Why Traders Slowly Drift Away From Their Own Strategy

Disclaimer these posts are for educational and/or entertainment purposes only. If you want more warnings and disclaimers check every footer on every page, or click here, it sends you to the same disclaimer if you're too lazy to hit [end] on your keyboard. These posts are not advice, trades you do are of your own, if we hint to try something, it's to test on a demo or at your own risk, we never guarantee profitability.

Let's Catch the Drift

Traders make a strategy; however, the rules are pretty loose. Whatever rules they made barely exist after a while. Why? Because they tend to drift away from them. They get distracted by other things, other rules that exist in other strategies that don’t apply to the one they made and are currently using.

It’s unfortunate. As I stated earlier, humans are very volatile in their actions. They’ll do one thing, change their mind, and do something else. They will say, “Oh yeah, let’s meet and have lunch one day,” and then cancel on you 30 minutes after you already arrived and the reservation is already set, so you end up eating on your own.

Back to the trade. The problem is they make a strategy that looks good in hindsight. For example, a breakout of a recent high with a pullback. It looks good because it showed it moved, but the market doesn’t move like that all the time.

Also, a lot of discretionary traders tend not to backtest or even look back in the past to see if it’s a repeating pattern that they can exploit. For example, that 1-2-3 pattern: price moves up, that’s one. Price retraces, that’s two. Price breaks out of that high, which was one — that’s the breakout entry.

Traders get deluded into thinking it happens more often than it does. Then they start trading it and notice the market gave them a good sampler. It tasted good, but when they got the product, there were some bad packages in it.

Bad Packages, Bad Strats

Packages such as false breakouts, instantaneous reversals, sideways markets with micro 1-2-3 patterns within a 1-2-3 pattern.

They’re already in a trade, but they’re seeing other setups while in the trade, so they start to question.

“Okay, I selected incorrectly. Maybe I should get out and then enter this one that’s coming in.”

Then they reverse, and then they lose.

It’s a downward spiral. The rules don’t exist anymore.

“I said they existed. I put them there. It didn’t mean I was going to follow them.”

Such an unfortunate trap traders face these days. This is why systematic trading is a better option, because once you compile, you can’t really break the rules as much as you want to. The strategy is armed against you breaking it.

You would have to disarm it, open the code editor, and delete some lines or edit some parameters to do trades that weren’t meant to be done.

At that point, if you’re really going to break the code, what is the point of even auto trading? You might as well go back to the discretionary path.

I know I’m rambling and I’m all over the place, but this is a trader. I’m a human and just laying it out there.

Feelings Filter

Another point I notice is traders filtering signals based on the last trade result and how it felt. The emotions tied to discretion break down most strategies.

The trader either skips the next setup or modifies the next setup or trade.

For example, it’s supposed to run a fixed stop with a risk of one and a reward of three. They see, “Well, it doesn’t go as far, so let me drop the reward to two.”

What they don’t notice is that by reducing the reward, the transaction costs weigh more on the entire trade, win or lose. Because it’s something you pay in the end.

Churn and Burn

When you lose, it gets taken. When you win, it gets taken.

So when you profit less, it doesn’t mean the broker and CME Group take less. They take a fixed amount. If you trade more, they love that, don’t they?

They like it when you churn because that’s exchange fees, NFA, clearing, commissions. It doesn’t matter if the trader won or lost. What matters is the exchange and the broker got paid.

As a trader, one must think: do I want to operate in this way and bleed out on transactions because I can’t even follow the rules of the strategy I made?

Many Forms of Backtesting

In terms of backtesting, there are many ways to do it. There’s Python. There’s NinjaTrader. I use NinjaTrader, there's CQG, TT, TradeStation, there’s Excel if you know how to use that. Or there’s a manual way of doing it.

For example, if you did have a risk of one and a target of three, that risk-to-reward ratio requires at least a 25% win rate to break even and cover transaction costs. So let’s put it at a 27% win rate.

With that said, maybe you manually backtest it. Manually visually backtest the strategy for a month. It’s going to take some time, but you’re a discretionary trader, right? You like to do things manually.

So I think it’s good to visually manually backtest, because you saw what happened in the past. Then you’re going to really test yourself.

If you would have taken that setup when you saw it — it’s already passed. You can’t do anything about it, but you’re going to sit and think. Then you’re going to be distracted by other movements that happened in the past.

You’re going to challenge yourself.

“Oh, there’s a spike the signal didn’t give. How could I capture that move?”

Discretionary trading is tough, folks.

The manual backtest — I remember doing this. I did this on weekends. I downloaded the data for the time period I wanted to test.

For example, since equity index futures are usually three or four months of life, I would download that. Then I would look at the data, the charting, apply the setup, mark it on the chart.

You should be able to use drawing tools, right?

Then calculate the P&L. Some drawing tools will calculate the P&L based on how you take it and what the end result would be.

You could log this in Excel or something like that.

It’s tedious, but that is the game for trading manually.

Don’t get me wrong. There are backtests where you can provide your setup and the backtesting engine will run it for you.

You can do something like this on NinjaTrader with the Strategy Builder.

But let me make it quick for you.

Millions of Lies in a Backtest

In NinjaTrader 8 you can game the backtester. So you would sabotage your backtest in the end by lying to yourself, thinking the backtest is legit and the strategy actually makes $1 million in a year.

TLDR — how do you do that?

If you make something in Strategy Builder, set it to fill on limit orders on touch and have a target that’s one point on the E-mini S&P.

You will have a very smooth upward curve, and you’re going to see that you made $1 million in three months.

You know how I know?

You’re not there yet, but I was 18 years ago.

I found out that flaw in the backtest engine back when it was NinjaTrader 7 (the testing issue still persists today). And guess what? I ran it in live markets.

You can only imagine the results of that. It was nothing close to what the backtest showed. I got margin called, end of that project.

Gravitational Drift

Lastly, it always gravitates back to the emotional aspect that causes the drift.

The trader was either not confident with the strategy in the first place, hence they started tweaking it, so it doesn’t look like the original product anymore.

In systematic trading and algorithmic trading, we do things like this, but it’s a procedural process. We add upgrades, patches, and improvements that improve the strategy.

We don’t throw things out there that seem like, “Oh, let’s try this and see if it sticks.”

“Let’s put a scalping parameter in a trend-following strategy. Let’s see if it will scalp around the trend.”

Goodness.

Methodology to Succeed

You have to be more methodical if you’re developing a strategy. If you set rules, you should follow them.

Because if you don’t, just delete the rules. You just have a strategy by name — it’s a banner you put up when you’re trading.

End of discussion.

I know that may have seemed like a long-winded rant, but I really sat and thought about what I had to say.

It may come off blunt to some people, but I’d rather be honest with you than lie to you with magical words.

You’re going to end up saving a lot of time in trial and error.

Because after reading what I said, you’re probably going to notice these pitfalls in the future.

You’re going to be like:

“Oh… he said that.”

Then you’re going to stop and watch what happens.

You’ll see the disastrous result unfold in front of your eyes — but you didn’t take the bad trade.

See?

Trade well everyone, until next time.

In closing

If you find these journal entries helpful, please consider subscribing here.

~Asymmetric_Vol