An LNG book is a calendar before it is a number. The tradable unit is a whole cargo tied to a loading window, a ship and a discharge slot - which is why two cargoes of the same size are not substitutes, why the quantity discharged is smaller than the quantity loaded by design, and why the most expensive decisions on an LNG desk expire on a date rather than at a price.
Most physical commodity books are held as quantities. You are long so much crude, so much copper, so much gas, and the operational question of how it gets where it is going is downstream of the position rather than part of it. An LNG book does not work that way. The tradable unit is a whole cargo, attached to a loading window at one end and a discharge window at the other, carried in a ship that is itself a scarce contracted thing, and evaporating steadily the entire time it is in transit. The energy on the invoice is a consequence of all of that rather than an input to it.
Once you see the cargo rather than the volume as the unit, a set of things that look like operational annoyances turn out to be the structure of the business. It is why two cargoes of the same size are not substitutes for each other, why the quantity discharged is smaller than the quantity loaded by design rather than by error, why the price index a cargo settles against can still be undecided while the cargo is at sea, why a delay costs you the commodity and not only the freight, and why the most expensive decisions on an LNG desk are dated rather than priced — they expire on a notice deadline that has nothing to do with the market.
This article sets out what you are actually holding when you are long a cargo, which of the things you hold are separate contracts with separate counterparties, where the delivered quantity comes from and who computes it, and what an LNG trade line has to state before it can be hedged, settled or reported honestly. It is written for the people who have to make those things agree — chartering and scheduling, the middle office, settlements, and traders who have to explain why a position that netted flat could not actually be delivered.
If you want the mechanism from the beginning, start with A cargo is indivisible and dated. If you already run cargoes daily and want the part that most often breaks the reporting, start with One cargo is a stack of contracts, not a line.
A cargo is indivisible and dated
The tradable unit in LNG is a whole parcel occupying a specific window, not a quantity of energy. A book of cargoes is a calendar before it is a number, and two cargoes of the same size in different windows are not substitutes for each other.
A pipeline gas position can be shaded. You can sell a bit more, buy a bit back, shift the shape within the month, and the increments are as small as you want them to be, because what moves is a rate of flow into a shared network. An LNG chain does not move increments. It moves ship-sized parcels, and the whole apparatus at both ends — the berth, the loading window, the vessel, the discharge slot — is sized and scheduled around a parcel arriving whole.
The consequence is that your flexibility is not in how much. It is in which window, which ship, which destination, and whether the cargo happens at all. Those are the levers, and none of them is continuous. You cannot deliver a partial cargo into a slot; you either use the slot or you do not.
This is why an LNG position expressed purely in energy units is misleading in a specific way. Adding two cargoes together produces a number that is arithmetically correct and commercially meaningless, because a cargo lifting in one window and a cargo lifting in another are not interchangeable — they need different ships, they reach different markets, they settle against different prices, and one of them may be deliverable while the other is not. The netting that a position report performs quietly assumes fungibility that the physical chain does not provide.
The cargo consumes itself while it travels
LNG stays liquid by boiling. Some of the cargo turns to vapour continuously in transit, and the vapour is either burned to move the ship or handled at a cost — so the quantity that arrives is structurally smaller than the quantity that left, and the shortfall is a function of how long the voyage takes.
Nothing keeps a cargo cold by refrigeration alone during a voyage. It is kept cold by allowing part of it to evaporate, which carries heat away and holds the rest of the liquid at temperature. The vapour that comes off is not waste in the ordinary sense: on most ships it is usable as fuel, which means the voyage is partly paid for in the cargo itself.
Two things follow that do not follow anywhere else in the physical book.
The first is that loaded quantity and discharged quantity are two different facts about the same cargo, and the gap between them is not a measurement error to be reconciled away. It is the voyage. Which of the two quantities the sale prices against is a contract term, and it decides who owns the difference — which is to say that the same physical cargo, sold on two different delivery terms, produces two different invoices for reasons that have nothing to do with the market.
The second is that time is denominated in commodity. A ship waiting at anchor for a berth is not only running up a demurrage claim; it is also still boiling. A slow steam to save fuel is a decision that trades one cost against another inside the same cargo. A diversion that adds days to the passage costs freight and cargo together. On most desks, the freight side of that calculation sits with chartering and the cargo side sits with trading, and the two are frequently added up after the fact by whoever is asked to explain the variance.
