1 · The board, in USD

All prices on this page are quoted in US dollars, because that is the currency the world actually trades energy in. The board below is live from the same feed the Commodities Analysis page uses.

2 · What moves the price — systematically

Energy prices are not mysterious. They are the outcome of a small number of forces, and almost every headline you will ever read is one of them wearing a costume. Split them into four layers: demand, supply, plumbing (the physical middle of the chain) and the paper market.

The two columns. Anything that raises consumption or removes supply pushes the price up; anything that cuts consumption or adds supply pushes it down. The strength of each arrow depends on spare capacity — see below.

2.1  Demand — who is burning it

• Economic activity. Oil demand tracks industrial output, freight and air travel closely. A global slowdown is the most reliable bear force there is, because it removes consumption without anybody deciding to.
• Weather. Cold winters raise gas, heating oil and electricity demand; hot summers raise electricity demand for cooling and, in gas-heavy grids, gas with it. Weather is the biggest short-horizon swing factor in gas and power, larger than most political news.
• Substitution. When gas gets expensive relative to coal or oil, power generators switch fuel where they still can. That switch caps gas prices in normal times — and is exactly what is missing where old coal and oil plants have already been shut.
• Efficiency and structural change. Slow, one-directional, easy to underestimate. Heat pumps, electric vehicles, insulation and industrial closures permanently remove some demand each year.

2.2  Supply — who is producing it

• OPEC+ policy. A cartel decision to hold barrels back or release them is the fastest deliberate supply lever on earth. Read the meeting outcomes, not the pre-meeting rhetoric.
• Spare capacity — the single most important number nobody quotes. It is the volume that could be brought online within about 30 days and held there. When spare capacity is large, disruptions get absorbed and prices barely move. When it is thin, the same disruption produces a violent move. The same news is worth a dollar or twenty dollars depending on this one variable.
• Outages and accidents. Field maintenance, storm shut-ins, fires, strikes, sabotage, sanctions and war. Mostly temporary, occasionally not.
• Investment and decline rates. Producing fields decline every year without new drilling. Money not spent today shows up as missing barrels three to five years out — which is why a long stretch of cheap oil quietly plants the next expensive one.
• Sanctions and embargoes. These rarely remove barrels from the world; they mostly re-route them, lengthen voyages and add insurance and freight cost. That still raises the delivered price even when global production is unchanged.

2.3  Plumbing — getting it from there to here

• Inventories. Commercial stocks and strategic reserves are the shock absorber. Low stocks going into a demand season turn a small disruption into a price event.
• Refining. Crude is not fuel. Refineries convert it, and their margin — the crack spread — can blow out even when crude is flat. A refinery fire raises the price of diesel without touching the price of oil.
• Freight and insurance. When a route becomes dangerous, tanker day-rates and war-risk premiums rise. The cargo is unchanged; the delivered cost is not.
• The US dollar. Energy is priced in dollars, so a stronger dollar makes it more expensive in every other currency even at an unchanged dollar price — and mechanically dampens demand outside the United States.
• Chokepoints. Straits, canals, pipelines and terminals. Covered in detail in sections 4 and 5, because this is where the fragility actually lives.

2.4  The paper market — and the shape of the curve

Far more barrels trade on paper than are ever delivered. That is not a scandal; it is how producers and airlines hedge. But it means positioning and financial flows can move the price for weeks without a single physical barrel changing hands.

The one piece of jargon worth learning is the shape of the futures curve, because it tells you what the physical market is actually short of:

3 · Seasonality — what to expect through the year

Energy is one of the few markets with a physical reason to be seasonal. People heat in winter and drive in summer, refineries shut for maintenance in the quiet months, storage is filled in one half of the year and emptied in the other. That leaves a footprint in the price history — a real one, but a weak one, easily buried by a war, a cartel decision or a recession.

The two charts below are computed live from the actual price history at load time, not typed in from a textbook. For each calendar month we take every month-on-month change in the record, average them, and chain the twelve averages into one typical year. The number of years behind each month is shown, because an average of four Januaries is not evidence.

