What Is Algo Trading? A Beginner's Guide for Indian Traders
Algo trading (algorithmic trading) means using a computer program to place buy and sell orders automatically, based on rules you set in advance: when to enter, when to exit, how much to trade and where to stop a loss. In India it is legal for retail traders through a SEBI-registered broker, and no-code platforms let you build strategies without programming.
What is algo trading, in simple words?
Think of it as writing your trading plan in a form a computer can follow. Instead of watching charts and clicking, you write the logic once:
- When to enter: "at 9:20", "when the price breaks yesterday's high", "when the 20-period average crosses above the 50".
- When to exit: a profit target, a stop-loss, a time, or an opposite signal.
- How much: a fixed number of lots, or a size based on the amount you're willing to lose.
The program checks those conditions on every price update and sends orders to your broker the moment they are met, whether you are watching or not.
A real-life example
Say Rahul trades NIFTY futures with a simple plan: buy one lot (65 units) if NIFTY crosses 25,000, take profit at 25,300, and cut the loss at 24,900.
- ₹6,500 the most he can lose (100 points × 65)
- ₹19,500 the target profit (300 points × 65)
- 1 : 3 risk to reward, decided before the trade
By hand, Rahul has to watch for the level, place the order, set the stop and watch the exit, every day. One meeting and he misses the entry. One hopeful moment and he moves the stop. As an algo, the plan runs the same way every time. These numbers are an example of a plan, not a prediction.
How algo trading works
The loop an algo strategy runs all day
Every algo trading setup has the same parts:
- Market data: live prices, indicators, option data, time.
- Rules: your entry, exit, stop-loss and sizing logic.
- An execution link: your broker's API, which carries orders to the exchange.
- Risk controls: per-trade stops, daily loss limits, automatic exits at a set time.
- Monitoring: logs, alerts and a dashboard so you can see what the strategy did and why.
On a cloud platform, all of this runs on servers, so your strategy keeps working when your laptop is closed.
Manual trading vs algo trading
| Manual trading | Algo trading | |
|---|---|---|
| Decisions | Judgement in the moment | Rules written in advance |
| Execution | Seconds to minutes, when you're watching | As soon as conditions are met |
| Emotion | Fear, greed and hesitation affect trades | Rules are followed as written |
| Consistency | Varies with mood and focus | The same every time |
| Instruments at once | A few | Many |
| Testing before risking money | Hard | Backtests and Live Offline runs |
| Main weakness | Discipline | A bad rule is repeated faithfully |
Manual trading still has a place: discretionary traders reading news or unusual situations may not want rules. For a fuller comparison, see algo trading vs manual trading.
Why traders use algo trading
- Discipline. Stops are never moved, entries never skipped, and you don't revenge-trade after a loss.
- Time. The strategy runs during market hours while you work.
- Scale. One person can run several strategies across indices, stocks and options.
- Testing. You can see how a rule would have behaved over years of data before risking money.
- Speed where it matters. Exits fire immediately when a stop is hit, even in a fast market.
Common types of algo trading strategies
| Strategy type | The idea | Example |
|---|---|---|
| Trend following | Prices that are moving tend to keep moving for a while | Buy when a 20-period average crosses above a 50-period average |
| Breakout | Big moves often start when price leaves a range | Buy above the first 15 minutes' high, stop below its low |
| Mean reversion | Stretched prices tend to return to their average | Buy below the lower Bollinger Band, exit at the middle band |
| Option selling | Option premium loses time value as expiry nears | A hedged NIFTY short strangle with stop-losses on each leg |
| Option buying | Profit from large, fast moves | Buy calls or puts on a breakout, with a tight stop |
| Rebalancing | Keep a portfolio at target weights | Rebalance 10 stocks to equal weight each month |
| Arbitrage | Capture small price gaps between related instruments | Future vs underlying; usually too thin for retail after costs |
Options strategies suit automation particularly well because they often have several legs, each with its own stop and adjustments, which are hard to manage by hand.
A short history of algo trading in India
Computerised trading started in US markets in the 1970s and spread to large institutions worldwide. In India, SEBI allowed direct market access for institutions in 2008, which opened the door to institutional algo trading, and exchange colocation followed. Retail traders came much later, through broker APIs and no-code platforms. In February 2025 SEBI issued a dedicated framework for retail algo trading, which now applies in full.
Is algo trading legal in India?
Yes. Retail traders can automate through their broker's API under SEBI's framework:
- Every algo order carries an exchange-issued Algo-ID, so it can be traced.
- API access uses keys issued per client from a static IP your broker has whitelisted.
- If you build your own algo and stay below the exchange's orders-per-second threshold (10 per second on NSE), you don't register it separately. Above that, it must be registered through your broker.
- Brokers are responsible for algos on their platforms; third-party algo providers must be empanelled with the exchange.
Details: SEBI's algo trading rules, the Algo-ID mandate and is algo trading legal in India?
