TradingBot series — 5 reasons to build your own trading bot
In this post, we introduce where a trading bot should be better than a human and why you might want to build one yourself
Reason 1 — A systematic strategy
Good human traders can be very skilled in identifying trade patterns in technical analysis charts. Humans are much better pattern recognisers than algorithms today. Bots cannot hope to compete on that basis today.
However, the systematic approach needed to develop trading bots is very helpful in making consistent profits:
- Identify and code up indicators and promising algorithms
- Backtest the strategies scientifically across multiple markets, time periods and market cycles
- Set the bot paper trading and then move to live trading once it has proven itself and the bugs have been ironed out
- Tweak, make improvements and repeat
This gives bots a great advantage —imposing a systematic cycle and scientific approach.
Reason 2 — Scan more assets
Whilst a good human trader may be able to analyse 10–30 assets a day, perhaps more with some technological support, a bot can analyse thousands of assets many times a day across many different candle time periods.
This is a skill that bots need to take advantage of in order to create consistent profits. It allows the bot trader to be more discerning in which trades you take. The bot allows you to look across a much wider set of assets to find the most attractive needles in this much larger haystack of assets.
Reason 3— Unemotional
I am a lousy human trader. I don’t stick to the system. I get bored easily. I often enter trades too late and stay in positions too long. I am prone to getting carried away with the emotions of winning and losing.
On the other hand a bot is completely unemotional and systematic. The human work is the interesting bit of identifying profitable algorithms, backtesting them, paper testing them and then letting the bot run. Unemotional traders are usually better traders.
Reason 4 — Run 24/7
As humans, we need eat, sleep, and also have a life. When not watching the active market we can miss good trade entries and exits. This leads to regret and introduces more emotion into trading.
In contrast, the bot can sit in the cloud and grind away day and night, collecting the market data, setting orders and managing positions.
Reason 5 — a full workout as a coding project
From a developer’s perspective, building a trading bot requires afull spread of technical skills. It is a great way to build out skills, including:
- Interacting with exchange APIs and Telegram APIs
- Front-end development for your bot’s web app
- Database design and structuring
- Code refactoring — vital to avoid the bot codebase becoming spaghetti
- Testing
- Security — you really need to secure your API keys
- Deployment in the cloud
The Trading Bot Series covers our project, building a complete trading bot using Python, exchange APIs, Backtrader and Telegram
