Position Aggregation and Delta Exposure
Managing Risk Across Complex Trading Books
FUNDAMENTALSBefore you can trade profitably, manage risk, or calculate P&L, you need to answer a deceptively simple question: What do I have?
Not approximately. Not "mostly gas." Exactly what you own, what you owe, when it delivers, and whether it's hedged.
Position management is the foundation beneath everything else in energy trading. Get this wrong, and every calculation downstream - P&L, VaR, margin requirements, delivery nominations - is wrong too. Energy trading risk management shows how a desk sets limits on these positions.
This article explains how positions work from first principles: what a position is, the one rule for adding trades together, how physical and financial trades are kept apart, and what delta exposure and hedge coverage tell you. It follows one small book of seven trades all the way through, and ends with the same calculation in a few lines of Python.
What is a position?
A position is your net obligation to buy or sell a commodity. If you have bought 300 MWh of gas and sold 120 MWh, your position is 180 MWh. You are long: you hold more than you owe, and you gain if the price rises. If you had sold more than you bought you would be short, and you would gain if the price falls.
Energy positions differ from positions in shares in one way that matters. A megawatt-hour is not just a megawatt-hour. It is gas or power, at a named hub, for delivery in a named period. Gas for January cannot be used to meet a sale for February. Gas at the British hub does not cancel gas sold at the Dutch one.
Signed quantities
To add trades together, turn "buy" and "sell" into numbers. A buy is positive and a sell is negative. A purchase of 300 MWh becomes +300. A sale of 120 MWh becomes -120. Once every trade has a signed quantity, working out a position is addition.
The aggregation rule
Trades may be added together only when three things match: the commodity, the delivery period and the book. A book is a named set of trades kept for one purpose. Here there are two: a physical book for trades that end in delivery, and a hedge book for financial trades that lock in a price.
Take a small desk with seven trades.
| Trade | Book | Commodity | Delivery | Action | Signed quantity (MWh) |
|---|---|---|---|---|---|
| 1 | Physical | NBP gas | January | Buy 300 | +300 |
| 2 | Physical | NBP gas | January | Sell 120 | -120 |
| 3 | Hedge | NBP gas | January | Sell 150 | -150 |
| 4 | Physical | NBP gas | February | Buy 200 | +200 |
| 5 | Hedge | NBP gas | February | Sell 50 | -50 |
| 6 | Physical | TTF gas | January | Sell 80 | -80 |
| 7 | Physical | GB power | January | Sell 40 | -40 |
NBP is the British gas hub and TTF is the Dutch one. Group the trades by commodity, delivery period and book, and add the signed quantities in each group. Only trades 1 and 2 share all three, so only they combine: +300 - 120 = +180.
| Book | Commodity | Delivery | Net position (MWh) |
|---|---|---|---|
| Physical | NBP gas | January | +180 |
| Hedge | NBP gas | January | -150 |
| Physical | NBP gas | February | +200 |
| Hedge | NBP gas | February | -50 |
| Physical | TTF gas | January | -80 |
| Physical | GB power | January | -40 |
That table is a position report. Whether a system holds seven trades or seventy thousand, the logic is the same: group by the three keys, add the signed quantities.
Why one total misleads
Add all seven signed quantities and the answer is +60 MWh. The book looks almost flat. It is not. It is long 150 MWh of February gas, short 80 MWh of Dutch gas and short 40 MWh of power. Those are three separate risks, at three separate prices, and none of them offsets another.
This is the commonest way a position report goes wrong. The total is arithmetically correct and tells you nothing. Each key in the rule is there to stop a different false offset: the commodity keeps hubs and fuels apart, the delivery period keeps months apart, and the book keeps delivery apart from price protection.
Physical and financial positions
A physical trade ends in delivery. Gas flows through a pipe, or power across the grid. A financial trade is settled in cash against a reference price, and nothing is delivered. Futures and swaps are the usual examples. Physical and financial energy trading covers the difference in full.