Test before you risk money: backtest, then Live Offline
There are two ways to check a strategy before real money, and they answer different questions.
A backtest runs your rules on historical data. It shows how the strategy would have behaved: profit and loss, drawdowns, losing streaks. It is the fastest way to throw out bad ideas. Its weakness is over-fitting: tune a strategy too closely to the past and it fails live. Include charges and slippage, and test on several years.
Live Offline runs your strategy on live prices: the strategy runs and records every trade, but nothing is sent to the exchange. It shows whether the rules fire as you intended in today's market, with today's liquidity.
- Write the rules down completely, including stop-loss, position size and what happens near expiry.
- Backtest on several years, with costs, and read the drawdown as carefully as the return.
- Run it Live Offline for a few weeks and compare its trades with the backtest.
- Go live at the smallest size your broker allows.
- Review weekly and change rules only for reasons you can explain.
How much money do you need to start?
There is no fixed minimum; it depends on what you trade. Live Offline needs no capital at all. For live trading, buying one NIFTY option at a premium of ₹100 costs ₹6,500 (100 × 65), while selling options or trading futures needs much larger margins. Start with an amount you can afford to lose while you learn, and size each trade so a single loss is a small fraction of it.
Myths about algo trading
- "Algo trading guarantees profit." Automation executes a strategy; it doesn't make it good. A losing strategy loses faster when automated. SEBI's own study found 91% of individual F&O traders lost money in FY25.
- "Only institutions can do it." Retail algo trading is formally allowed and widely used through broker APIs.
- "You need to code." No-code builders let you assemble rules from conditions.
- "Set it and forget it." Broker sessions expire, margins change, orders get rejected. Strategies need watching.
- "Faster is always better." Retail strategies don't win on speed; they win, if at all, on logic and discipline.
For the profitability question in depth, read is algo trading profitable in India?
Is algo trading right for you?
Pros
- You have, or can write, clear rule-based ideas
- You struggle to follow your own stops and plans
- You can't watch markets all day
- You're willing to test before trading real money
Cons
- You don't yet have a defined approach
- You expect automation itself to make money
- You won't monitor a live strategy
- You want to skip testing and go straight to live
If most of the first list fits you, algo trading is worth learning. If most of the second does, start by defining and testing a simple plan by hand first.
Risks to manage
- Strategy risk: the rules have no real edge, or are over-fitted to the past.
- Execution risk: slippage, rejected orders, thin liquidity in far strikes.
- Technology risk: expired broker sessions, connectivity, wrong settings.
- Leverage risk: positions too large for the account.
- Regime change: the market stops behaving like your test period.
Every one of these is managed the same way: test honestly, size small, use stops and daily limits, and monitor. See reducing drawdown and how to build a bullet-proof strategy.
Glossary
| Term | Meaning |
|---|---|
| Backtest | Running a strategy on historical data to see how it would have behaved |
| Live Offline | Tradetron's test mode: the strategy runs on live prices and records every trade, but nothing is sent to the exchange |
| Live Auto | Orders go to your broker and the exchange |
| Drawdown | The fall from a peak in account value to the following low |
| Slippage | The gap between the price you expected and the price you got |
| Square-off | Closing an open position |
| Algo-ID | The exchange-issued tag every algo order carries |
| OPS | Orders per second; above the exchange's threshold a self-built algo must be registered |
| Static IP | A fixed internet address your broker whitelists for API access |
Frequently asked questions
What is algo trading in simple words?
Using a computer program to buy and sell automatically according to rules you set in advance, such as when to enter, when to exit and how much to trade.
Is algo trading legal in India?
Yes. Retail traders can automate through a SEBI-registered broker's API. Algo orders carry an exchange Algo-ID, and self-built algos above the exchange's orders-per-second threshold must be registered through the broker.
Do I need coding knowledge for algo trading?
No. No-code platforms such as Tradetron let you build strategies from conditions and keywords. Coding helps for unusual logic but isn't needed to start.
What is the difference between a backtest and Live Offline?
A backtest runs your rules on past data to estimate how they would have performed. Live Offline runs them on live prices and records every trade without sending orders, to check they behave as intended today.
How much money do I need to start algo trading?
None to test: backtests and Live Offline need no capital. For live trading it depends on the instrument; start with an amount you can afford to lose and size trades so one loss is small.
Is algo trading profitable?
Only if the strategy has an edge after costs and is managed sensibly. Automation improves execution, not the strategy. Most retail F&O traders lose money, with or without algorithms.
Can beginners learn algo trading?
Yes. Start with one simple rule-based strategy, backtest it, run it Live Offline, and go live small. Learn from each step before adding complexity.
What happens to my strategy if my internet goes down?
On a cloud platform like Tradetron, strategies run on the platform's servers, so your own internet connection doesn't stop them. Your broker connection must stay active, so check that your broker session is logged in each day.