The delivered quantity is a calculation, not a reading
Nobody meters an LNG cargo the way a pipeline is metered. The quantity is computed from measurements taken in the ship's own tanks and from an analysis of what the liquid is, using a procedure written into the contract — which makes the procedure itself a negotiated term and makes a quantity dispute a dispute about an input to a formula.
Custody transfer happens by survey. The ship's tanks are gauged before and after transfer; temperature, pressure and the vapour space are recorded; a sample is analysed to establish the composition of the liquid and therefore its density and its energy content; and the delivered energy is calculated from those inputs. Both sides typically have the right to be represented while it happens, and the method by which it happens is agreed in advance rather than imposed by the network.
That is a genuinely different arrangement from the one that governs piped gas, and the difference matters for how disputes work. In a network, the delivered quantity is determined by someone who is not your counterparty, and you receive it as a statement. In LNG, the delivered quantity is produced jointly, from a procedure the two of you agreed to, on the cargo itself. There is no third party to appeal to and no allocation to wait for — but there is a formula, and every input to it is contestable.
The practical failure this creates is quiet. A trading record that stores the final energy figure and nothing else has discarded the part of the record a dispute is actually about. When a counterparty questions a cargo months later, the question is never "was your number wrong"; it is "which gauge reading, which analysis, which tables, whose representative signed". A book that cannot produce those without going back to the ship's paperwork is a book that will concede things it did not need to concede.
Destination is a variable, so the price can be decided after the cargo exists
A cargo with destination flexibility is an option that stays live while the cargo is moving. Which market it goes to determines which price it settles against, which hedge is the correct hedge, and which slot at which terminal gets used — and that determination can be made after loading.
For a fixed-destination sale, none of this arises: the cargo goes where the contract says, and the price is whatever the contract's formula produces. Where the sale is flexible, the cargo becomes a position on the difference between delivered markets, net of what it costs to reach each of them.
The word "net" is carrying real weight there. The spread between two delivered prices is not the value of a diversion. The value of the diversion is that spread less the additional freight, less the additional cargo consumed on the longer passage, less whatever the abandoned slot costs you, and subject to whether a ship and a berth are actually available in the window. A desk that watches the headline spread and not the deliverable spread will see opportunities that do not exist and miss the constraint that made the spread appear in the first place.
Two reporting consequences follow. A hedge placed at load can become the wrong hedge without any price moving, simply because the cargo changed destination and the index it settles against changed with it. And a mark on a flexible cargo is a mark on an option, not on a fixed sale — a book that carries it at one market's price is carrying an assumption as though it were a term.
One cargo is a stack of contracts, not a line
Being long a cargo usually means holding several separate agreements at once — a right to lift it, a ship to carry it, a slot to discharge it, and a sale to deliver it into — each with a different counterparty and a different deadline. The deadlines expire on dates, not on prices, and they expire whether or not anyone is watching.
The single line on the position report is an abstraction over a stack. Underneath it there is typically an offtake or tolling arrangement that gives you the right — and often the obligation — to lift in a window, with a nomination process running well ahead of it and, in many arrangements, a right to decline the lift at a stated cost. There is a shipping arrangement, which may be a vessel you own, one you have on charter, or one you have not secured yet. There is capacity at the receiving end: a berth, a discharge window, and regasification or storage that is itself contracted and scheduled. And there is the sale.
Each element has its own calendar, and most of them have a use-it-or-lose-it property. Capacity you hold and do not use is generally not refunded. A cancellation right that is not exercised by its notice date stops being a right and becomes an obligation. A window you miss is not rescheduled to suit you. This is the sense in which LNG decisions are dated: the deadline for deciding whether to cancel arrives on a day set by a contract, and the information you would like to have before deciding often arrives afterwards.
What makes this hard to see in a trading system is that the stack nets to something that looks simple. A book that reports an energy position can show you flat when you are in fact long a cargo you have no ship for, holding a discharge slot for a cargo you have already cancelled, and carrying a notice deadline that lapses next week. None of those is a price exposure. All of them are money.