The two books answer different questions.
- The physical book: can we meet our deliveries? For January the desk must take 180 MWh of gas at NBP, net, and find 80 MWh at TTF and 40 MWh of power to deliver.
- The hedge book: how much of our price risk have we locked in? The desk has sold 150 MWh of January gas forward, and 50 MWh of February.
Both are measured in megawatt-hours. A hedge position of -150 means "price locked in on 150 MWh", not "£150".
What is delta exposure?
Delta exposure is the volume still exposed to price once the hedges are counted. For one commodity and one delivery period:
Delta exposure = physical position + financial position
For January NBP gas that is +180 - 150 = +30 MWh. For February it is +200 - 50 = +150 MWh. The Dutch gas and the power have no hedges, so their delta is the physical position itself: -80 and -40.
Delta turns a position into money. A delta of +150 MWh means that if the February price rises by £2 per MWh, the book gains 150 x £2 = £300. If the price falls by £2, it loses £300. The sign tells you which way you are exposed, whether you meant to be or not.
The word comes from options, where delta measures how much a contract's value moves when the price moves by one. For a forward, a future or a swap that number is one, so the exposure is simply the volume. Options change that, and they are a subject for later.
Hedge coverage
Hedge coverage is the share of a physical position that financial hedges cover.
- January NBP gas: 150 hedged against 180 physical, which is 83%.
- February NBP gas: 50 hedged against 200 physical, which is 25%.
A firm's risk policy is usually written in these terms. If the policy says 80% of the coming month must be hedged, January passes at 83%. February, at 25%, shows where the open exposure sits. "Adequately hedged" becomes a number that can be checked.
The same calculation in Python
The whole article fits in a few lines of pandas. This is the core of every position report, in a spreadsheet or a trading system.
import pandas as pd
trades = pd.DataFrame(
[
("Physical", "NBP gas", "Jan", "Buy", 300),
("Physical", "NBP gas", "Jan", "Sell", 120),
("Hedge", "NBP gas", "Jan", "Sell", 150),
("Physical", "NBP gas", "Feb", "Buy", 200),
("Hedge", "NBP gas", "Feb", "Sell", 50),
("Physical", "TTF gas", "Jan", "Sell", 80),
("Physical", "GB power", "Jan", "Sell", 40),
],
columns=["book", "commodity", "delivery", "action", "mwh"],
)
# Buy is positive, sell is negative
sign = trades["action"].map({"Buy": 1, "Sell": -1})
trades["signed"] = trades["mwh"] * sign
# The aggregation rule: commodity, delivery period, book
positions = trades.groupby(["commodity", "delivery", "book"])["signed"].sum()
# Delta exposure: physical plus financial, for each commodity and period
delta = trades.groupby(["commodity", "delivery"])["signed"].sum()
print(positions)
print(delta)
The last line prints +30 for January NBP gas, +150 for February, -80 for TTF and -40 for power: the figures worked out by hand above.
Can an ETRM system calculate position exposure?
Yes. It is the first thing an energy trading and risk management (ETRM) system is for, and What is an ETRM system? explains where it sits. The system applies the rule above to every trade, usually at a finer grain: each day, or each half-hour for power.
It is still worth being able to do it by hand. A position report is only as good as the tags on the trades. A trade booked to the wrong book, the wrong hub or the wrong month produces a report that adds up and is wrong. The people who catch that are the ones who can rebuild a position from the trade list.
Building or replacing this on your own desk? It is the work A115 does. See the systems work
Common mistakes
- Netting across hubs. NBP and TTF are both gas, and they are separate positions at separate prices.
- Ignoring the delivery period. A long position in February does not cover a short one in January.
- Mixing units. British gas is quoted in therms and power in megawatts over a period. Convert everything to megawatt-hours before adding. The example here is already in megawatt-hours.
- Counting a trade twice, or not at all. A trade loaded from two sources, or missing from one, moves the position without anyone trading.
