Data Science — Fri Mar 06

← Home | ← data-science

Global Shock Part 6: What We Can Conclude

Fri Mar 06
#Data Science
oil-price-chart — brent & wti spot / 1986–present
// brent latest loading...
// wti latest
// all-time high (brent)
// all-time low (brent)
$
brent wti spike normalization restoration
$ fetching price data
// shock events — click to highlight phases

Conclusions

At the start of this project, I was really trying to answer three simple questions:

  • How does the heating oil industry actually work?
  • What will this new disaster push oil prices to?
  • How long do I need to ride it out?

Those questions were the reason for building this in the first place.

Not to produce a dramatic chart. Not to build a neat little event timeline. But to get closer to a more honest answer.

And the first thing this work makes clear is that you have to compare as close to apples with apples as possible.

Not every crisis belongs in the same category. A war involving a major producer, a shipping disruption, a sanctions regime, a refinery outage, and a demand collapse may all look like “global shocks” from a distance, but they do not move the market through the same mechanism. That means comparisons only become useful when the structure of the disruption is broadly similar. Ukraine and Iran, for example, may both be geopolitical events, but the useful comparison is not that both are scary. The useful comparison is how each one changes perceived supply risk, transport risk, escalation risk, and the market’s expectation of containment.

That is also why I broke the model into phases.

A shock is not just a moment on a chart. It is a sequence.

Spike

This is the first violent repricing.

It is the moment where the market reacts fastest, usually because it fears lost supply, blocked transport, or a wider escalation that has not yet been measured properly.

Normalization

This is the adjustment phase.

The first panic has happened, but now the market starts working out what is real, what is temporary, and what can be absorbed. This is where rerouted shipping, reserve releases, production responses, sanctions, and sentiment all start getting priced in more rationally.

Restoration

This is the longer move back toward equilibrium — if equilibrium returns at all.

Sometimes restoration means a drift back toward pre-shock levels. Sometimes it means the market settles into a new normal instead. That matters, because not every event ends with a neat reset.

That phase structure does not give me a magic answer to all three questions. But it does improve the quality of the answers.

It helps answer how the heating oil industry actually works by showing that heating oil prices do not simply respond to headlines. They sit downstream of crude, refining, transport, storage, wholesale pricing, and local delivery. So when disruption happens, what matters is not just the event itself, but where that event hits the chain and how the market thinks it will spread.

It helps answer what this new disaster might push oil prices to by forcing a more disciplined comparison. Not every event is a useful precedent. The better question is: what past event behaved most like this one? Is this closer to a short-lived security premium, a wider supply shock, a transport disruption, or a collapse in demand? That does not produce certainty, but it does produce a more grounded range of possibilities.

And it helps answer how long I need to ride it out — although here the answer is still necessarily imperfect.

I cannot put an exact date on how long someone needs to tighten their belt with heating oil.

What I can say now, though, is more useful than a false promise.

I can monitor the pattern.

I can look at whether the market still appears to be in spike, whether it has started to normalize, or whether it is moving toward restoration. That means the response does not have to be blind panic. It may be more sensible to buy smaller amounts while the market is still in the shock and adjustment stages, rather than assuming the first violent move is the final price the whole season will be built around. Then, if the pattern starts to look more like restoration, buying decisions can become less about fear and more about timing.

That is really the conclusion this project leads to.

I started out wanting certainty.

What I got instead was something better: a way to read the pattern.

This chart does not tell me exactly what happens next, and it does not promise that every crisis will follow the same timetable. But it does make one thing much clearer: the shock itself is never the full story. What matters is how the market absorbs it, whether the disruption spreads, whether it gets contained, and whether prices begin to bend back toward normality or settle somewhere new.

So if this project answers those original three questions at all, it answers them like this:

  • the heating oil industry works through a chain, not a headline
  • a new disaster only matters in price terms if it changes the expected balance of that chain
  • and riding it out is less about waiting for a date on a calendar than learning to recognize whether the market is still spiking, starting to normalize, or finally moving into restoration Gareth Winterman