3.1  The calendar, and the reason behind each month

4 · The sea lanes — where it physically comes from

Roughly six out of every ten barrels of oil consumed in the world travel by ship, and essentially all liquefied natural gas does. Ships are cheap and flexible — until they have to pass through a gap a few kilometres wide that somebody can close. Every serious oil price shock of the last fifty years has been about one of the points on this map.

Red numbers are chokepoints; amber lines are the main crude and product sea routes; the dashed amber line is the Cape of Good Hope detour that everything falls back on when the Red Sea route is unusable.

• 1Strait of Hormuz — the single most important passage on earth for energy. In a normal year around a fifth of the world’s oil supply and a similar share of its LNG pass through a channel whose shipping lanes are a few kilometres wide. Saudi Arabia, Iraq, the UAE, Kuwait, Qatar and Iran all ship through it. There is no full alternative: the two bypass pipelines (see section 5) together move only a fraction of the normal flow.
• 2Strait of Malacca — the artery between the Indian Ocean and East Asia, and the busiest oil route in the world by volume. At its narrowest it is under three kilometres wide. Alternatives exist (Lombok, Sunda) but add days and cannot take the full traffic.
• 3Suez Canal & the SUMED pipeline — the short way between the Gulf and Europe. The canal has closed before for years at a time, and the parallel pipeline exists precisely because the canal is a single point of failure.
• 4Bab el-Mandeb — the southern gate of the Red Sea. Close this and the Suez Canal is pointless for Gulf–Europe trade, because nothing can reach it. This is why attacks in the southern Red Sea move European fuel prices.
• 5Turkish Straits (Bosphorus and Dardanelles) — the only way out of the Black Sea. Russian and Caspian crude, on a waterway that runs through the middle of a city of 15 million and is regularly closed by fog and weather.
• 6Danish Straits — the exit from the Baltic, and the route for Russian seaborne crude out of its Baltic ports.
• 7Panama Canal — carries products, LPG and LNG between the US Gulf and the Pacific. It is unique among these points in being limited by rainfall: the locks are fed by a freshwater lake, and in a drought the canal cuts transits and drafts. A weather event, not a political one, throttles it.
• 8Taiwan Strait — the approach to the world’s largest cluster of LNG importers (Japan, South Korea, China, Taiwan). Not narrow, but strategically loaded.
• 9Cape of Good Hope — not a chokepoint but the release valve. When the Red Sea is unusable, cargoes go around Africa: roughly nine to ten extra days from the Gulf to north-west Europe, more fuel burned, more ships needed to move the same volume, and freight rates up for everyone.

5 · The pipelines — where it comes from by land

Pipelines are cheaper than ships and far less flexible. A tanker can be redirected mid-voyage; a pipeline goes where it was built, through whichever countries it was built through, and if it stops there is no substitute on the other end. That inflexibility is exactly why pipelines are political.

Blue = gas, amber = oil, dashed grey = built but idle, shut or destroyed. Routes are schematic: the corridor is right, the exact trench is not.

