Is your discretionary trading strategy becoming too systematic?
Stop trading like an algorithm and start trading like a human
I recently completed my biggest-ever journal review to date. Going through every trade I’d ever made, I classified each one by exact setup. At first, I thought it would be simple enough, but I ended up with 31 different types of setup. And even with this many categories, it still felt reductionist.
Next, I looked at the expectancy of each setup. It was clear that some setups performed a lot better than others. So I created an alternate strategy in which I only took the highest probability setups (only 7 out of the 31 setups). The results were incredible. Not only would I have exceeded the returns I made trading every setup, but it took 70% fewer trades to get there.
It felt like I’d found the cheat codes. I altered my risk strategy accordingly. I didn’t want to stop trading all the other setups, as I believed my performance would continue to improve in them. But I wanted to reduce the risk I took in those weaker-performing setups.
I took the new strategy live. Naturally, I expected to reach new heights of profitability with less drawdown; the holy grail for any trading strategy. What happened next was truly spectacular, but not in a good way.
An unexpected downturn
I began to enter my worst drawdown to date. Despite lower overall risk levels than before, I made new records both in total drawdown and in the number of consecutive losers.
It didn’t make any sense. How could my performance have deteriorated after having made such a breakthrough? Surely this knowledge was worth something?
After an endless amount of investigating and soul-searching, I discovered why my performance had taken such a nosedive.
Before my huge journal review, I used to see every trade as unique. No two setups were the same, and each setup was made up of countless elements with so much nuance it would have been impossible to classify them. Yet I found a way to categorize them, putting them into neat little boxes with probabilities and expectancies, and other juicy bits of data. But in doing so I unwittingly changed my entire trading process.
A human algorithm
My process of trade selection was to spend at least a few minutes on each chart, reading the story that price action was telling. I would narrate this story out loud to myself, trying to make sense of what was going on. If there were strong enough signals that the market would head in a certain direction, I’d look for a setup.
But now I looked for the setup first. Like a pattern recognition robot, I scanned the chart for setups, spending only thirty seconds or so on each chart. If I found a setup I’d refer to my journal statistics to decide if I should take it or not. I no longer carefully analyzed the story of the price action leading into it, instead, I viewed each setup in isolation.
I was just setup-spotting. My eyes scanned the chart for a setup the same way a child looks for Wally in a “Where’s Wally?” book. Sure, I was getting good at finding Wally, but I was missing all the other details in the picture. My pattern recognition skills were getting sharper, but I was missing the bigger picture of what the market was actually doing.
Insert pic of where’s Wally with the rest of the picture blurred and then a chart with a clear setup and the rest of the picture blurred.
I’d become nothing more than a human algorithm – an embarrassingly basic one. Was there a setup on my chart or not? If so, what is the probability of success and what risk should I use?
Armed with this insight, I looked back on my recent string of losers. They all had one thing in common. Objectively, they all had setups that at face value looked worthy of trading. Using only logic it was reasonable to expect there would be a reaction at my entry level. But when I looked closer at the price action preceding the setup, it was telling an entirely different story.
The public relations story
Whenever a corporation is involved in some scandal, its public relations team will always find a way to spin it to preserve its public image. They present this phony version of the truth in such a way that most find it plausible without looking too much into it.
But underneath this carefully constructed narrative lies the real story. And if you look deeply enough at all the facts, it’s usually pretty obvious.
Price action tells a story of buyers and sellers in an endless battle for control of the market. But there are usually two versions of that story occurring at the same time.
The first is the face value, “public relations” version of the truth. It’s full of false advertisements and traps that you can easily fall into if you’re too quick to believe it. It’s the one that novice traders tend to believe, where every price swing looks like a trading opportunity they can’t afford to miss.
But underneath this facade of phony setups and false narratives is the real story. If you take the time to look deeply enough, you’ll see it. You don’t need any fancy indicators or years of experience reading order flow. All you need to do is read each candlestick on the chart like the chapter of a story.
In my haste to spot tradeable setups, I was getting misled by the “public relations” version of the story. I no longer took the time to seek the truth of what was actually going on. The result was falling into trap after trap, with every trade I took immediately going against me and my hard-earned year-to-date profits steadily disappearing before my eyes.
Fortunately, it was an easy issue to fix. I simply had to go back to my old process. As soon as I did, my results immediately began to improve, and I was profitable once more.
What’s the real story?
Now whenever I take a trade, I force myself to narrate the story the price action on each timeframe is telling. This added step in my trading process prompts me to look beyond the surface-level story and prevents me from falling into those traps.
I often mention Trader Dante, as above any other trading mentor, his teachings were what helped get me on the path to consistent profitability. His ability to read the story of price action, and explain that story in clear terms, taught me how to decode the price action for myself. He would always be skeptical of the false advertisements of the market. Instead, he would look at all the clues in the price action and get to the truth of what was really going on.
He would often try to “talk himself out of taking trades”, rather than the more common and less profitable approach of talking yourself into taking trades. This approach helps to highlight all those factors that suggest a trade won’t work, and helps uncover the real story. It’s similar to how a scientist tries to disprove theories, rather than prove them to be correct. This way we avoid all the inherent biases of trying to convince ourselves we’re correct.
Next time you’re perusing through charts looking for trades, really stop to read the story that price is telling. Start a dozen or so bars back, and read each candle like a chapter of a book. What’s happening in each candle? Did the story go as expected or was there a twist of events? Do sellers keep getting squeezed despite plenty of opportunities for a sell-off? Or do buyers keep gaining an advantage no matter how strong the overhead resistance is? And the most important question of all – what’s going to happen in the next chapter?
The market tells the same old stories again and again. As soon as you start paying attention to them, you’ll quickly learn how to predict what will happen in the next chapter.
There is great value in thorough analyses of your journal. Knowing the win rates and expectancies of each setup has allowed me to adopt a safer risk strategy. I now only risk the highest amount of 2% on the setups that I know perform well. For lesser setups, or setups that don’t yet have enough data, I opt for either 1% or 1.5%. Over time this has significantly reduced my drawdown and was the saving grace in my misadventure as a human algorithm.
But you have to be careful that when using this data, you don’t interfere with your process. While systematic traders can make consistent profits using only data, as discretionary traders our process outweighs any data we have, no matter how strong it may appear.
After realizing I’d become nothing more than a human algorithm, it struck me that it wasn’t the first time this had happened. Just a year prior, I’d run into a similar issue. Rather than a setup-spotting algorithm, I was a level-spotting algorithm. Again it occurred after spending weeks collecting and analyzing data, a process that disrupted my usual process. It led me to go level-spotting every time I saw a chart, ignoring the narrative of the price action. Again my results had suffered, but again it was an easy issue to fix once I realised what was happening.
It’s a humbling experience to see firsthand that your edge as a trader can vanish at any time. Not only that, but it can occur as a result of trying to become more profitable. All it takes is something to interfere with your process.
Take the time to look at your current trading process and your latest results. While you’ll hit periods of drawdown that result from random probability or market conditions, I’ve found my tendency to blame bad results on these factors has only held me back. You need to be on constant alert for anything that could be interfering with your edge.
It helps to look back on your last 30-50 trades, to keep track of your performance and see if you’re deviating from your norm. If you are, you need to find out why as soon as possible. Could it be that in your haste to become more profitable you’ve turned into a human algorithm? Are you just going level or setup-spotting while missing the bigger picture? If so, you know what to do.
Now go forth and trade like a human, for the last thing the market needs is another algorithm.