability but no realized value per MWh, which number currently tells you whether the chain from forecast to settlement added value last month or quietly gave some back?