• APower of Siberia (Russia → China) — the flagship of the eastward pivot. Gas that used to have only a European outlet now has an Asian one, though at a price set by a single buyer.
• BCentral Asia – China (Turkmenistan, Uzbekistan, Kazakhstan → China) — the largest gas import route into China, and the reason Turkmen gas never needed to reach Europe.
• CTurkStream (Russia → Turkey, onward to south-east Europe) — after the Baltic routes went dark, this became the main surviving Russian pipeline path toward the EU.
• DSouthern Gas Corridor — TANAP and TAP carrying Azerbaijani gas from the Caspian across Turkey and Greece into Italy. Europe’s only large non-Russian, non-LNG eastern gas route.
• ETrans-Med and GreenStream (Algeria and Libya → Italy) — North African gas under the Mediterranean, plus Medgaz directly to Spain. Politically exposed at both ends.
• FNorwegian export system (Norway → UK, Germany, Poland via Baltic Pipe) — now the largest single pipeline gas source for the EU, which makes Norwegian maintenance schedules a European price event.
• GDruzhba (Russia → Belarus, Poland, Germany, Hungary, Slovakia, Czechia) — the longest oil pipeline network ever built. Partly sanctioned, partly still flowing to landlocked buyers, repeatedly interrupted.
• HBTC and Kirkuk–Ceyhan (Azerbaijan and Iraq → Turkish Mediterranean) — both exist to reach open water without passing a Russian or Iranian chokepoint. That is their entire purpose.
• ICPC (Kazakhstan → Novorossiysk) — Kazakh crude that must cross Russia and then exit through the Turkish Straits. Two dependencies stacked on one route.
• JSaudi East–West (Petroline) and ADCOP (UAE → Fujairah) — the two Hormuz bypasses. They move oil to the Red Sea and the Gulf of Oman without entering the strait. Their combined capacity covers only part of normal Gulf exports, and the Red Sea end has its own risk.
• KESPO (Russia → Pacific coast and China) — the eastern oil equivalent of Power of Siberia.
• LNorth American systems — Keystone (Alberta → US Gulf), Trans Mountain (Alberta → Pacific) and the Trans-Alaska line. Internal to a continent that is a net exporter, which is precisely why North American prices behave differently from everyone else’s.
• ⚊Idle or destroyed — Nord Stream 1 and 2 (damaged 2022, out of service), the Yamal–Europe route through Poland, and the Maghreb–Europe line through Morocco. Built capacity that exists on the map and delivers nothing. A pipeline is only an asset while the politics hold.

6 · Water, weather and heat — the risk nobody prices until it arrives

Straits and pipelines get the headlines. In practice, a surprising share of energy disruption over the last decade has been caused by something with no politics in it at all: not enough water, or water that is too warm. It hits three separate things at once — transport, cooling and generation — and it hits them in the same weeks, because the cause is the same.

6.1  Rivers as fuel corridors

• The Rhine carries on the order of 200 million tonnes of cargo a year — diesel, heating oil, coal, chemicals, grain — into the industrial heart of Germany, eastern France, the Netherlands and Switzerland. Loading is set by a reference gauge at Kaub, whose long-run average is about 208 cm. Below roughly 77 cm barging becomes structurally uneconomic; around 40 cm commercial traffic effectively stops. The previous record low was 25 cm in October 2018. In the summer of 2026 the gauge fell to single digits — about 6 cm in mid-August, the lowest since records began in 1880 — with barges loading a fraction of capacity and liquid-bulk freight rates from Rotterdam to points south of Kaub rising several-fold in two months (I).
• There is no easy substitute. It takes roughly 60 trucks to replace one 1,500-tonne barge. Europe does not have that many spare trucks or drivers, and rail is already near capacity on the same corridors. A river that stops does not shift to road — it mostly just stops.
• The same story elsewhere: the Danube (riverbed blasting and dredging in Romania in 2026 to keep a channel open), the Mississippi (low water repeatedly restricting the grain and fertiliser barge trade), the Yangtze (the 2022 drought cut Sichuan’s hydro and forced industrial shutdowns), and the Panama Canal, whose locks run on rainwater and which cuts transits in a dry year.

6.2  Water as coolant and as fuel

• Thermal plants need cooling. Nuclear and coal stations that draw from rivers face two limits in a heatwave: too little water to draw, and rules capping the temperature of what they discharge. Output gets curtailed exactly in the week the grid is under maximum cooling load — the worst possible correlation.
• Hydro is a weather derivative with a turbine attached. A dry year in a hydro-heavy system is a direct energy shortfall. Whole national grids run on it: Norway, Brazil, Colombia, Ethiopia, Kenya, Laos, much of Central Asia, and the Zambia–Zimbabwe grid at Kariba, where low reservoir levels have produced some of the longest sustained blackouts of the decade.
• Snowpack is storage. In alpine and Nordic systems the snow that falls in winter is, in effect, next summer’s battery. A warm winter with rain instead of snow moves the inflow forward by months and leaves late summer short.

