Good Weather In Latin America And Coffee Peaks In May

Commodity Trader’s Almanac observes that coffee prices have a seasonal pattern of peaking in May. This weakness provides a seasonal tradable short opportunity in coffee futures in May.

Being an agricultural crop, coffee is affected by the global production. Latin America – mostly Columbia and Brazil – is a major producer of coffee. If the weather there stays good in winter – no or minimal frost – then it results in normal to good crop. This means that coffee prices in the futures market start to decline in May. This weakness persists for 2-3 months. Off course this seasonal pattern could be upended by a number of factors, but almanac notes that in normal course of events, this trade has a fairly good success rate – 70% over a 37 years period.

I have made few modifications to this seasonal strategy. The strategy calls for entering the short trade in May when it closes at 10-Day high and exiting on a 10-Day low close after 54 trading days. The 10-Day high close filter does not trigger a trade every year but it does improve the results.

Below is the hypothetical results of this strategy since 2001. I have used a $140,000 equity for each contract based upon the max draw down.

Net Profit $31,838
# of Trades 10
Win % 70%
Avg. Ret (Per Trade) 2.3%
Avg. Win $6,316
Avg. Loss $(4,125)
Max Draw Down (27)%
Avg. Trading Days Held 60
Avg. Calendar Days Held 88
% of Time Invested 22%

Here is how these hypothetical trades would have fared since 2001:

Entry Date Entry Price Exit Date Exit Price Net
1-May-01 64.50 26-Jul-01 51.30 4,950
16-May-02 50.75 15-Aug-02 46.50 1,594
1-May-03 68.00 22-Jul-03 59.50 3,188
4-May-04 70.10 27-Jul-04 67.95 806
26-May-05 123.70 15-Aug-05 96.15 10,331
14-May-07 106.85 16-Aug-07 111.30 (1,669)
9-May-08 136.55 13-Aug-08 135.55 375
1-May-09 120.40 14-Aug-09 128.95 (3,206)
3-May-10 138.55 20-Jul-10 158.55 (7,500)
2-May-11 305.10 19-Jul-11 243.85 22,969

Starbucks Corp (SBUX), being a distributor of coffee is inversely affected by the coffee prices. It could be used as a substitute. Below is the hypothetical results for SBUX using $10,000 trade amount.

Net Profit $4246
# of Trades 9
Win % 56%
Avg. Ret (Per Trade) 4.7%
Avg. Win $1,643
Avg. Loss $(993)
Max Draw Down (37)%
Avg. Days Trade Held 62
Avg. Calendar Days Held 90
% of Time Invested 20%

Here is how these hypothetical trades would have fared since 2001:

Entry Date Entry Price Exit Date Exit Price Shares Net
31-May-01 9.76 20-Aug-01 9.35 1,024 (420)
6-May-02 11.33 12-Aug-02 9.86 882 (1,297)
8-May-03 11.50 31-Jul-03 13.67 869 1,886
7-May-04 19.08 27-Jul-04 23.89 524 2,520
11-May-06 36.54 17-Aug-06 31.29 273 (1,433)
4-May-07 30.70 23-Jul-07 28.17 325 (822)
11-May-09 13.45 30-Jul-09 17.63 743 3,106
6-May-10 25.61 9-Aug-10 25.66 390 20
9-May-11 36.12 31-Aug-11 38.62 276 690


The Fine Print: Before employing this strategy in your live account please understand the rationale behind this seasonality pattern. Then run some back tests for the trade start-date and holding periods. Make sure that the risk level, max draw down, win ratio and average loss falls within your comfort zone and that you can withstand the draw down and the associated risk. Also read our disclaimer.