What is algo investing and trading?

Algo investing and trading means following a set of written rules, automatically, every time. The rules decide what to buy, when to sell and how much to hold. A computer then follows them exactly, without second thoughts.

That’s all it is. It doesn’t have to be fast, complicated or secretive.

Two parts, two benefits

Rule-based: consistent and objective. A rule is a decision you make once, calmly, instead of again and again under pressure. Because it is written down, you can test it on past data and see how it would have behaved, including its bad years, before you trust it with money. Your decisions stop depending on mood, news or a tip.

Automated: emotion taken out. Most investors know what they should do. The hard part is doing it when markets are falling and every instinct says sell, or rising and every instinct says chase. Automation follows the rules when it is hardest to: in volatile markets, on bad days and on days you are simply busy.

The myths that put people off

“Algo means high-frequency trading.” That is one corner of it. A rule that checks a portfolio once a month and rebalances it is an algo too. Several of the strategies on this site trade a few times a month or a few times a year.

“It has to be complicated.” The opposite is usually true. Simple rules are easier to understand, easier to test honestly and harder to fool yourself with. Many of the ideas we test fit in five lines.

“It’s only for traders.” Many investors already use one without calling it that: a SIP invests a fixed amount automatically every month, with no new decision each time. Rules work just as well for long-term investing: which stocks to hold, when to rebalance, how much to keep in each. For example, our monthly momentum and low volatility studies are simple investing rules applied once a month or once a quarter.

“Algo means guaranteed profits.” No. A bad rule followed perfectly loses money perfectly. Rules remove emotion, not risk. That is why every strategy here is tested the same way, with costs, on data it never saw, and published whether it passed or failed. See how we score strategies.

What rules can’t do

• They can’t remove market risk. A well-tested rule still has losing months and deep drawdowns.
• They can’t make a weak idea strong. Testing tells you which rules are worth following and which aren’t.
• They can’t run themselves without care. Data, brokers and servers fail in practical ways, and an automated system needs checks of its own.

Where to start

• Strategies: real rules, tested honestly, with the results, the risks and our verdict.
• Landmines: the traps that make bad strategies look brilliant in a backtest.
• Launchpad: how to build and run your own algo-bot safely, and lessons from real failures.

Nothing on this site is a recommendation to buy or sell anything. See the disclaimer.