6.3  The wind and sun version of the same problem

Wind and solar have a weather risk of their own, and it is correlated across a continent. A still, overcast, cold high-pressure system — the Germans call it a Dunkelflaute — can hold wind and solar output near zero across north-west Europe for several days in midwinter, when demand is at its annual peak. Solar output in high latitudes is at its yearly minimum in exactly the months heating demand is at its maximum. This is not an argument against either technology; it is an argument for being honest that a grid is only as strong as its worst simultaneous hour, not its annual average.

7 · Electricity — every source has a transport loss and a supply risk

Electricity is not an energy source. It is an energy carrier — a way of moving energy from where it was released to where it is used. That distinction is the whole of this section, and it produces one rule with no exceptions:

The chain, with the losses that are usually left out of the brochure. Percentages are typical ranges, not constants: they vary widely by technology, age, distance and country. The headline number to keep is that between the energy in the fuel and the energy doing work at the socket, a large minority — often a majority — is gone.

7.1  Where the energy actually goes

• Conversion is the big one. A conventional thermal power station turns roughly 33–45% of the fuel’s energy into electricity; a modern combined-cycle gas turbine reaches about 55–62%. The rest leaves as heat. Hydro turbines, by contrast, convert around 90% of what passes through them.
• Transmission and distribution lose roughly 8% of generated electricity worldwide. Well-run grids run 4–7%; weak or over-extended ones lose 15–25%. High-voltage DC lines lose on the order of 3% per 1,000 km plus a couple of per cent in the converter stations at each end — which is why long-distance power transfer is possible but never free.
• Storage costs a round trip. Pumped hydro returns about 70–85% of what you put in; grid lithium batteries about 85–92%; hydrogen used as power-to-power storage returns only about 25–40%. Storing energy is buying insurance, and the premium is paid in kilowatt-hours.
• The LNG chain is the equivalent tax on gas that travels by sea: liquefaction alone consumes roughly 8–10% of the gas, and shipping, boil-off and regasification add several per cent more. Gas that arrives by ship is structurally more expensive than the same gas in a pipe, before anybody argues about price.

7.2  The sources, and what each one is actually exposed to

8 · Nuclear fuel — uranium is a fuel, not a perpetual motion machine

Nuclear power is often discussed as though a reactor, once built, simply produces. It does not. A reactor consumes fuel, continuously, and that fuel has a supply chain as long and as concentrated as any pipeline. A typical plant is refuelled every 12 to 24 months, replacing roughly a third of its core each time. Stop the fuel chain and the reactor stops — later than a gas plant would, but it stops.

8.1  Where the uranium comes out of the ground

Share of world uranium mine production. Around three-quarters of it comes from just three countries. Figures are approximate and move a few points year to year (II).

Kazakhstan alone produces about 40% of the world’s mined uranium — a larger share of its commodity than Saudi Arabia has ever held of oil. Canada and Namibia follow, then Australia, Uzbekistan, Russia, China and Niger. Most Kazakh output is extracted by in-situ recovery, which is cheap, and much of it has historically left the country on routes that cross Russia.

8.2  The real bottleneck is not the mine

Uranium out of the ground is useless to a reactor. It has to pass through four more industrial steps, and each step exists in only a handful of places on earth:

• Milling turns ore into uranium oxide concentrate — yellowcake. Done at the mine.
• Conversion turns the oxide into uranium hexafluoride gas. There are only about five commercial conversion plants in the world, one each in Canada, China, France, Russia and the United States (III).
• Enrichment raises the fissile fraction from 0.7% to the 3–5% a power reactor needs. Capacity sits with Russia, China, the United States, the European URENCO consortium and Orano in France. Russia holds on the order of two-fifths of world enrichment capacity — which is why sanctions on Russian energy have consistently carved out nuclear fuel.
• Fabrication presses the enriched material into pellets and assembles fuel rods — and assemblies are reactor-specific. A Russian-designed reactor cannot simply take a Western fuel assembly without a qualification programme that takes years. This is the single most under-appreciated lock-in in the energy world.