How This Fits the Curriculum
This guide is foundational. You cannot calculate P&L without positions. You cannot optimize spark spreads without knowing your generation positions. You cannot manage storage without tracking injection and withdrawal positions. Every subsequent topic assumes you understand position mechanics. This is where the mathematics of energy trading begins.
Prerequisites
Complete the Physical Foundations guide first. You need to understand commodity types (gas vs power), delivery periods (day-ahead vs forward), and why physical delivery obligations matter.
Mastery Tip
Recreate the seven trades above in Excel or Python. Start with trade data (Trade ID, Action, Commodity, Quantity, Delivery, Book). Apply signed quantities. Aggregate by the three-key rule. Split physical from financial. Calculate delta and hedge ratios. Then modify one trade - change a Buy to a Sell, or move a trade from Physical to Financial - and watch how delta and hedge ratios respond. The sensitivity will become intuitive, not theoretical. When you can predict the impact before running the calculation, you've internalized position mechanics.
What the full guide adds
This article covers the rule and one small book. The full guide goes further:
- a reference portfolio carried through every step, from the trade list to mark-to-market P&L;
- hedge coverage measured against a firm's natural exposure, and what the difference tells a risk manager;
- positions across several commodities, and the basis risk that the physical and financial split reveals;
- how to organise books for analysis and for operations, and when to add across them;
- the full workflow: load the trades, sign them, aggregate, measure exposure, join market prices and value the book;
- the pitfalls that break real position reports, each with an example, and a checklist for an accurate one;
- exercises with answers.
The full guide is in the community on Skool, with the rest of the series, and it is updated regularly as the markets change. If you want to work on a trading desk, read it to understand the subject properly. To join, start at the Energy Trading Fundamentals course page.
What You'll Learn
- ➕ Signed Quantities: The foundational concept - Buy = positive, Sell = negative. Converts every trade into a number you can add.
- 🔑 The Three-Key Rule: Sum signed quantities grouped by Commodity + Delivery Period + Book. The universal aggregation rule.
- 📊 Physical vs Financial Splits: Separate delivery obligations from price hedges - revealing operational risk vs market risk.
- 📈 Delta Exposure: Delta = Physical + Financial. Measure unhedged risk and know which way you're exposed to price moves.
- 🎯 Hedge Ratios: Calculate what percentage of physical exposure is hedged. Turn "adequately hedged" into precise metrics.
- 💻 Python Implementation: Build a position engine - the foundation every trading system uses.
Energy Trading Fundamentals Series
A structured curriculum for mastering energy markets - from physical infrastructure to P&L calculations
Prerequisites:
- → Physical Foundations of Energy Commodities
- → How Power Trading Works
- → The Global Gas Market
- → How Gas Trading Works in Great Britain
- → LNG Pricing Explained
- → Spark Spread & Power Plant Economics
- → Dark Spread & Coal Power Plant Economics
Next in Series:
Daily valuation, unrealized gains/losses, risk reporting
Related Guides:
Desks pay £850 a day for developers who understand this.
You can now aggregate a position and read its delta. The next thing a desk asks is where the price for each delivery period comes from, and how to build the curve.
£850 is the UK contract median for energy trading developers (ITJobsWatch, 2026), and analysts and business analysts who understand the trade are hired for the same reason. The free Fundamentals path takes about a week part-time. If you have a date, an interview, a start, a go-live, say so when you join and Jordan Dimov will give you the reading order. On Premium, Jordan reviews your code and your CV.
Day rates: IT Jobs Watch, UK contracts, six months to 9 November 2026.
"Jordan's in-depth explainers on topics ranging from interconnector flows to viewing gas-fired generators through option frameworks helped me get my role as an analyst on a power and gas trading desk." Tom R.
Aiming for a role? Quant developer · ETRM business analyst · Power market analyst · Trading technology teams
Training a team? VIP is one invoice, with seats you can reassign. See the VIP plan
Your employer can pay by invoice. How to ask