You sit with the traders and work out what electricity will cost tomorrow, next week and next winter. You take in weather forecasts, power plant outages, gas prices and flows across borders, and turn them into a forecast: which plants will run, and what the price will be in each quarter-hour of tomorrow. Then the trader asks the question that matters: where is the market's price wrong, and why? Your answer is called a view. It is a number and a reason, and you are judged on whether the desk made money on it.
Not the same job: consultancy analysts (Aurora, Cornwall Insight) write reports for clients, regulatory analysts read the rules, data analysts build dashboards. This guide is about the analyst whose forecast gets traded.
Why it is in demand right now
The price now follows the weather. Germany's power price was below zero for 575 hours in 2025, because wind and solar made more power than anyone needed. When the wind drops, the swing goes the other way: on 8 September 2026 Great Britain paid a record £33.6m in one day to keep the grid balanced, mostly to gas plants asked to switch on. Someone has to say, the day before, which of those two days tomorrow is.
There are four times as many prices to call. Since October 2025 the European day-ahead market sets 96 prices a day, one per quarter-hour, instead of 24. Trading is also moving closer to delivery: EEX intraday volumes were up 39% year on year in August 2026. A forecast now has to be finer, and updated during the day.
Gas drives the price again. The European gas price (TTF) rose from about €29 per MWh in January 2026 to over €74 in early September, and Germany started September with its gas storage 53% full, a record low. Gas plants set the power price most hours, so every power analyst is half a gas analyst.
Who is hiring. This year: ExxonMobil, Shell and SSE in London; Vattenfall, EP Commodities, DXT and Modo Energy in Europe.
What it pays (London, 2026)
| Role | Pay |
|---|---|
| Analyst, 2 to 5 years | £55k to £90k |
| Senior, desk-side, base | £90k to £165k |
| Head of Fundamentals | £150k to £250k |
| Bonus, trading house | 30% to 100% |
Indicative. Anchored on a September 2026 London posting at £55k to £65k, a December 2025 Senior Power Trading Analyst posting at £125k to £165k, and 2026 recruiter guides. Trading houses pay top, utilities and vendors below; Germany and the Nordics 10% to 20% lower.
What the interview really tests
- Tomorrow's evening price in Germany is €95. Too high, too low, or about right? The reason, not a number.
- The wind forecast drops 10 GW after the day-ahead auction. What happens in intraday, and at the gas plants?
- A model a trader actually used: what did it get wrong, what did you change?
- Where does your data come from, and what do you do on the morning it is late?
Where it goes from here, honestly
Net positive, with a catch. The tailwinds are real: weather-driven price swings are here to stay, batteries are a new kind of player to model, and gas is volatile again. The catch: the analyst being hired is not the analyst of 2019. Employers now call Python and SQL "essential" even for juniors (DXT) and ask for hands-on use of AI tools (Modo). Recruiters say analysts who can model and code now "shape strategy" while traders execute. The analyst who describes yesterday's price in a spreadsheet and a morning note is being replaced, partly by software that writes the morning note.
Put together: the job is safe for the person who owns a model the desk trusts and can say where the market is wrong and why. It is not safe for the commentator. The rest of this guide is for the first kind.
The roadmap
Four stages, in order. Each stage names the guides and projects to work through in the Energy Trading & ETRM Skills library, roughly how long it takes, and the thing you will have at the end that you can show someone. Guides are textbook-style PDFs; projects are code you build.
| Stage | Time |
|---|---|
| 1. Read the machine | about 20 hours |
| 2. Read the markets | about 25 hours |
| 3. Build the model | about 40 hours |
| 4. Own the view | about 20 hours |
Around 105 hours in total, or nine weeks at twelve hours a week.
Stage 1: Read the machine
Energy Trading Fundamentals course. The cheapest plants run first, and the most expensive plant still needed sets the price. Learn that machine before the markets.
- Forming a View on Power Prices
- Physical Foundations
- The Global Gas Market
- The Spark Spread and Power Plant Economics
- Interconnectors and Carbon
What you leave behind: a hand-drawn merit order for one country and one hour, naming the plant that set the price, and one page on why yesterday's evening price was what it was.
Stage 2: Read the markets
Fundamentals and Practitioner Track. Where the prices come from, in what order during the day, and what the data looks like when it lands on your desk.
- Day-Ahead and Intraday Power Markets
- Forward Curves and Market Data
- Market Microstructure and Order Books
- Market Data Operations
- Cross-Border Capacity Trading and FTRs
What you leave behind: one page on a day when the intraday price moved away from the day-ahead price, and the forecast change that caused it.
Stage 3: Build the model
Python for Commodities, the Systems Builder Track and one BESS guide. A merit-order model in code, on public data, tested against what actually happened. The thing a hiring manager will open.
- Writing Professional Python
- Project 1: Price Curve Simulator; phase 7 is the merit-order dispatch
- Production Python for Trading Systems
- Dispatch Optimisation (BESS course): a battery in the stack
What you leave behind: a public repository: a merit-order model for one market on ENTSO-E data, giving 96 prices for tomorrow and showing how far off it was over thirty days.
Stage 4: Own the view
Practitioner Track and BESS course. What the forecast is for, what it costs when wrong, how batteries bid, and the one rule an analyst who sees outage data must never break.
- Power Trading Strategies
- Risk Management Fundamentals
- Revenue Markets and Revenue Stacking, BESS course
- Energy Market Regulation and Compliance, the REMIT inside-information chapter
What you leave behind: four weekly one-page views: what the market said, where you disagreed, why, and what happened. Keep the wrong ones.
If you already work as an analyst: skip the Stage 1 guides you could teach, but not the model or the four views. Desks hire the record, not the CV.
The first thirty days
The roadmap is nine weeks. The first four decide whether you finish. Twelve hours a week, split roughly two thirds reading and one third building.
| Week | Do | Done means |
|---|---|---|
| 1 | Forming a View. Physical Foundations. Start Project 1 (phases 1 to 3). | You can name the type of plant that set yesterday's German price, any hour. |
| 2 | The Global Gas Market. Spark Spread. Interconnectors and Carbon. Project 1, phases 4 to 7 (the merit order). | Your simulator runs a merit order. You can work out a gas plant's margin by hand. |
| 3 | Day-Ahead and Intraday. Forward Curves and Market Data. Market Data Operations. | Your one-page write-up of a day when intraday moved away from day-ahead is done. |
| 4 | Writing Professional Python. Start the merit-order model on ENTSO-E data for one country. First backtest. | Repository public. 96 prices for tomorrow and a chart of the error over thirty days. First view written. |
Weeks five to nine: harden the model, add the battery, read Strategies, Risk and the REMIT chapter, write three more views.
Prove it: what to put in front of a hiring manager
Not a certificate. Five things they can open.
- A merit order drawn from memory. One market, one hour, the plant that set the price, and why.
- The model on GitHub. Public data in, 96 prices out, thirty days of results, the error stated honestly.
- Four weekly views. What the market said, where you disagreed, why, what happened. Keep the wrong ones.
- One day explained. Day-ahead, intraday and imbalance prices, and what the weather forecast did in between.
- A data map. Every source the model uses (ENTSO-E, Elexon, ECMWF, AGSI+), what arrives late or wrong, and what you do then.
Take those five to an interview and the conversation changes. Nobody asks whether you understand the market. They ask what your view is for tomorrow.
What a power market analyst learns on the courses, and where to start.