9 · Who gets hurt first — exposure by country

The measure below is net energy import dependency: the share of a country’s total energy use that has to be brought in from outside. Red means highly dependent on imported oil, gas or fuel — the countries that feel a disruption first and hardest. Green means broadly independent, or a net exporter. Amber is in between.

EU figures are the published 2024 import-dependency rates (IV); non-EU figures are approximate and given as bands. Dependency is a structural measure, not a forecast of who suffers most in any particular event.

10 · What high energy prices actually do — fast and slow

An energy shock is not one event. It is a sequence, and the parts arrive on very different clocks. Knowing which effects are days away and which are years away is the difference between preparing and panicking.

The transmission timeline. The top band moves before most people have finished reading the news; the bottom band is still moving when the price itself has long since come back down.

10.1  Within days to weeks

• Pump and heating prices. Retail fuel follows the wholesale price within one to two weeks — faster upward than downward, which is a real and well-documented asymmetry, not a conspiracy.
• Freight and air fares. Fuel surcharges are contractual and reprice almost immediately; airline fares follow within weeks where hedges roll off.
• Industrial curtailment. Energy-intensive producers — aluminium, zinc, glass, cement, ceramics, paper, chlorine, steel — simply stop when power costs more than the product is worth. The decision is arithmetic and takes days.
• Fertiliser. Ammonia is made from natural gas, and gas is most of its cost. Ammonia plants shut within weeks of a gas spike — which quietly loads the next planting season, and therefore the food price after that.
• The currency of an importing country weakens as the import bill rises, which raises the local-currency energy price further. Importers get hit twice.

10.2  Within one to twelve months

• Household bills reset as regulated tariffs and fixed contracts roll over — often with a lag of a quarter to a year, which is why the pain peaks long after the headline.
• Headline inflation, then core. Energy enters directly, then indirectly through every transported good, then through wage bargaining. That last step is what turns a price spike into an inflation problem.
• Monetary policy. Central banks face the worst trade-off there is: a supply shock raises prices and cuts output. Whichever they choose, someone is hurt — and interest rates on mortgages and business credit move accordingly.
• Government intervention: subsidies, price caps, windfall taxes, strategic reserve releases, export restrictions. Each helps someone immediately and distorts something else later.
• Demand destruction. People drive less, heat less, fly less. This is the mechanism that ends the shock, and it is not painless — it works by making energy unaffordable for the people with the least slack.

10.3  Over one to ten years

• Industry relocates — permanently. When a smelter or a chemical plant closes because of energy costs, it usually reopens somewhere with cheap energy, not where it was. The price comes back down; the factory does not come back. This is the most consequential and least reversible effect on the list.
• The investment cycle turns. High prices fund drilling, terminals, mines and reactors. Five years later that supply arrives, often all at once, and produces the bust. High prices are the cure for high prices — on a multi-year delay.
• Efficiency and substitution get built. Insulation, heat pumps, electric vehicles, industrial process changes. Slow, boring, permanent — and the only structural defence there is.
• Fiscal divergence. Importing governments borrow to subsidise; exporting ones accumulate. Sovereign credit quality moves in opposite directions on the same event.
• Politics. Energy prices have preceded governmental instability often enough that it should be treated as a normal consequence rather than a surprise, particularly in economies where fuel is a large share of household spending.

11 · Fusion — when, realistically, and whether ever

Fusion deserves a straight answer rather than either the hype or the sneer, because the honest position sits between them.

Where fusion actually stands. The physics milestone is real and already passed. The engineering and economic milestones are not, and they are the ones that decide whether it ever powers anything.

11.1  What has genuinely been achieved

In December 2022 the National Ignition Facility in California produced more fusion energy from a fuel capsule than the laser energy delivered to it — about 3.15 MJ out for 2.05 MJ in. That is a real, historic result: scientific ignition. The caveat is equally real. Producing that laser pulse drew on the order of 300 MJ from the wall. Scientific gain and engineering gain are different things, and only the second one sells electricity.

