The fuel oil market’s hidden challenge: navigating global blending economics and dynamics

With the upcoming release of our fuel oil vertical, we had the opportunity to speak with Hoa Nguyen, our commodity owner for fuel oil at Sparta to gain insights into its development, impact, and what it means for the fuel oil market.

During our conversation with Hoa, we discussed the current challenges facing participants in the fuel oil market and explored the world of fuel oil, along with the motivation behind Sparta’s latest vertical.

Sparta for Fuel Oil’s proprietary blender, combined with its extensive coverage of blend component differentials, offers an unparalleled forward view of blending dynamics across major fuel oil grades at global blending and bunkering hubs. The full release is scheduled for the end of March 2025, with early access available now.

A conversation with Hoa Nguyen, our man leading the charge of Sparta’s upcoming fuel oil vertical. 

Please give a brief introduction to yourself and your experience.

I’m Hoa Nguyen. I joined Sparta early last year to head up the development of our new Fuel Oil vertical. Before Sparta, I spent four years at Trafigura in Singapore and was in charge of building and running the Fuel Oil analytics infrastructure at the company.

I built an extensive range of tools there, not only the more traditional tools but also those crafted with the trading team to cater specifically to trading strategy needs.

Before Trafigura, I spent a brief stint in emerging market bond trading in Boston and equity research in New York City. I’m originally from Vietnam, but I’ve spent half of my life in Singapore and America.

In your view, what persistent challenges are faced by fuel oil traders?

Many people tend to focus very heavily on supply and demand fundamentals. Of course, those are very important and give us valuable suggestions for structuring our trading strategies. But after a while, you get the sense that most people in the market tend to look at almost the same things, using similar tools and often coming to similar conclusions.

For example, we often attribute big bullish moves in spreads to major balance draws driven by fundamentals. But when price action doesn’t align as expected, we tend to retroactively adjust our balances to fit the market or explain the divergence as the market ‘plays’—or simply accept that price action doesn’t always follow fundamentals.

These are all valid points, but the real question is, how can we do it better? How can we anticipate market moves earlier or trade with more conviction, even before the complete fundamental picture is clear?

Think of the market as a complex risk allocation mechanism. Even if it’s not perfectly efficient, prices react fast, especially near a dislocation.

And it’s not just finished grade prices (or premiums) that are more susceptible to actions by big players with an agenda; it’s also things like FOB prices for dozens of blending components, regional and inter-regional freight rates, forward arbitrages, and landed crude premiums.

The challenge is distilling insights from these countless price signals in new and better ways.

When extracting insights from prices, traders face two classic challenges: access and efficiency, especially in the opaque fuel oil market. Smaller traders struggle with visibility on physical prices without extensive operations.

At the same time, larger firms have the data but face the headache of scattered information buried in PDFs, Excel sheets, and endless emails. Turning all that into actionable insights takes a huge amount of time and effort.

Our first set of features for fuel oil focuses mainly on the blending aspect of the market; why is a lack of insight into blending dynamics such a big problem for fuel oil market participants?

We can think of the fuel oil market as three key segments: bunker fuel, refinery feedstocks, and power generation. We’re building our fuel oil vertical to help traders and analysts navigate these areas, starting with bunker fuel, the backbone of the market. Since most finished-grade fuel oil comes from blending, especially in Singapore, shifts in blending dynamics significantly impact trading.

Many traders have access to some blending components and track their prices intuitively. But manually gathering, remembering, and analysing price movements, especially when signals are mixed, is time-consuming and limiting.

This challenge is even greater for those without direct access to physical prices, meaning valuable market signals often go unnoticed.

Most traders still rely on supply and demand balances, which are useful, but fuel oil is unique; low and high sulphur balances hinge heavily on arbitrage flows, not just local demand and production. These arbs aren’t just for finished grades but also for blending components moving into hubs like Singapore, Fujairah, ARA, and the MED.

Calling these arbs correctly is critical, as they can swing by hundreds of thousands of tons monthly, often driven by blending economics.

And how does the first feature set of Sparta for Fuel Oil help solve this problem?

When I think about market fundamentals, I focus on the big picture: what problem the market is facing, what the solvers are, and whether the market has acted on them. I apply this same approach to price insights.

