I'm long sugar (and to a lesser extent coffee) and currently down ~2% on the trade (though at one point was up 15% but didn't take anything off). The thesis is basic – they’re below in their long-run costs of production and they represent markets that are broadly uncorrelated to other financial assets. Over-supply has been an issue; a 25% increase in the Maharastra cane acreage will push India's output beyond what was produced during the 2017-18 season and also allow for an early crush start in October. Further out, Thailand is targeting 210 million metric tons of cane by 2024 (roughly doubling current production to put it as the third-largest producer, ahead of China but behind Brazil (1st) and India (2nd)).But most recently weakness in the Brazilian real has been hurting sugar and coffee the most. These commodities and the BRL are a natural hedge to each other, similar to the Chilean peso (CLP) and copper, reflecting the idiosyncratic relevance of these commodities to export activity as the world's top producers of each by wide margins. You can hedge long sugar/coffee by going long the USD/BRL, though you currently have to pay ~460 bps of carry on that pair (~640 selic rate minus ~182 fed funds), which is expensive insurance. Also, sugar is a contango market, so there is negative roll yield. London sugar (#5) is actually backwardated to the December 2018 contract, which reflects short-term pessimism. But with New York sugar (#11) you have to pay 9.5% roll yield out one year, which isn't helping the trade either. The New York May 2021 contract is trading at a 26% premium to spot, which isn’t very favorable. With coffee you pay 11% roll yield (for the July 2019 contract) with Sept 2020 trading at a 26% premium over spot, which is also unfavorable. So even though the longs are correct if you're thinking long-term – that's why the forward curve is so steep – remaining over-supply, BRL weakness, and yield chasers are keeping the price down. These trades require patience, similar to cocoa in 2017.