When moving outside the normal noise in equity and bond markets, the big story in the Global Futures markets has largely centered around the Cocoa markets (there are two of them). Here, a large part of the profits for trend followers have been generated. And the decision to have this peculiar market in the portfolio largely determined the YTD for trend based strategies.
Last month, the market was rife with calls for Cocoa to continue to trend up. Pierre Andurand, someone who was accused of being a Cocoa Tourist, made a prediction of 20k a few weeks ago. We, like many other overly Futures Interested Persons, have developed a keen interest in the gyrations of the Cocoa market.
As we will show, the Cocoa market has been a really uninspiring to trade over the last decade. The Future that could not trend, suddenly became one of the trendiest trades around.
What was looking like an unstoppable rise came to an abrupt halt in April as the rain clouds moved in over the Ivory Coast. The supply outlook brightened. As the weather change appears to be somewhat sustainable, Cocoa prices reversed with a vengeance.
After a stratospheric rise in price from mid-2022, Cocoa saw a ~500% increase a price. In the last couple of weeks, it has reveres from highs of 12,000 down to close to 8,000 per ton. Post the reversal, the market seems to have calmed down and recent movements have been small compared to the recent past.
To illustrate the difficulties trading Cocoa, we simulate a simple Breakout strategy, i.e. if prices are higher/lower than in the prior X days, we go long/short. While no particular CTA is using a simplistic strategy like this, it gives an indication of how CTAs have been positioned.
Since 2010, until today, only the longer time frames seems to have been able to pick up any sustainable pnl over time. In this example, we are using a volatility based position sizing, reducing the giveback in the end. As a comparison, we simulate the same strategy, but without dynamic position adjustments.
Without volatility-based position adjustment, all of the time frames would have been loss-making, despite the recent rally. The chart below shows how a volatility sensitive trend strategy would have adjusted position levels. We observe two effects, the increase in price and the increase in volatility leading to a dramatic volatility reduction. This resulted in a 95% reduction in positions for a generic trend follower.
We also note that only the shorter time frames have switched position in Cocoa, from long to short. As volatility and price have subsided a bit, CTAs have started to add back risk.
If trading Cocoa futures was a true one-hit wonder, or if the market has changed the dynamic completely is something best left for the more knowledgeable crowd. Here, we note that dynamic position sizing seems to have reduced drawdowns and managed the pnl slightly better over time, compared to non-vol adjusted trend strategies.
Cocoa is not alone of being unprofitable for trend traders. Silver is another classical example that continuously suffers from whipsaw losses. But occasionally they will make the year for trend managers. True outliers are difficult to predict.