The decisions that set a cargo's profit sit in different teams' files
Every mechanism above lands in the same short list of activities: nomination and scheduling, shipping and vessel operations, inventory and the quantity-and-quality determination, demurrage and laytime, invoice checking, and reconciliation across systems that were never designed to agree. Those are the steps where the working system of record is commonly a spreadsheet and a mailbox, and where the trading system is the place a number gets booked once people have already agreed on it.
That much is not unique to LNG. What is unique is which of those steps carries the commercial outcome. In a book of pipeline gas, the back office is where the quantity is established. In a book of cargoes, the quantity is established at the ship's rail — but the deliverability is established in scheduling, the price can be established in chartering, and the option value is destroyed or realised by whoever does or does not remember a notice date. The decisions that decide an LNG cargo's profit are distributed across teams that report to different people and keep their records in different files.
Four parts of a cargo have an owner, and the part in transit does not
| Right to lift | Shipping | The cargo in transit | Discharge capacity | The sale | |
|---|---|---|---|---|---|
| What you hold | A window and a quantity you may or must lift | A vessel owned, chartered, or still to be found | A quantity that is changing while you hold it | A berth, a slot, and space to put it in | An obligation to deliver into a market |
| Who it is with | The liquefaction or offtake counterparty | The owner or the charter market | Nobody — it is physics under a contract term | The terminal or capacity holder | The buyer |
| What the deadline is about | Nominating, or declining to nominate | Fixing a ship for the window | Voyage time, and time spent waiting | Using the slot or losing it | Delivery window and documents |
| How it goes wrong | The cancellation right lapses unexercised | The cargo exists and the ship does not | The shortfall is treated as a discrepancy rather than the voyage | Held for a cargo that was cancelled | Hedged against the index the cargo no longer settles on |
| Where the fact first exists | Nomination correspondence and the offtake schedule | The fixture and the charter | The ship's log and the surveyor's report | The terminal's schedule | The trade capture record |
The column worth sitting with is the middle one. Four of the five have a counterparty, a document, and someone whose job it is to watch them. The cargo in transit has none of those, and it is the one that is changing.
An LNG trade line has to state nine things before it can be settled or hedged
| The line states | Which question it answers | What happens when it is missing |
|---|---|---|
| Load window and discharge window, as dates rather than a month | Whether this cargo is deliverable at all | Two non-substitutable cargoes netted against each other |
| Loaded and discharged quantity as separate fields, and which one prices | Who owns the voyage difference | A shortfall investigated as an error rather than read as the voyage |
| Delivery term, and what it makes the seller responsible for | Where title, risk and cost actually sit | The wrong party chasing the wrong claim |
| The measurement procedure and who witnessed it | What a dispute would be argued from | A final number with nothing behind it |
| Price basis, and the destination it depends on | Which index this settles against | A hedge against an index the cargo left behind |
| Shipping arrangement, and whether it is secured | Whether the position can move | A flat book that cannot be delivered flat |
| Discharge capacity held, and its window | Whether it can land where it is going | A cargo afloat with nowhere to go |
| Diversion and cancellation rights, with their notice dates | What you can still decide, and until when | Optionality that expires without anyone deciding |
| Demurrage exposure at both ends | What waiting is costing while it happens | A claim that first appears as an invoice |
None of these fields is difficult on its own. What makes the line hard to assemble is that its fields are owned by different desks and arrive at different times — trading holds the first and the fifth, chartering the sixth and the ninth, scheduling the seventh, operations and the surveyor the second and the fourth — and that a cargo can look completely booked while the fields that decide whether it is deliverable are held somewhere else entirely.
Questions people ask about this
How is LNG trading different from trading pipeline natural gas?
The unit is different, and almost everything else follows from that. Pipeline gas is a rate of flow into a shared network, so it is continuous, divisible, and its delivered quantity is determined by the network after the fact. LNG is a discrete parcel that has to be lifted whole, carried in a specific ship, and landed in a specific window — so its quantity is determined at the ship by an agreed procedure, and its real constraints are calendar constraints rather than volumetric ones. A desk that runs both is running two different businesses that happen to price off related curves.
Why is the quantity we are invoiced for smaller than the quantity we loaded?
