Why retail algo-traders fail?

Trading is a risky and difficult endeavor. Many traders who execute trades manually fail, and I suppose the failure rate for algorithmic traders is also high. Although the failure rate of manual traders and algorithmic traders may be similar, I think the reasons are different.

Here are my top reasons why retail algo-traders fail.

Philosophical Reasons

1. Trading systems that do not match their personality.

   1. Algorithms aren’t just code. They reflect the personality and assumptions of the developer. If you are a laid back person like me, you may rather trend following algorithms. If you are a hyper focused control freak (lol jk) you may need to scalp or do mean reversion trades. You have to match the style to your personality. Because you need a system you are comfortable with, and a system you can consistently trade.

2. Searching for the holy-grail

   1. By searching for the holy-grail they jump from system to system and never go deep on any system. If trend-following didn’t work this week, they jump to mean reversion. This will only result in you losing money and always being out of sync with the market. Many successful algo-traders specialize with a specific style, market and timeframe.

3. Some think it is an opportunity to get rich quick

   1. Some beginners think algo-trading is a method to get rich quick. They create a beautiful backtest and can’t see how this would fail. Only to realize if it was that easy, everyone would be doing it.

Technical reasons

1. Trading low time frame systems.

   1. Algorithmic trading is not high frequency trading. Retail algo-traders seem to be attracted to finding a trading edge in the lower time frames. Lower timeframes being 5 minute candles and below. If you are trading in these timeframes, you are essentially trading market noise. Obviously its possible to do this, but ask yourself, do you have the equipment to compete against high frequency trading firms that have the clear advantage over you? HFT firms spend thousands to be located as close as possible to the exchange they are trading, to cut down on latency. They invest in the best technology, and hire the smartest people from the best universities. Why would you decide to compete against that from your home on your internet connection? Save yourself time and money and pick a better fight.

2. Focusing on metrics that are not indicators of forward success

   1. The Sharpe ratio is not the holy grail. There I said it. Despite what many books, and talking heads may tell you. The highest Sharpe ratio is not the metric you want to optimize for. There is a lot of risk inherent in the Sharpe Ratio, which I will explain in another blog post. Algo-traders essentially build algorithms and try to optimize for metrics that do not predict forward success. Therefore, when they run the algorithm live, performance degrades.

3. They do not understand probabilities and statistics.

   1. If I lose 60% of my trades, but when I win trades, they are 5 times larger than my loses, will I be profitable in the long run? Yes. Unfortunately, many beginner algo-traders do not understand this, and opt for the opposite, negatively skewed systems. These systems inevitable fail in the long run.
      

4. They trade high position sizes.

   1. I did this when I was starting out. I would trade very large positions, putting my account in jeopardy with every trade. If a month of trading can wipe out your account, you need to reduce your position size and get better at risk management.

Emotional reasons

1. Fear of losing

   1. Just because you made an algorithm, doesn’t mean you are ready for the stress of watching it lose your money over and over… and over again. Retail algo-traders, especially begineers, are fearful of losing money. So when their algorithm is in an open equity drawdown, they may intervene and close the trade, thinking that they are helping by reducing loses. But your algorithm may have recovered that open equity drawdown and close the trade with a profit.

2. Fear of being wrong

   1. I get it, you are smart and you want to be right. But being right does not mean you will make money algorithmic trading. The fear of being wrong pushes a lot of algo-traders to optimize systems for a high winrate. Then these high win rate systems, end up being curve fitted systems that are destined to fail in the live market. If you are doing this, just stop.

3. Lack of patience

   1. A person with no patience will favor the quick result, lower time frame trades and day trading. They may also trade scalping systems, or mean reverting systems. For whatever reason, they are either afraid of being in the market long, or they want the quick win and dopamine hit. This leads them to trading systems that are negatively skewed and destined to blow up.

   2. Another aspect of no-patience algo-traders are those that purchase a system, because they do not want to take the time to learn, build and study algorithms for themselves.

How to avoid failure?

Avoiding failure is simple. Do the exact opposite of previous lists.

1. Trade a system that matches your personality.

2. Stop looking for the holy-grail.

3. Understand this is not a get rich quick opportunity.

4. Trade higher time frames (30 minute candles and above.)

5. Focus on metrics that are better predictors of long term performance. (Profit factor, Profit Expectancy, Return/Drawdown ratio, etc.)

6. Improve your probability of profiting from algorithmic trading. This might require some basic education in probability and statistics.

7. Trade smaller position sizes. Risk 0.1-0.25% of your account on each trade.

8. Release your fear of being wrong and losing money. This is an inevitable part of leveling up your trading performance.

9. Remove the stress of trying to get rich overnight. Be patient, read the books, complete the courses, and do it the right way. There are no shortcuts.