11.2  What is still missing

• Tritium supply. The fuel is deuterium and tritium. Deuterium is abundant in seawater; tritium barely exists in nature and decays with a half-life of about 12 years. A power plant must breed its own tritium from lithium in a blanket around the reaction — and no machine has yet demonstrated a self-sufficient breeding cycle. This is a bigger open problem than plasma confinement.
• Materials. Fusion neutrons are far more energetic than fission neutrons and progressively wreck the structure containing them. No material has yet been qualified for a plant lifetime under that flux.
• Duty cycle. Experiments run for seconds. A power station must run for months, restart reliably, and be maintainable by remote handling inside an activated structure.
• Economics. Even a working reactor must produce electricity at a price that competes with solar plus storage and with fission — both of which will also have moved on by then.

11.3  The dates, as they currently stand

• ITER, the international flagship, rebaselined in 2024: research operations now start around 2034, with full deuterium–tritium operation around 2039. It was originally meant to see first plasma in 2025.
• Private compact tokamaks — principally Commonwealth Fusion’s SPARC, using high-temperature superconducting magnets — target net plasma energy gain in the late 2020s, with a first commercial follow-on machine aimed at the early-to-mid 2030s. Timelines here have already slipped once and should be read as intentions, not schedules.
• Other approaches — stellarators, field-reversed configurations, inertial and magneto-inertial concepts — now number dozens of ventures across more than a dozen countries. One has sold a power purchase agreement for the late 2020s. None has delivered a kilowatt-hour to a grid.

12 · What to watch, and what to do about it

The purpose of everything above is to be useful before the event. Here is the short list worth checking once a month, and the short list of things that are worth doing while nothing is happening.

12.1  The monthly watchlist

• Spare production capacity. Thin spare capacity turns small news into big moves. This single variable explains most of the difference between a shrug and a spike.
• Inventories against their five-year average — crude, products, and gas storage fill percentage going into the heating season.
• The shape of the futures curve. Backwardation means physical tightness; contango means surplus. See section 2.4.
• Freight and war-risk premiums, and transit counts at the chokepoints in section 4. Transit counts fall long before anyone announces a closure.
• River gauges and reservoir levels in the basins you depend on, and the refinery maintenance calendar.
• Scheduled decisions — cartel meetings, tariff resets, central-bank dates. Known in advance, and reliably under-prepared for.

12.2  What is worth doing in calm weather

• Fix price when nobody is frightened. Contracts are cheapest precisely when the news is boring. Fixing during a spike is buying insurance after the fire has started.
• Buy efficiency rather than forecasts. Insulation, a better heating system, a more efficient vehicle and lower consumption pay out in every scenario, including the ones nobody predicted. No forecast is required for the return to arrive.
• Hold a buffer. A full tank, a filled oil or pellet store, a few weeks of the things that go scarce. Modest, cheap, and it removes you from the queue.
• Count your independent routes. For a household: how many separate ways can you heat and cook? For a business: how many suppliers, and do they share a single upstream chokepoint? One supplier with two contracts is still one route.
• Do not buy into the spike. The most expensive decisions in every energy crisis are made in the first fortnight, by people reacting to a headline with a full-size order.

13 · Data freshness — checked monthly

Half of this page is live: the price board and the seasonality charts recompute from the market data feed every time the page loads, so they cannot go stale. The other half — volumes, capacities, production shares, project dates — is written text, and written text rots. It is reviewed on a monthly cycle and flagged here.

• Live, never stale: the USD price board and every number derived from it; the seasonality averages, hit rates and year counts; the date range and row count behind each chart.
• Reviewed monthly: chokepoint volumes and status, pipeline status (which lines are flowing, idle or destroyed), uranium production shares, import-dependency figures, and the fusion project dates in section 11.
• Expected to change: anything in section 11, and the status of any single pipeline or strait. Anything describing a live conflict should be assumed out of date and verified before it is relied on.