Because part of the cargo evaporates in transit, and on most voyages the vapour is burned to move the ship. That is how the cargo stays cold; it is not a leak and it is not a measurement problem. What varies between contracts is which quantity the sale prices against, and therefore who absorbs the difference. If the shortfall on your cargoes is being opened as a discrepancy every time, the problem is not the shortfall — it is that the trade record holds one quantity field where it needs two, and nobody downstream can tell which one they are looking at.
What does it actually mean to divert a cargo, and when can we do it?
It means sending the cargo to a different discharge point than the one originally intended, which is possible only if the sale contract permits it and if a berth, a window and enough voyage time exist to make it work. The commercial point is that diverting changes which price the cargo settles against. The trap is valuing the diversion on the headline difference between two delivered markets: the number that matters is that difference less the extra freight, less the extra cargo consumed on the longer voyage, less the cost of whatever slot you abandon, and only if a ship and a berth are actually available. A spread that looks unusually wide is often wide precisely because moving cargo between those two points has become hard.
Who decides how much LNG was delivered?
The two parties do, jointly, through a procedure their contract specifies. The ship's tanks are gauged before and after transfer, the liquid is sampled and analysed to establish what it is, and the delivered energy is calculated from those inputs — normally with both sides represented while it happens. This is unlike a pipeline, where an operator determines your quantity and sends it to you. It means there is nobody to appeal to, and it means every input to the calculation is a thing that can be argued about later. Keep the inputs, not just the answer.
Do we need to own or charter a ship to trade LNG?
To move a cargo yourself, you need access to a vessel for that window — owned, on charter, or fixed in the market. To trade the price without moving anything, you do not. The distinction is worth being explicit about internally, because the two look similar on a position report and behave nothing alike: a paper position can be closed by trading, and a physical cargo with no ship against it is a problem that trading cannot solve inside the window. If your book contains both, the question "which of these can we actually deliver" should be answerable without asking three people.
Why does our LNG position look flat when we cannot actually deliver it?
Because netting in energy units assumes that a unit is a unit. A long cargo in one window and a short obligation in another window net to zero on a report and to nothing at all in reality — they need different ships, they reach different places, and the offset only exists if the physical chain can be made to connect them. The same effect hides missing links: the report cannot show that one leg has no vessel, or that the discharge slot belongs to a cargo that was cancelled. A position that is flat in energy and open in logistics is the normal way an LNG book goes wrong.
What is a cancellation right worth, and when do we have to decide?
It is an option, and like any option it is worth something up to the moment it expires and nothing afterwards. What is unusual about it is that the expiry is a notice date written into the lifting arrangement, not a market event — so it can pass unexercised simply because the deadline was in a contract that nobody opened this month. The information that would make the decision easy frequently arrives after the deadline has gone. That is uncomfortable but it is not a reason to avoid deciding; it is a reason for the deadline to be a visible field on the position rather than a line in a document.
Can we run an LNG book in spreadsheets?
Plenty are, and the tool is not really the issue. The specific thing a spreadsheet handles badly here is that one cargo is several contracts with several counterparties and several dates, and a row is a poor container for that. The failure mode to watch for is the one where the cargo is corrected in one file — a changed window, a diverted destination, a cancelled lift — and the other files continue to describe the cargo as it used to be. After that you have several internally consistent files that disagree with each other, and no record of which one changed. If you are on spreadsheets, the highest-value single addition is a field on every cargo saying which stage each of its numbers came from and when it last changed.
What should we fix first if our LNG settlements keep getting disputed?
Find out what the disputes are actually about before changing anything, because two very different problems present the same way. If they are about the delivered quantity, the fix is upstream of settlements: capture the measurement inputs and the witnessing record at the time of transfer, because a dispute is argued from those and not from the final figure. If they are about which quantity, which price basis or which window applies, the fix is that the trade record is holding one field where the business needs several, and settlements is reconstructing the distinction by hand each time. Automating the reconstruction is the wrong move for either problem — the first needs a record that does not currently exist, and the second needs the record to stop collapsing three facts into one.
Where this lands in a trading system
Nothing above is an argument for buying software. It is an argument about which facts have to exist, on whose record, and by when. Some of this article's conclusions do map onto capabilities that are published, and it is worth being specific about which ones — and about where the mapping stops.