Our first feature set, including the fuel oil blender, isn’t just about blending your own components, though it can be used that way, especially with upcoming full customisation. It’s designed to show blending from the market’s perspective as an ‘omniscient blender’, that is, the market itself.

We gather extensive FOB (or CIF) diffs, freight, and paper values, turning them into landed diffs for blending hubs worldwide.

Using our proprietary cloud blender, built with input from quants and experienced blenders, we generate the most cost-effective blending recipes for each region and each grade.

So, how does this help traders and analysts? There’s a wealth of trade signals in this data and even more in our breakeven blender feature.

The best way to learn is to chat with the team, but consider this: If cheap heavy crude floods the market, Aframax rates spike, Suezmax stays flat, and mid-sulfur barrels shift toward high sulfur instead of low sulfur, how will that impact VLSFO blending costs over the next three months?

Should you lock in physical premiums now or wait? Buy or sell spreads? Will a swing barrel move East instead of West? Or even simpler, at what level is Al Zour LSSR landing in Fujairah?

These are the kinds of questions we help answer through price and blending economics.

Can you elaborate on Sparta’s new breakeven blender feature?

‘The breakeven blender is a first-of-its-kind tool, requiring significant computational power and advanced algorithms, something not available in the market today. While breakeven pricing is a familiar concept, this tool quantifies value dynamically, factoring in component relationships.

For instance, in a market short on mid-sulfur barrels, those components should command a higher premium than a simple sulfur penalty suggests.’

We’ll dive deeper into its intricacies in our next conversation with Hoa. If you would like to see our breakeven blender in action, book a demo with us today.

What will differentiate Sparta for Fuel Oil from any other market intelligence tool available on the market?

Here’s what sets us apart:

1. We are the only provider of market intelligence that collects a very extensive range of FOB/CIF diffs for blending components in a notoriously opaque market. For Singapore VLSFO blending alone, we collect FOB diffs on over 30 different streams, which are included in our calculations for the blender. This provides an unmatched view of the market as a whole.

2. We are the only provider in the market that runs a live blender for fuel oil that builds on the success of our very strong quantitative team when it comes to building optimisers (the existing gasoline blender as an example) and the extensive market data we are collecting across both physical and paper.

3. We are the only provider in the market that runs a powerful and innovative tool called breakeven blender.

4. We model and give our clients unparalleled access to spot and forward freight rates for up to 150 fuel oil routes at the moment, which will soon be extended to over 400 in later updates this year. What that means is that you don’t have to spend time talking to your chartering team, which then needs to talk to their brokers to get a quote for a vessel that you can use to calculate your arbs.

And these are just the first set of features we are releasing. We have many more projects in the pipeline to address other aspects of fuel oil gradually.

Who is Sparta for Fuel Oil built for?

Sparta is built specifically with the trader’s and analysts’ needs and workflows in mind, which has always been our principle. Think of Sparta as your trading companion, someone you ask for a second opinion on a trade from a (different) perspective.

In your view, how has the fuel oil market changed since you started?

Fuel oil has been around for a long time, but trading interest has surged in recent years, driving demand for better analytical tools and market expertise.

Significant shifts like IMO 2020 rendered many old models obsolete, while new players with deep pockets and high-risk tolerance have entered the space.

Since 2022, Russian high-sulfur barrels that once stayed in the West have been flowing East in huge volumes, with China, India, and Saudi Arabia emerging as key buyers.

The upcoming MED ECA in the next six months will potentially shake up blending dynamics in Europe, impacting everything from ultra-low sulfur fuels to 0.5% blends, mid-sulfur barrels, and heavy sweet crudes.

What’s next for Sparta’s fuel oil vertical? Are there any milestones or goals you’re mainly focused on?

We pride ourselves on having very fast product iterations. In the coming weeks and months, we plan to add more capabilities to our existing blender, including additional signals for paper trades and a dedicated arb’s dashboard for finished grades, blending components, and feedstocks.

Full customisation for the blender is also on my mind, and we may even venture into feedstock valuation, leveraging our existing capabilities with the LP on the crude side. Lots of exciting things are coming!