#1 Not learning to code
This one is the most important, before starting anything you should learn about programming. Coding will make you assimilate a certain logic that’s close to mathematical formulas and can help you formalize your trading process. It’s essential to be able to understand everything that’s “under the hood”, what if you strategy starts to slow down after a few months and you’re not able to improve it yourself.
You won’t learn programming in a day, you should take your time to learn and understand the process. Fortunately, there are multiple free methods you can use to learn about Python. You can use websites like EDX, Coursera, and Udacity.
#2 Backtesting and training on the same period
Let’s say you found the perfect strategy that makes +300% in the 2014 period, you may want to backtest it on a different period, the strategy may work in that specific time but it could make you lose a lot on another period. This beginner mistake has a name: overfitting. Ideally you want to split your data set into at least 2 parts: train and test. But if you want to have a rock-solid performance, you can try K-Fold cross validation, it’ll split your data set into K parts, train 1 part and test it on the other ones, and so on.
#3 Not backtesting enough
Backtest, backtest and backtest. Use different time periods, adjust the trading size, the strategy could work by buying 100$ worth of stocks at a time but what if you want to scale it ? You could introduce slippage and of course broker fees.
Backtesting is good but paper trading is better, you should run the strategy in real-time but without any broker connection, this way you can simulate how it’s going to behave with current market situation.
#4 Not having a risk management strategy
Risk management is going to make a difference during bear markets or high-volatility periods. You can limit the maximum exposure and ignore any buying signal if you hit the limit, or automatically close any position older than a few days. These are suggestions, it’s important to make sure you won’t get stuck with a growing loss over time.
#5 Having unreliable data
Your strategy will be based on financial data, either real-time, minute or daily data, a single data point can destroy your profits. You need to make sure it’s coming from a reliable source and not some random websites, a good source is Quandl, some of their datasets are free.