The first conclusion is that a cargo is a stack of obligations that has to be held as one thing: a physical lifting, a shipping arrangement, a hedge and a settlement, related to each other rather than meeting at month end. That is the job of a CTRM/ETRM system providing complete trading lifecycle management for commodity traders, integrating physical trade, financial hedging, risk control, and settlement in one platform; Time Dynamics' Fusion is one, and the parts that bear on this article are complete physical trade lifecycle management and execution, end-to-end logistics coordination and execution management, real-time inventory tracking and optimization, and automated financial settlement and payment processing. The trades are recorded there. X-Ray is the other half of the same system. It unifies scattered data sources, provides one-click reports, risk alerts, and enables data-driven decision making, and the components named below are part of it.
The second conclusion is about dated obligations — the notice deadline that lapses, the slot that is held for a cargo that will not arrive, the cancellation right that expires while the decision sits with someone else. Watching a value against a threshold and raising something when it crosses is the shape of real-time risk alerts and monitoring, which is what X-Eagle does; its published description points that capability at trade finance field warnings, and whether it is pointed instead at a notice date approaching is a configuration question rather than something a platform decides for you. A deadline that has to be noticed before it passes is an ordinary shape, not an exotic one.
The third conclusion is that the facts which decide a cargo live outside any trading system. Nominations sit in scheduling correspondence, the fixture sits with chartering, the discharge window sits in a terminal's own schedule, and the measurement record arrives as a surveyor's document. That is what X-Ray addresses: a non-invasive data processing and analysis platform, whose collection toolkit XDK performs non-invasive automated data collection from databases, Excel files, and web interfaces, and which is built to collect data without disrupting existing systems or workflows. The last property is the relevant one, because the terminal's schedule is not going to be replaced to suit you and the chartering file is working.
Re-cutting a position by window, by vessel, by counterparty and by stage is the fourth, and one-click generation of personalized reports with Excel-like interface — X-Sheet — is that kind of work.
One conclusion of this article is deliberately not mapped. The destination-optionality argument — what a diversion is worth once freight, boil-off and the abandoned slot are taken out of it — does not correspond to anything on the published capability list. Valuing an option and optimising a voyage are not on it, and the honest thing to do is leave that conclusion unmapped rather than stretch a nearby capability across it.
And here is where the rest stops. None of this measures LNG. It does not gauge a tank, does not analyse a sample, does not decide a density or an energy content, and has no standing in a quantity dispute. It does not fix a ship, does not obtain a berth or a discharge window, and does not decide whether to divert, to cancel, or to wait — those are commercial decisions with money on both sides of them. It will not make a terminal give you a slot that is already taken. What a system can settle is narrower and duller than the subject of this article: which of a cargo's obligations you are actually holding, which of its numbers is a request and which is a fact, where each came from, when each changed, and whether the person who has to answer for the cargo can see all of that without asking four people in three teams. That is a large part of the difficulty in an LNG book. It is not the whole of it.
A note on the evidence in this article
There are no statistics in this article, and that is a decision rather than an oversight. The observations behind it — which steps in the trade lifecycle are still run out of spreadsheets and mailboxes, and how much of a cargo's outcome is decided in them — come from material we are not in a position to publish as a citable figure. A number we could not source would make the argument look stronger and would make it worth less, and it is exactly the part of an article most likely to be quoted onward by someone who never checks it.
So the argument is structural instead. Every claim here is made from the mechanism, and every one is stated in a form you can test against your own book rather than accept on our authority. The test worth running is the one implied throughout: take a cargo that settled recently and write down, side by side, its lifting right, its ship, its loaded and discharged quantities, its price basis, its discharge slot, and every notice date attached to it — together with where each of those lives and who would have noticed if it changed. The count of fields you cannot produce from a system is the finding. If your own records contradict what is written here, your records are the better evidence.
Practices also differ by contract and by market. Which quantity a sale prices against, who bears the voyage difference, how custody transfer is performed and witnessed, whether a cargo may be diverted and on what notice, and what happens to unused capacity are all set by the agreements you actually signed and by the rules of the terminals you actually use — none of it is universal. Where this article describes a mechanism, it describes the shape that mechanism takes rather than any one market's or any one contract's terms. The terms that apply to you are in the documents on your own desk.
The volume is what gets reported. The cargo is what gets delivered.