Commodity resources
12 March 2022
Spread trading against seasonality
16 March 2022
After last week's sugar (Interesting calendar spread on sugar), today I am analysing a calendar spread on another "soft" commodity, coffee. I take my cue from one of the spreads recommended by SpreadCharts, which uses its own system for choosing futures and spreads based on artificial intelligence to identify the best opportunities of the moment. The spread I am going to analyse, as constructed by SpreadCharts, is KCH23-KCU22 and I will show you why this spread caught my eye.
Above is the Seasonality stacked chart and you can see that the spread this year is well below all the previous nine years. Not only that, it is the only spread that is in backwardation at this point in the year. I will explain the reasons for this at the end of the analysis when I outline the coffee fundamentals.
Confirmation of this anomaly can also be seen by looking at the Continuous histogram below.
In addition to the fact that the current spread is at an extreme level, well outside the 5th-95th percentile range, in the top left-hand corner I can read that over the last 20 years the spread has been above zero (i.e., in contango) 94.33% of the time and that therefore a negative spread is an anomaly.
To complete this part of the analysis, I show you how a backwardation at this point in the season represents an anomaly by taking the chart of the coffee term structure.
Comparing the current term structure with those of the 5- and 15-year averages, I can clearly see that the latter are in contango, with each delivery priced higher than the previous ones.
The motivation, therefore, that drives me to trade this spread is certainly not the seasonality, however I also show you the chart with the 5-, 15- and 30-year seasonal patterns.
The three seasonal patterns are bullish and also well correlated. So, my trade does not go against seasonality. My analysis continues by checking how hedge funds are positioned on coffee. I do this by taking the chart with the net position of the C.O.T.
Investors' interest in coffee has been waning for some weeks now, so much so that the net position returned last Friday (although it would be more accurate to say last Tuesday since the data refers to that day) below, albeit slightly, the 40K "resistance". So, it is possible that in the coming weeks the price of coffee could fall, which would allow a return of the term structure in contango.
Other charts do not interest me since the trade, as mentioned, is based on an anomaly and not on seasonality. To conclude, let me turn to the fundamentals of coffee. First, in the second half of 2021 coffee crops in Brazil were hit first by abnormal frosts and then by a drought that significantly reduced their yield. This was compounded by logistical problems. So, there is a reduction in supply behind the sharp rise in coffee prices in 2021 and in the first weeks of 2022 (Brazil is the world's largest coffee producer and exporter).
Now, however, the situation seems to have settled down at least. There are three pieces of news that struck me most:
- Brazil exported 3.1 million 60-kg bags of green coffee in February, 14.3% less than a year ago but up from the previous month as global trade logistics improved.
- The forecast is for a crop of 63.38 million 60kg bags next season (2022/23). Brazil is estimated to produce a crop of 54.40 million bags in 2021/22, pushing the global arabica market into a large deficit of 8.45 million bags.
- The latest news can only be about the war in Ukraine (unfortunately) with coffee prices in India crashing. Russia accounts for 74% of India's coffee imports while Ukraine accounts for 20%. However, the main reason for the fall in prices is not the stalling of exports in either country but uncertainty and fear in European markets over the impact of the war.
This problem is also expected to affect Brazil to some extent. The ban on ships docking in the two countries (Russia and Ukraine) and the removal of some Russian financial institutions from the SWIFT are creating difficulties that should be reflected in upcoming exports to Russia. Russia is the sixth-largest importer of Brazilian coffee.
My analysis has come to an end. Before drawing conclusions, I want to go back to the beginning of the article and show you the SpreadCharts signal regarding the KCH23-KCU22 spread.
Conclusions. Everything agrees on a possible drop in coffee, a realignment of prices (term structure) and a return to normality. This would also lead to a rise in the KCH23-KCU22 spread which is the subject of my analysis. Certain is that the period is not exactly calm and from one week to the other the situation can be very different. Therefore, always make much attention and be prudent. Sometimes certain "occasions" can carry to losses much heavy one. I know many traders that to shorts the wheat they got burned.
I take my cue for my analysis from one of the spreads recommended by SpreadCharts, which uses its own system for choosing futures and spreads based on artificial intelligence to identify the best opportunities
8 Comments
Thank you. I don’t trade sorts due to the high ICE data fees but is is great to see your analysis in action. Loving your books!
Thank you, Keth! I don’t have real-time data for softs; with spread trading, they are not so important. 10 or 15 minutes of delayed data is fine too.
Hi David – Its a great analysis. I have been tracking coffee for a while especially post the recent peaks. What is the time frame for this trend (spread) to normalize ?
what is the ideal entry time frame ?
Hi! Your question is the classic million-dollar question. To answer it would take a long time, I will try to summarize. There is no rule, in trading, it’s a question of probability. Based on your knowledge of the market, you have to identify the levels at which a reversal of the trend is most likely to occur. Personally, I think the spread can still fall a bit (not much), so I would try to buy the spread, if possible, in the area -5,85/-6,35 (but it’s not said that it will get there). Then, it also depends on how many contracts you want to buy. With more contracts, the entries can be staggered making attention that the total stop-loss corresponds to the maximum loss that you have decided. I repeat it’s all a matter of probability, of how much probability you give to the spread to invert the bearish trend at a determined price. You have to use what is called “subjective probability”. However, the subject is much more complex and can’t be explained in a few lines.
Thank you David
I’m just at the beginning and it’s very usefull to meet so short and strong analizes
Hi Andrey, I’m glad my analyses are useful to you. Keep following me and write to me if you have any doubts.
For anyone who is short the near month coffee. https://www.brecorder.com/news/40162876
Thank you!