How to Build Profitable COT Trading Systems

COT positioning reports can give you an edge and make you a better, more profitable trader.

Today, we're going to be going through three specific examples of profitable trading systems that you can trade, all of which incorporate COT positioning data.

This is the third article in a set of three.

In our first article, we talked about COT report 101. What is the report? Why does it matter?

In the second article, we talked about commodity market participants, with a focus on hedge funds.

And now in this third article, we're going to tie everything together.

This is the fun stuff.

We're going to be talking about how you can build real systematic trading strategies based on COT data, based on what we talked about in the first two articles.

Today, we're going to be going through specific trading systems for three different agriculture commodity markets:

  • Oats

  • Live Cattle

  • Kansas Wheat

All three of these trading systems use a pretty similar big picture idea:

Sometimes hedge funds get too long and sometimes hedge funds get too short.

We're going to be building systems that take advantage of either short squeezes or long liquidation events.

We’ll be going through these examples in TradeStation using TradeStation EasyLanguage, but if you're a discretionary trader, if you want to do this in Excel, or you want to build it in Python, that's no problem. We’ll use TradeStation so we can look at some basic profit statistics. But, most importantly, let's focus more on the logic of the systems and less on the specific code itself.

THRESHOLD TRADING

Now for this first trading system, we're going to be going through a very basic example of what we call COT threshold trading. We'll use Oats as an example. The questions we’ll be asking: Is the hedge fund net position above a certain threshold? Are hedge funds too long and vulnerable to a liquidation event?

We can build a very simple trading system using just two lines of logic that says whenever that non-commercial net position is above 2,200 contracts, i.e., when hedge funds are extended long, then we want to sell short Oat futures the next day on the market open. And after 21 days have gone by (about a month), we want to exit that short position.

 

Sample oat strategy with 2 lines of logic

 

Now, if we look at the strategy performance report, that's a nice upward sloping equity curve, meaning this strategy has generally made positive cumulative profits over time. We can see that it's made money in 12 out of 18 trades over the past few years.

 This shows promise using a very simple approach that we could trade the Oats market using that non-commercial position as a threshold to establish shorts.

Equity curve line for sample Oat trading strategy

BIG FLOWS

Now for our second trading system example, we're going to be talking about big hedge fund flows.

Sometimes weekly COT reports show that hedge funds bought or sold a large number of contracts, and often that can send important forward-looking price signals about that market.

Let's look at an example for the Live Cattle market.

If we look at data from the August 24th, 2021 COT report, we can see that non-commercial hedge fund traders added 14,760 contracts of new longs during the week. That's the third-largest inflow week on record.

Now, if we look at the other 20 largest inflow weeks since 2006, we can see that Live Cattle prices generally drop during the one-month and three-month periods that follow.

live cattle market showing price changes after long inflows

Another way to say this is hedge funds tend to be “long and wrong” when they buy a lot of Cattle futures.

Hedge funds buy live cattle futures, prices go up, the market runs out of buyers, prices naturally settle back in the one- and three-month periods after.

We can see that happened in real-time. After we sent this note to clients on August 28th, 2021, if we look at what live cattle futures have done…they’ve settled lower, as predicted.

Live cattle market

THRESHOLD TRADING + PRICE SIGNAL

Now for our third and final trading system, we're going to be using the same threshold idea that we talked about in the first trading system, but also incorporating a price signal to trade.

Threshold trading is great. It's easy to understand, but it has one big shortcoming and that is hedge funds can stay extended long or extended short for long periods of time. It often takes some catalyst, some fundamental or non-fundamental trigger that moves prices against hedge fund positioning and drives all these funds to liquidate their big positions at the same time. We're going to build a trading system that uses this price trigger idea to trade profitably.

Now let's take the same code that we used in the first example for the Oats market, and let's apply it to Kansas Wheat. So in this case, if the non-commercial net position is above a 2,000-contract threshold, then sell short Kansas Wheat next bar at market.

 

kansas wheat sample strategy

 

We can see if we look at the strategy performance report, this is a decent trading approach - we've got a nice upward sloping equity curve.

This approach would've made $30,500 over the last 10 years using this approach. Not bad.

Kansas wheat equity curve

Now we can improve this a little bit.

We can also layer on a price signal.

Instead of only using a COT position threshold, we’re going to use a threshold PLUS a price signal of a two-day-low stop. So the system will sell short only if prices break through the lowest level of the past two trading sessions.

Kansas wheat sample strategy with price signal

Whenever momentum starts turning lower, this system is then going to sell instead of just selling above that threshold. Now, if we look at our strategy performance report, it's gotten a little bit better. Our P&L has improved.

Our profit line looks better. More profits, fewer drawdowns.

Kansas wheat equity curve with price signal

Now in both the Oats system in the first example and the Kansas Wheat system in the third example, we talked about selling short when hedge funds are extended long.

But there are, of course, other ways that you can trade COT data:

  • You could add more price filters to some of these systems or a seasonal filter or volume or momentum.

  • You could also incorporate positions from some of the other market participants we talked about in our second video: other reportables, non-reportables, commercial positioning.

There's a lot more you can do here.

If you are looking for more ideas on how to build trading systems around COT data, these are fantastic resources:

The Commitment of Traders Bible by Stephen Briese

Trade Stocks and Commodities with the Insiders by Larry Williams.

Hopefully this article gave you a good sense for how you can build profitable trading systems using COT data.

Any questions? insight@peaktradingresearch.com