Moving from Commodities to Global Macro

An article about my move from Commodities at Citadel to Global Macro at Point72 was recently covered in the financial press.

Some might wonder what the reasons were for a move from a specialist field to a generalist field, so I want to add some colour to it:

"Be a specialist, not a generalist!"...This phrase is something everybody who joins a trading desk will have heard at one point or another. Is it good advice?

In my opinion, that depends. In statistics, you have cases when you should use Ordinary Least Squares (a specialist model) an other cases when you should use Generalized Linear Models (a generalist model). Why should it be different in trading? In the past year, I was part of the European Natural Gas trading desk at Citadel - a rather narrow focus in terms of both, geography and asset class. Being so specialised is great, you can work on in-depth analytical tools and you can at least attempt to know everything about the space you are trading in. Ideally, this means that you will not miss any good trading opportunities. However, there are clear risks to this specialised focus: The biggest risk is that you can easily lose sight of the bigger picture. If you dive into European gas markets in so much detail that you try to predict where individual LNG vessels are going or how much gas a specific storage site can inject at any given stock level, you can be prone to falling into the trap of viewing the gas market as an independent market. But gas demand and supply is not independent, there is major benefit to understanding (a) gas markets on other continents, especially North America and Asia, and (b) power markets, oil markets, CO2 markets, as well as the state of the broader global economy. So in this context, specialists are more likely to see opportunities but also more likely to miss risks. Which one has more value, getting opportunities right or getting risk right? There is no right or wrong answer surely, but you know how they say in sports: "Offense wins games, defense wins championships"?

Another reason for my switch to the more general Global Macro trading field is uncertainty about Europe's energy future. With coal plants being phased out all over the continent, who can really tell where gas usage will be in 15 years? If you attend gas conferences, you will always hear a very positive story, one where natural gas will still be a large part of the energy mix in 100 years. If you attend renewable energy conferences, you will hear quite a different story. The truth is, nobody knows and it will largely depend on progress in the batteries industry over the next decade. And this is something that current gas trading portfolio managers do not have to account for, they will be retired by that time. But individuals in the early stages of a trading career should give a lot of thought to the question "What will this market be like in a decade?"

This last point also spills over into my third reason for the move to Global Macro: systematic trading (automation). I feel like natural gas trading is still a long way from very successful systematic trading strategies, there is a lot of talk about "systematizing analytics", but no real commitment to it even on leading gas trading desks - and while it is understandable that portfolio managers of successful desks want to maintain the status quo, it seems somewhat myopic, because even if you are successful now and take an "Any model I don't fully understand goes straight to the bin" view, that does not mean that others will not optimise their forecasting methodology and rely on experts in the time series and machine learning fields. This is different in Global Macro - with its abundance of data there is an undeniable need for systematic elements and I feel like this is the perfect position to be in the trading space: a discretionary desk (acknowledging that systematic trading still has a some shortcomings) with a strong push towards systematizing a large part of analytics (to be fit for the future of trading that seems most likely).

Leave a Reply

Your email address will not be published. Required fields are marked *

Protected with IP Blacklist CloudIP Blacklist Cloud